U.S. Army Medical Department, Office of Medical History
Skip Navigation, go to content







AMEDD MEDAL OF HONOR RECIPIENTS External Link, Opens in New Window






Chapter XIV







At the time the United States declared war on Germany, effective methods for treating various types of wounds, with some few exceptions, had been established by surgeons of the Allied armies. Among these exceptions were wounds of the chest. There were wide and some irreconcilable differences of opinion as to the care of intrathoracic injuries. These differences were caused by failure to understand the interdependence of the functions of the circulatory and respiratory apparatuses and to appreciate the contribution of their functions to the powers of resistance, defense, and repair, powers which must be conserved and developed in order to facilitate immediate recovery and to minimize the extent and duration of subsequent disability.

The possibilities for rendering service to the wounded in this direction, as well as in others, through investigations of inherent problems were recognized early in the history of the American Expeditionary Forces by the then chief of the research division of the American Red Cross in France. He secured early opportunities to begin these studies, and, in so far as he was permitted, continued to give assistance; special apparatus was provided for experiments; also supplies of oxygen and nitrous oxide gases, and transportation, which were essential and otherwise unobtainable, were provided.1

Work was begun by the writer in 1917 at the Ambulance de l'Ocean, at La Panne, Belgium, under Col. A. Depage of the Belgian Medical Department, and included clinical, post-mortem, and experimental observations. This work sufficed to disclose the more significant problems and to indicate means for their solution.

 Late in 1917, the chief consultant, surgical services, of the American Expeditionary Forces, recommended continuation of the work, and in 1918, the chief surgeon, A. E. F., detailed for this purpose medical officers, nurses and enlisted men for duty at the Central Medical Department Laboratory at Dijon.2

Experimental studies, carried on in the section provided for surgical research, with the help and advice of the director and of the chief of the division of surgical research, were continued until the Chateau-Thierry operation began.

After this time, and until the beginning of the armistice, a unit a composed of medical officers, nurses and enlisted men appointed by the chief surgeon, .A. E. F., was detailed to sundry field hospitals for the care of nontransportable wounded and to mobile hospitals in zones where there was active fighting.

The directions of the chief consultant of the surgical services were (1) to discover the physiologic interrelationships between the circulatory and respiratory mechanisms in order to determine the functions which need the protection

a The members of the unit detailed to the study and treatment of thoracic injuries were: Col. J. L. Yates. M. C.. in charge; Maj. W. F. Verdi, M. C., surgeon; Capt. W. S. Middleton, M. C., and Capt. M. A. Blankenhorn. M. C., physicians; Capt. Robert Drane, M. C., and Capt. J. M. Steiner, M. C., radiologists; Capt. J. T. Gwathmey. M. C., and Anna Fitzgerald, A. N. C., anesthetists; Anne Bernard, A. N. C., surgical nurse.


to assure the largest opportunities for immediate and remote recoveries; (2) to develop the simplest effective surgical methods compatible with physiologic requirements; (3) to apply these methods to the wounded whenever there was any possibility of saving life and without regard to the high mortality rate that would inevitably accompany acceptance of the gravest risks; (4) to follow each fatality with necropsy to determine what should not be done, to trace the results in those who recovered and thereby discover the dependence of degrees of functional rehabilitation upon methods employed in order to get better methods; and (5) to make eventually a report indicating how soldiers suffering from intrathoracic injuries could be the more certainly protected against death and disability.

What follows in this chapter is in consequence of compliance with these directions. No attempt has been made to preserve chronologic sequence in observations; facts as they now appear important are presented. Statistics are avoided as far as possible since, because of the many inevitable sources of error, they can not fail to be even more than usually unreliable.


Two factors, peculiar to thoracic injuries, are of sufficient importance to be kept constantly in mind: First, mere existence, as well as additional activities, including the capability of withstanding the extraordinary stresses imposed by wounds and by surgical treatment, depends fundamentally upon ability to provide oxygen for tissue metabolism. Second, chest injuries impose not only the burdens incidental to other tissue insults but also definite restrictions to the supply and delivery of oxygen.

Suitable methods of treatment will protect the means whereby oxygen is supplied to the body and delivered throughout the body and will be determined by knowledge of the activities upon which the functions of supply and delivery of oxygen depend. A summation of all relevant observations made by the writer during and since the World War was presented before The American Association for Thoracic Surgery in 1924.3 It is given here in full, with a few minor changes, as it indicates the essential activities and functions, and how they may be conserved and rehabilitated.


A biologic aphorism, no life without breathing, indicates in a general way the importance of respiration. Activities sufficient merely to support life or to realize the utmost physical and mental powers, including defense and repair, are produced by metabolic processes which are dependent primarily on oxidation. This explains why man, although he may survive for weeks without food and for days without water, can exist for only a few minutes deprived of air. It also explains why any reduction in supplies of oxygen to the body, in deliveries of oxygen throughout the body and in utilization of oxygen by the body, imposes a corresponding degree of disability.

The many diseases affecting the thorax and its contents, the enormous totals of transient and permanent disabilities and the large number of deaths


they cause constitute a serious problem. More effective therapy is needed to provide greater limitations of disability and to reduce mortality.

There is a direct road to this accomplishment. Reduction in vital capacity is so constant a result of thoracic diseases that the extent of this reduction is an accurate measure of the disabilities attributable to them.b In order to obtain better methods of treatment, it is essential to know the structures and the functions of the structures that determine vital capacity, how vital capacity is affected by various lesions and consequent malfunctions of these structures. and how therapeutic methods can protect vital capacity against reduction during aggressive phases of diseases and facilitate its rehabilitation during regressive phases.

Four distinct phases are notable in completed respiration: (1) Ventilation of intrapulmonary air effected by breathing; (2) external respiration, the interchange of gases between the intrapulmonary air and intravascular blood through alveolar and capillary walls; (3) transportation of aerated blood to somatic cells and of blood needing aeration to lungs by a coordination of the pulmonary and systemic circulations, and (4) interchange of gases between blood and tissues through capillary walls and cell surfaces, or internal respiration. Although most essential, internal respiration is germane to this subject only as it is influenced by the other steps in complete respiration.

Clinical interest centers in the finer workings of the mechanisms of breathing, of circulation, and particularly of external respiration, because of its direct relationships to vital capacity and to life itself. External respiration not only accomplishes aeration of blood throughout life, no matter bow great or how little are the demands for oxygen, but this accomplishment has to be relatively complete at any or all intervals. In order to simplify the immediate presentation, let it be supposed that the walls of all of the alveoli are fused into one sheet of epithelium against which is applied a similar sheet composed of the pulmonary capillary endothelium. Suppose further that the intrapulmonary air is in an even layer on the epithelial surface and the intrapulmonary blood is flowing in another even layer on the endothelial surface.

Conditions being normal, the oxygen in the layer of air suffices to aerate homogeneously and completely the layer of blood as it passes over the endothelial surface. Should the amount of blood be increased, either the area of the sheet of endothelium. must be enlarged or the layer of blood must become thicker. If the layer of blood becomes too thick, it will be aerated neither homogeneously nor completely. On the other side, if the intrapulmonary air be in too thick a layer, because the area of epithelium is too limited, aeration is likewise defective. The states of defective aeration, anoxemia and cyanosis, are abnormal. Therefore, it is presumable that there is a natural control which correlates the volume of ventilated air and the area of alveolar epithelium

b Estimations of vital capacity are not always the only or even the most dependable sources of information. Frequently, it is impracticable to make them. Occasionally, they can lead to erroneous interpretations. As will be made clear later, vital capacity is influenced by the circulation. An individual may have an apparently competent circulation while at rest and show at such times an approximately normal vital capacity; yet that same individual, if the reserve cardiac energy were limited, would show a material reduction in vital capacity after exercise which would be less of a tax on the myocardium than many types of diseases or operations. Other examples need not be cited. This suffices to emphasize the fact that vital capacity readings, like other single sources of clinical information, are reliable when intelligently interpreted in conjunction with all other evidence.


(degree of inflation) with the expanse of endothelium (cross section and length of capillaries) and the amount of blood (unit volume). Moreover, since the demand for oxygen comes from the tissues and is manifested by increments of carbon dioxide in the blood, it is likely that the unit volume of blood in the pulmonary capillaries exerts a telling influence on the three other variables.

A further step in presenting the problem is a supposition that the sheet of alveolar epithelium has been turned into a single sac communicating with a bronchus, and the layer of blood is confined in loops of contiguous capillaries connected with the pulmonary artery and vein. Conditions affecting each of the many alveoli which constitute the portion of the breathing unit directly concerned in external respiration would then be illustrated by a single large alveolus.

Obviously, the first and great commandment is to discover the means whereby the amount of blood requiring aeration establishes conditions suited to this accomplishment.

Knowledge of the structures constituting the apparatuses of breathing and of circulation can alone determine therapeutic priciples. It is necessary to know how they operate during periods of rest; what adjustments occur when activities are gradually or abruptly increased and diminished; what adaptations are employed to meet handicaps imposed by disease; what are nature's methods to increase resistance and to hasten repair; what forms of treatment cooperating with natural methods of adjustment, compensation, and adaptation will augment defense most effectively and facilitate functional recovery.

Breathing, external respiration, and the deliveries of blood to and from the lungs are effected by extrathoracic and intrathoracic structures which are so intimately associated physically or physiologically, or both, that they must be considered together.

The extrathoracic portions need not be described in detail. It is only necessary to recognize that the chief control of the distribution of blood in the systemic circulation and the rate of flow in both systemic and pulmonary circuits, as well as the rate and depth of breathing efforts, is in the central nervous system.

The intrathoracic portions not only provide external respiration but they also are the structures affected by the diseases here considered, and give diagnostic, prognostic, and therapeutic indications.

An accurate measure of the efficacy of external respiration is vital capacity, which is usually defined as the amount of air expelled by the most complete expiration after fullest inspiration. This means that when the air cells are (distended to physiologic limits, inflation, and therefore the total area of alveolar walls, are greatest for normal conditions and are thus disposed to aerate equivalent amounts of blood.

Aeration is an exchange of oxygen and carbon dioxide through alveolar and capillary walls. Natural economy has apparently established a means of control whereby the total expanse of alveolar wall as determined by the extent of pulmonary inflation compels corresponding variations in the total expanse of capillary wall and the amounts of intracapillary blood. Grades of pulmonary


inflation are actively determined by variations in rate, depth, and force of breathing movements, and, as will appear later, are influenced passively by the unit volumes of blood in the pulmonary capillaries. Length, diameter, and, therefore, the contents of pulmonary capillaries can be attributed to three forces, since there is no effective vasomotor control of the pulmonary circulation. These forces are the blood pressures in the pulmonary circulation, fluctuations in intrapleural negative pressures, and alternating intravascular aspiration and expression caused by elongation and shortening of blood vessels with the inspiration and expiration of each breathing cycle.

In addition to central control by the nervous system, there is a peripheral intrathoracic governor which so coordinates the activities of the breathing and circulatory units as to keep volumes of ventilated air present in the alveoli which will aerate at any and all times those amounts of blood being driven through the capillaries. The governor simultaneously correlates the other two factors, namely, the total area of alveolar epithelium and the entire expanse of capillary endothelium.

Knowledge of the structure and function of the governor or gear which coordinates the two sets of variables, volumes of ventilated air and area of alveolar epithelium, with expanse of capillary endothelium and amounts of blood to be aerated, is of basic importance, since it determines how external respiration is affected by disease and can be benefited by treatment.

Moreover, such knowledge answers two other significant questions: Why are reductions in vital capacity so accurate a measure of disabilities? What are the structures and what are their activities that influence vital capacity?

Investigation of the various actions and reactions exhibited by the units concerned in respiration during health and the changes imposed by diseases should answer the question. It is necessary to observe the changes in those units under various conditions.

Normal respiration is possible so long as movements of thoracic parietes, particularly of the diaphragm, are unrestricted; intrapleural negative pressures are undisturbed, and pulmonary elasticity is not impaired, provided the right heart delivers adequate amounts of good blood in the absence of obstruction due to incompetence of the left heart.


During periods of rest, expenditures of energy and hence demands for oxygen are least. In consequence, the volume of tidal air, that exchanged with each complete breathing cycle, is lowest, as is the tidal blood, the amount driven through the pulmonary capillaries during the same breathing cycle. Aeration of tidal blood of rest is accomplished by the tidal air of rest which produces the lowest level of pulmonary inflation with minimal expenditures of energy. This is the period of greatest physiologic economy in which the storage of energies exceeds expenditures by the largest margin.

The reserve supply of air stored in lungs during periods of rest is five or six times larger than the tidal air. Approximately three-fifths of the reserve air is available for sudden physiologic requirements, leaving two-fifths to meet


urgent pathologic demands. A comparable amount of reserve blood available for similar emergencies is stored, less actively circulating, in larger pulmonary vessels.

Tidal air, approximately three-eightteenths of all the air concerned in breathing, serves to ventilate the intrapulmonary air more than enough to keep its oxygen content effectively high. Although one-third of the tidal air is required to fill the trachea and bronchi, the remaining two-thirds causes a change estimated at three-twentieths in the sizes of the air cells.

Thus there is an established relationship between the volume of tidal air (three-eighteenths) and variation in sizes of air cells (three-twentieths). Presumably the same relationship obtains between the amount of tidal blood and the length and cross section of capillaries. All combine to afford the means to aeration of the blood circulating through the pulmonary capillaries.


As activities gradually arc increased after a period of rest, the slowly progressing demands for oxygen are met by proportionately higher levels of inflation produced by larger volumes of tidal air and by increased amounts of tidal blood driven by higher pressures. Relationships between tidal air and alveolar size and between tidal blood and capillary dimensions occurring during rest are but little distorted though produced less and less economically as activities increase.

A level is reached which may be termed optimum, whereat the tidal air equals the vital capacity. This is the upper limit of economic expenditures of energy at which extraordinary activities can be prolonged without causing early exhaustion. When this level is unusually high, it gives the remarkable endurance exhibited by exceptional athletes, by superior race horses, and by certain individuals in meeting the stresses of disease.

Activities raised above the optimum level initiate compensatory responses and mark uneconomic expenditures of energies which assure exhaustion. In order to supply the oxygen demanded for internal respiration, breathing is more rapid and the heart beats faster to develop higher pressures. Finally a still higher level is reached which may be called maximum, because at this level the utmost powers are realized for the brief period before exhaustion forces reduction in activity.

When the sequence is reversed, the processes are orderly reductions in breathing and cardiac rates and diminutions in volume of tidal air, pulmonary inflation, the amounts of tidal blood and capillary expanse. The relationships between the total areas of alveolar walls and capillary walls are adapted to the blood that requires aeration and the intrapulmonary air, properly ventilated, which supplies the oxygen.

Gradually increased activities provide opportunities for orderly adjustments in the actions of respiratory units and develop the largest total power of which an individual may be possessed. Hence the desirability of a suit- able warming-up process before severe contests.



If a resting individual should suddenly engage in a critical physical contest, the demands for oxygen would jump in a few seconds from lowest to highest. limits. The reserves of air and of blood stored in the lungs would be utilized until compensation could be accomplished. It has been estimated that the amount of blood delivered by the pulmonary circulation can be increased abruptly more than tenfold. Or should the individual while driving at the maximum level of action suddenly change to complete rest, the rate of external respiration measured by deliveries of aerated blood would decrease with almost the same rapidity.

The relationships between volume of air and total area of air-cell wall and amount of blood and expanse of capillary wall must be preserved with a fair degree of accuracy not merely to effect proper aeration of blood but to continue life. A very few seconds suffice to cause dilatation of the right heart or to produce edema of the lungs. These complications are seen infrequently because of a peripheral means of control that maintains the air-cell-capillary balance.

Abrupt changes are not alone less economic than are gradual but also are more dangerous, as they permit less opportunity for orderly adjustments and are operated almost entirely by expenditures of reserve energies or margins of safety.

This also applies to the handicaps imposed by disease. The more gradually induced are the less immediately dangerous, and this principle should be recognized in methods of treatment. Changes affecting the relationships involved in aeration of blood should be minimized in extent and in the rapidity with which they are produced.


Vital capacity is a suitable term. It does not occur post mortem. When it is less than the tidal air of rest, life can not continue. If it is equal to the tidal air of rest, existence is possible until the relationships effecting external respiration are disturbed by increased demands for oxygen. The excess of vital capacity over the tidal air of rest measures ability to work and the margin of safety.

Observations of natural adaptations to the handicaps of diseased states are as valuable as they disclose the compensatory changes that assure the nearest approach to normal external respiration of which the measure is vital capacity.

Normally, a volume of ventilated air on one side of a sheet of alveolar epithelium of exactly proportionate dimensions is separated by the interposition of another wall of vascular endothelium just as exactly proportionate in area to an equivalent amount of blood needing aeration. Normally, these variables fluctuate synchronously and almost equally. The upper limits are set by the total areas of epithelium and endothelium which can be presented, since the volume of air and blood can be further increased by raising the respiratory and cardiac rates for brief periods.

Limitations in any of the four factors-volume of ventilated air, total expanse of alveolar epithelium, total area of vascular endothelium or unit


volume of good blood-reduce the essential function, external respiration, which is manifested by corresponding reductions in vital capacity.

Adaptations strive to maintain the broadest margin of safety or the largest vital capacity. When a part of the alveolar epithelium is incapacitated, the remaining portions are expanded so that the total area available for external respiration may remain as nearly normal as possible. This is compensatory or physiologic emphysema. Likewise, the unit volume of blood is correspondingly increased in the capillaries in contact with hyperfunctioning alveoli. This is compensatory or physiologic hyperemia. Moreover, the unit volume of blood is reduced in the capillaries adjacent to the incapacitated alveoli. Similarly, increase or decrease in the unit volume of blood delivered through capillaries is associated with equivalent inflation or deflation of corresponding alveoli. Passive congestion is quite different. The capillaries are engorged, but the unit volume of blood passing through them is reduced. Pulmonary elasticity is restricted, and the total area of alveolar epithelial surface is diminished. Reduced vital capacity here also indicates the limitations in external respiration.

The important connection in maintaining the relationships which determine external respiration lies between the air cells and the capillaries. The interdependence of alveolar inflation and volume of air on one side and of capillary size and amount of blood on the other side is obvious.

Air cells and capillaries are not only intimately associated functionally but closely related physically. Capillary loops, arranged in a meshwork, surround, separate, and yet connect alveoli. When inflation increases, capillaries are straightened and elongated; when it decreases, they are shorter and more tortuous; hence, the aspiration and expulsion of blood with inspiration and expiration. Contrariwise, if tortuous capillaries are straightened and elongated by increased unit volumes of blood, the walls of corresponding air cells are carried with them and inflation results. Also, if this action, which corresponds to a positive phase in erectile tissue, is reversed, capillary walls are less tense, vessels shorten, and their lumens are decreased, air cells contract, and some deflation occurs.

This is the peripheral control, the intrapulmonary governor, and may be called the air cell-capillary gear.

The existence and functions of an air cell-capillary gear are not generally recognized and are disputed by clinicians as well as by physiologists.c The

c Haldane remarked the need of a governor: "We have no guaranty that even during quite normal breathing the distribution of air in the individual lung alveoli corresponds exactly with the distribution of blood to them. Unless this correspondence is exact some alveoli will receive more air in proportion to their blood supply than others, and, as a consequence, the mixed arterial blood will be a mixture of more or less fully arterialized (aerated) blood with some of the consequences first discovered (anoxemia). It is probable indeed that in some way or other the air supply is proportioned to the blood supply whether by regulation through the muscular coats of the bronchioles or regulation of the blood distribution; but it is also certain that this proportioning is only an approximation" (Haldane J. S.: Respiration, Yale University Press, New Haven, Conn., 1922, 137).
Certain it is that all the alveoli are unequally supplied with air. Fluoroseopic observations of normal breathing prove this in the increased inflation of the zone of lung adjacent to the diaphragm during each inspiration. The control of air and blood, as suggested by Haldane, would scarcely afford an approximation in the proportioning. On the other hand, the air cell-capillary gear provides exact proportioning for each alveolus though not the same proportioning for all alveoli, else there could be no compensation in health nor adaption in disease. Inequalities in alveolar inflation and in the ventilation of intra-alveolar air would occur even in health, and could explain the mixtures of more or less perfectly aerated blood noted by Haldane.


latter demand properly controlled crucial animal experiments for proof that would he acceptable to them. They are unaware that one animal, even though denied the distinction of being included among laboratory species, has been giving conclusive although spontaneous demonstrations for centuries. Clinicians have recognized, in those demonstrations, the signs and symptoms of natural adaptations to intrathoracic diseases which occur in man.

Animals are divisible, as Miller4 showed, into those having thin pleura? and those having thick pleurae. Animals with thick pleurae have mediastina which are impervious to air and to water. The blood supplied to lung parenchyma and visceral pleura comes chiefly from the bronchial artery. Animals with thin pleurae have mediastina which are pervious to air and to water. Their lung parenchyma and visceral pleura obtain blood supply more largely from the pulmonary artery. Man is of the thick pleura type. Observations made on the usual laboratory animals (cat, dog, and rabbit), which are of the thin pleura type, are not directly applicable to man. Attempts to use such experimental observations directly in explanation of human physiologic and pathologic manifestations have led to confusions which are the bases of most misconceptions.

The evidence for an air cell-capillary gear is direct and indirect.

If the pulmonary artery is ligated or obstructed in man or in other animals with a thick pleura, the portion of the lung supplied is deprived of function. It atrophies, contracts and becomes airless. There is no infarction. If a bronchus is ligated or obstructed and pneumonia is not occasioned, the portion of lung supplied is deprive(l of function. It atrophies, contracts and becomes airless. There is no infarction.

In man, no portion of lung which has had either its air or pulmonary arterial supply destroyed can be functionally rehabilitated. Obviously, if either part of the air cell-capillary gear is incapacitated, the other part is simultaneously disabled.

If the rate and depth of respiration be increased above normal, the carbon dioxide content of the blood is reduced and apnea results. The unit volumes of blood in the pulmonary capillaries are increased because inflation is increased. Extraordinary aeration results, not because of tissue demands expressed through central nervous system control, but because of the coordinating action in breathing and circulatory units effected through the peripheral governor.

If the rate of cardiac contractions is suddenly increased by emotions, there is a correspondingly increased inflation; respirations are deeper and more rapid. There is a transient sense of air hunger. This is a common experience. Inspirations are involuntarily deeper because a larger unit volume of blood is delivered through pulmonary capillaries at higher pressures and inflation of alveoli is inevitable.

Animal experiments are thus far unsatisfactory. Negative-pressure cabinets are required to eliminate complicating factors incidental to disturbed intrapleural negative pressures and to the intratracheal positive pressure usually employed. Pneumothorax or hydrothorax induced in an animal with a thick pleura causes a contralateral compensatory emphysema proportionate tothe amount of air or water introduced. The animal will tolerate increasing


positive pressures so long as it is able to develop and to maintain the higher pulmonary arterial pressures required to develop the degree of contralateral compensatory emphysema required for external respiration. If the animal be fatigued previous to the experiment, the limit of tolerance is lowered because cardiac energies are less. If the same experiments are made on animals with a thin pleura, pneumothorax and hydrothorax are soon bilateral; compensation is less possible because intrapleural pressures of both chest cavities vary together and toleration is limited. Further confirmation lies in the fact that as animals with a thick pleura are progressively exhausted, the tolerance of intrapleural positive pressures falls until it equals the intolerance of animals with a thin pleura.

Observations have been made often enough during operations on human beings to furnish sufficiently reliable controls to satisfy all requirements for accuracy. The following examples are pertinent. When chronic pleuritic adhesions are divided before pulmonary elasticity has been permanently destroyed, the underlying lung, relieved of restraint, bulges outward if the patient is in fairly good shape even when no differential pressures are employed. If, during an open thoracotomy performed on a strong patient under positive pressure anesthesia, the positive pressures are reduced or stopped, the mediastinum bulges toward the open side. A focal parenchymatous hemorrhage is surrounded by a halo of emphysema. The focal pressure of a finger on visceral pleura will produce emphysema in the adjacent lung.

Other observations, after lobectomies, are notable. Each lobectomy causes greater compensatory emphysema in the remaining lung. The limit of lung capable of supporting life is that which will provide a vital capacity slightly in excess of the tidal air of rest.

Attempts to observe actions of the air cell-capillary gear in excised lungs are quite futile. Post-miortem changes occur rapidly. Blood pressures in the bronchial artery are absent. It is impossible to wash out all the blood from the capillaries. The effects of negative pressures are lost. The amount of force which must be applied to drive water through the pulmonary arteries exceeds the strength of the capillary endothelium and alveolar epithelium. Leakage through the air passage results. Inflation of lung bv forcing air into bronchi will not cause the pulmonary artery to aspirate fluid because the capillaries are plugged or collapsed. Injecting awater into the pulmonary veins causes no inflation just as might be expected because passive congestion is known to reduce vital capacity.

On the other hand, sufficient air pressure in the pulmonary artery will cause inflation of the lung and aspiration of fluid by the bronchus before the capillary walls are ruptured.

The simplest experimental demonstration is to ligate with suitable precautions in a strong living animal the pulmonary artery of one lung and to observe by Röntgen-ray examination the increased compensatory emphysema produced in the contralateral lung.

All the manifestations noted can be explained by the interactions of pulmonary inflation and deflation and of the amounts and pressures of blood in


pulmonary capillaries effected through the air cell-capillary gear. They are supplemented by the forces of blood in the bronchial arteries which fluctuate with physiologic activities of the lung, since they are under control of the systemic vasomotor mechanism.

Dunham had recognized the air cell-capillary governor and constructed a model to illustrate its actions as shown in Figure 180. The jar represents a pleural cavity. It contains a rubber bag, an air cell instead of a lung, connected with the middle glass tube instead of a bronchus. The glass tube at the left is connected with a coil of guinea pig's intestine glued to the air cell to imitate a pulmonary capillary. The tube at the right makes it possible to establish negative or positive (intrapleural) pressures. Suitable variations of air pressures exerted through the three tubes demonstrate the effects of breathing, fluctuations in unit volumes of blood delivered to the pulmonary capillaries, and chances in intrapleural pressures. The model has been criticized because the pulmonary capillaries have been represented as end vessels. A moment's consideration will explain how difficult or impossible it would have been to reproduce a complete circulation and how unnecessary. Whether or not some blood passes through the capillaries, a sufficient unit volume of blood would cause them to straighten and to elongate.

FIG. 180.- Dunham's original model of the air cell-capillary gear.

The adaptations are few in number. Their effectiveness varies directly with grades of competence or incompetence of the circulatory unit. Hence, there are many variations in extent though none in character of responses.


Collapse results when intrapleural negative pressures are neutralized byan open thorax or intrapleural exudates of appropriate amounts. One or both lungs may be affected. The extent of collapse is determined by elastic recoil of the lung opposed by the expansive forces exerted through the pulmonary and bronchial arteries. The lower the blood pressures, the greater the collapse. The greater the collapse, the less reserve air and blood in the lung, the less the residual air is ventilated, the less blood is in circulation and the more urgent the


need for contralateral compensation to protect external respiration. Unfortunately, the greater the need, the less the capacity to develop compensation.

Another important influence is the rapidity of reductions in negative pressures. As already noted, adjustments are most effective when they are induced gradually and thus occasion less unfavorable expenditures of energies.

Observations on the effect of war injuries and surgical wounds of the thorax have proved that ai competent individual can survive a wide parietal opening even when suddenly produced, whereas a comparable individual, made incompetent by exhaustion, exposure, starvation, hemorrhage, dehydration, and infection can barely tolerate a small opening. Likewise, it has long been known that a gradually increasing pleuritic exudate is of far less moment

FIG. 181.- Sheep's lung five weeks after ligation of the artery supplying the left upper lobe

than one produced rapidly, and that life is possible when the former exceeds in bulk amounts which are fatal in the latter. The extraordinary dangers of abrupt production of bilateral pneumothorax or the rapid formations of bilateral pleural effusions are well known, even to laymen. Graham and Bell,5 working with animals of the thin pleura type, in which, unlike man, a unilateral pneumothorax can not be maintained, found that the evil effects of the rapid bilateral reductions in intrapleural negative pressures could be measured by the size of parietal defects. Other experimental evidence has been presented to indicate that if negative pressures are reduced sufficiently gradually, almost normal inflation of the lung can be maintained when parietal defects are created subsequently.


All of these manifestations can be explained physiologically. Nature attempts to maintain suitable areas of alveolar air and blood requiring aeration in order to protect external respiration on which life depends. Should intrapleural negative pressures be reduced sufficiently gradually in a competent individual. there is reason to believe that redistributions of blood in both circuits and increased blood pressures can by action through the air cell-capillary gear maintain full normal inflation after the negative pressures are neutralized. When negative pressures are so rapidly reduced in the competent that compensator adjustment can not be made, or gradually reduced in the incompetent, who are unable to make adjustments, corresponding grades of deflation are caused.

FIG.182.-- Sheep’s lungs five months after ligation of the artery supplying the inferior margin of the left lower lobe, showing adhesions produced by simple thoracotomy

Pulmonary deflation inevitably measures an equivalent reduction in the corresponding pulmonary blood supply. It should be recalled that the pulmonary circulation is not provided with an effective vasomotor control. The blood normally destined to reach an area of deflated lung is delivered to the nearest areal in whlich the ilitracapillary pressures will permit circulation to occur. Wherever an increased amount of blood is driven through pulmonary cappilaries there is produced a corresponding degree of inflations, a compensatory emphysema. Thus, if the area of deflationl is small, the compensatory emphysema will appear in the same lobe; if larger, in the same lung; if larger still, in thee opposite lung. Compensation is produced by the amounts and pressures of blood delivered and is, therefore, most effective in the virile. Limitations of compensatory pressures can be estimated in the weak by determining reductions in circulatory competence as well as by measuring the deficit. in vital capacity.


These are nature's methods of maintaining the air alveolar capillary blood relationships that underlie external respiration. The effectiveness of the relationships is measured by vital capacity. If it is less than the tidal air of rest, life is impossible; if equal, bare existence is possible; and as it exceeds the tidal air of rest, it measures the margin of safety, the limits of possible physical and mental activities, including defense and repair.


Collapse is the deflation caused by the elastic recoil of lung when intrapleural negative pressures are neutralized. The amount of deflation is controlled by the opposing forces, the blood pressures in the pulmonary and bronchial arteries. External compression is produced when a force exceeding the pressure of an atmosphere acts on a lung from without and produces a grade of deflation corresponding to the preponderance of compressive force from without the lung over the expansive resistance of the blood pressures within the lung. On the release of external compression the lung reexpands to the grade of deflation due to collapse. This is often noted when chronic pleural exudates are divided or removed or when pleural effusions are removed.

Deformity of chest walls, intrathoracic tumors, aneurysms, intrapleural transudates, exudates, spontaneous and artificial pneumothorax, thoracoplasty and intrathoracic operations are causes of compression. An outline of changes produced by the common cause, pleuritic exudates, suffices for all.

Exudates formed in a pleural cavity free from adhesions eventually gravitate. Should they increase, the lung is floated upward so far as its hilum attachments permit. As the level of the fluid rises out of the costophrenic sinuses, the lung is elevated and negative pressures are reduced to zero after which external compression begins. It is exerted first on the supra-adjacent lung. This is exactly contrary to the effects of diaphragmatic contractions, which diminish pressures on supra-adjacent lung, and are readily seen fluoroscopically in the zone of increased inflation just above a contracting diaphragm.

Pulmonary vessels carrying low pressures are easily compressed: air cells are smaller; and this zone of lung becomes proportionately inactive physiologically so that the bronchial arterial blood supply is also correspondingly reduced by the usual vasomotor responses. The same series of changes, becoming more and more accentuated, develop as the exudate increases. With increasing exudation, the significant adaptive changes become more noticeable.

At first, compression of the pulmonary circulation diverts blood to the nearest margin of lung under less compression and creates there a zone of physiologic emphysema in the same lobe. Gradually, this extends to higher lobes and finally to the contralateral lung. Thus are produced zones of skodaic resonance and the contralateral emphysema long noted with pleuritic effusions.

Should the exudate form rapidly, as it often does in streptococcus infections, patients may succumb in a few hours because of their inability to deliver the requisite blood pressures to continue the adaptive emphysema. Physiologic emphysema, be it recalled, is merely a natural method of developing a sufficiently large area of functioning alveolar epithelium in one part of a lung to compensate


for reductions elsewhere, and thus to provide enough external respiration to support life. Again, the exudate may be produced more gradually, and, though it cause complete external compression of the entire lung on the affected side, the patient is able to survive. Moreover, the more gradual productions of exudates are favorable to the formation of adhesions between visceral and parietal pleurae and thus to localization of the process.

Another point is noteworthy. Intense irritation of visceral pleura tends to produce a subserositis or cortical pncumonitis that may be effective in reducing pulmonary elasticity, which interferes with vital capacity.

There is a parallelism between the adaptive processes in the cerebral and pulmonary circulations neither of which is under direct vasomotor control. Both can compensate for gradually increasing antagonistic pressures remarkably well. Both are lethally incapacitated if the counter pressures rise more rapidly than the patient can develop compensatory rises in blood pressures to offset their effects which are respectively cerebral anemia and impaired external respiration.


Internal compression differs from external in that the force is exerted within the lung instead of on it. The common causes are chronic passive congestion, pneumonias, tumors, foreign bodies and parenchymatous hemorrhages.

The effects of internal compression are the same as those arising from external compression. They differ in extent and in distribution. Compensatory emphysema develops about a focus of internal compression by the same air cell-capillary gear. The blood being diverted from a focus of greater to surrounding zones of lesser pressures produces an enveloping layer of hyperemia and consequent emphysema. If the foci of internal compression are sufficiently large and numerous, they can produce a manner of skodaic resonance and even lead to contralateral emphysema.

Should the force exerted by foci of internal compression be sufficient, large branches of pulmonary arteries, bronchial arteries or bronchi can be occluded. The effectsof such occlusions will be noted later.


Abnormalities in the arterial circulation are gradual occlusions of pulmonary arteries producing equally gradual increments in peripheral resistance, sudden occlusions of larger branches by injuries, ligations and emboli causing abrupt increments in peripheral resistance.

Gradual occlusions or obstructions to arterial circulation occur in chronic external and internal compressions of the lung in pleural effusions, tuberculosis, bronchiectasis and pathologic emphysema. High peripheral resistance leads to hypertension in the pulmonary circulation by extraordinary exertion of the right heart which results in hypertrophy and diminished reserve power. This explains why individuals so affected are especially sensitive to sudden changes in intrathoracic pressures and why all grades of pulmonary compression should be minimized.


Sudden occlusions or obstructions to main branches of the pulmonary artery are produced by ligation and by emboli. They cause a sudden shunting of considerable volumes of blood into other vessels and an abrupt rise in peripheral resistance. A strong individual tolerates such a change without notable variations in pulse rate or systemic blood pressures. A weakened heart may be incapacitated almost immediately. Permanent occlusion of a pulmonary artery causes atrophy, shrinkage and atelectasis in the lung affected. Ligation has been practiced and is still advocated in treating tuberculosis and bronchiectasis. It is in effect a physiologic lobectomy when the principal artery to a lobe is tied. But it does not cause gangrene or infarction because the parenchyma and visceral pleura are supplied by the bronchial arterial circulation.

Experiments indicate that life is possible after four-fifths of the pulmonary arterial circulation has been destroyed in successive stages, as in repeated lobectomies. Sudden occlusions of much more than three-fifths are incompatible with life.

Two erroneous notions about the effects of pulmonary embolism are prevalent. One is that embolism is the cause of gangrene and infarction of the lung; the other, that deaths from emboli can always be attributed to their size.

Embolism can be an indirect cause of gangrene or infarction if it produces secondarily enough internal compression to occlude a bronchial artery. Usually, gangrene and infarction are occasioned by thrombosis in pneumonia or occur in lungs with circulatory impairments incidental to cardiac lesions or consequent on pleural effusions.

The victims of pulmonary embolism expire or they recover, perhaps to die later of pneumonia. Survivors suffer only from such ill effects as may be attributed to a physiologic lobectomy. Deaths from pulmonary embolism in debilitated patients are often due to emboli sufficiently large to raise peripheral resistance abruptly above the limits of their restricted powers of cardiac compensation. There are, however, many deaths of relatively healthy individuals, as, for example, a few weeks after a simple herniotomy, which fail to be thus explained. Barcroft 6 has suggested that an embolus can be caught close to one of the nerve endings present in the pulmonary arterial wall and may, by irritation, interfere reflexly with normal cardiac impulses.

Operative removal of pulmonary emboli has been recommended and, in one instance, has been accomplished without killing the patient. Removal of emboli could not be sufficiently prompt to obviate the sudden deaths. It could therefore only be employed to rehabilitate a fraction of the pulmonary circulation or prevent embolic pneumonia. The very patients needing to have a portion of their pulmonary circulation restored are those so debilitated that thoracotomy would be a fatal burden. None could say that any given embolus was going to cause metastatic pneumonia, or that its removal would prevent the pneumonia or would help to reduce the dangers of septicemia. Hence, the removal or attempted removal of a pulmonary embolus is almost without exception a highly dramatic example of surgical malpractice.


Interference with the circulation in the pulmonary veins is due to obstructions caused by incompetence of the left heart or resulting from pulmonary compression. Both add to peripheral resistance in the pulmonary circulation.

If the obstruction develops abruptly and the use of positive pressure with too great anesthesia is a nice example, acute cardiac dilatation is constant and often fatal. Acute venous obstruction can become so critical that systematic venesection is indicated. Gradually increased obstruction leads to chronic passive congestion with its series of handicaps. Abrupt increments in chronic obstruction can cause edema of the lungs.


Bronchial arteries carry six times the pressure of the pulmonary artery, and, as stated, supply the major part of nutrition to lung parenchyma and visceral pleura. Hemorrhage from rupture of the bronchial artery into a bronchus causes exceptional deaths with hemoptysis; into a lung, the diffuse parenchymatous hemorrhagic infiltration called splenization. Obstruction of this blood supply can produce infarction and gangrene. The principal significance is surgical. Splenization is commonly a positive indication to excise the portion of lung affected. Incisions and resections of lung require accurate ligation of severed branches of the bronchial artery which are easily recognized by the spurting of red blood. Usually, it is safest to remove all lung bereft of its bronchial arterial blood supply.


Obstruction of bronchi deprives the corresponding lung of its supply of air and assures atelectasis because the air present at the time obstruction occurs is rapidly absorbed. Commonly, an obstruction of larger bronchi leads to pneumonia through infection added to the pressure exerted by retained secretions.

The surgical significance is clear. Bronchial defects must be repaired accurately, and, when such repair is impossible, the corresponding lung should be removed.


The more acute and diffuse the inflammation, the greater and more abrupt is the internal compression and increase in peripheral resistance in the pulmonary circulation. Processes of this type are more apt to be accompanied by restricted movements of the diaphragm on the affected side, which are almost certainly inhibited if pleurisy develops. The danger lies chiefly in the extra load thrown on the right heart in addition to the burdens of myocardial injuries from toxemia. It is particularly desirable to reduce the expenditures of energy by the heart. Benefits of early and repeated aspirations of pleural exudates introduced years ago by Bowditch 7 are too important to be neglected. Occasionally, the diaphragm is but little affected. Paralysis obtained by injecting the phrenic nerve can be helpful. The main obligation is to reduce cardiac labor in order to aid compensation and to increase pulmonary blood supply.

Chronic pneumonia reduces pulmonary elasticity and increases peripheral resistance in the pulmonary circuit.



Pathologic emphysema means reduced pulmonary elasticity, increased peripheral resistance in the pulmonary circulation and a limited capacity for compensation. It commonly indicates a narrower margin of safety than would be suspected. The right heart has been working for a considerable period against abnormally high peripheral resistance, and its store of reserve energy is subnormal. The bronchial arterial supply is reduced, and the powers of resistance and repair in both parenchyma and visceral pleura are accordingly restricted.

Intrathoracic operations on patients suffering from emphysema are extraordinarily hazardous. Such patients are likewise less able to tolerate pleural effusions and should be protected by early and repeated aspirations, preferably bv continuous one-way drainage.


Lung becomes atelectatic when its supplies of air or of blood from the pulmonary artery are stopped. It is physiologically inactive and receives the least blood through the bronchial arteries. Such lung atrophies and is cicatrized, both leading to contraction. Contraction is powerful enough to carry with it adjacent lung so that an area of atrophy, which is but a part of a lobe, may reduce the whole lobe to a small mass puckered about the hilum.

In the performance of intrathoracic operations, this must be considered. Lung that has become permanently atelectatic or is likely to become so is more safely excised as a rule.

Observations of the actions and reactions in the breathing and circulatory units occurring under physiologic and pathologic conditions show that nature attempts to maintain such interrelationships as will assure preservation of the basic function of external respiration with the largest margin of safety therein.

Control of the basic relationships, namely volume of ventilated air, area of alveolar epithelium, expanse of capillary endothelium and unit volume of blood, is in part in the central nervous system, but to a larger extent is vested peripherally in the air cell-capillary gear.

Vital capacity measures the efficacy of the relationships so that, in effect, nature strives to maintain a high normal vital capacity, a lead that therapy must follow.

Vital capacity is determined in the last analysis by the integrity of the pulmonary circulation affected by and affecting the state of the breathing unit. Preservation and rehabilitation of the pulmonary circulation, the quality, quantity and pressures of the blood delivered by it, are the most important part of intrathoracic therapeutics.


Nature's methods of meeting irritations from injury or disease by increasing resistance and hastening repair through reduced but not inhibited function are more evident and perhaps more significant in the chest than elsewhere.

Irritations affecting thoracic parietes or viscera almost constantly restrict parietal movements. Costal excursions may be more affected than diaphragmatic


or vice versa. Both are influenced because there is a common control. This is well illustrated if one side of the diaphragm is paralyzed by blocking the cervical portion of the phrenic nerve when there is an immediate, if transient, limitation in the costal excursions on that side.

The effects of restricted motion are to encourage somatic rest, to conserve energy, to limit dissemination of irritants and to prevent atrophy and hyphemia inevitable with nonuse. They may be noted in parietes, in lung and in pleurae.

Restricted motion in extrapleural parietes subject to irritation is of no especial moment other than assuring the richest blood supply compatible with minimal expenditures of energy.

Restricted motion in the parietes has very definite influence on the lungs. When restricted motion is considerable, as in more intense types of irritations, the diaphragm is more or less relaxed and is prone to be forced by intraabdominal positive pressures into an unusually high position. Pulmonary excursions are thereby reduced, and more important still, the total volume of lung or the grade of inflation is less. It has been found experimentally that a lung in approximately a mean position between extremes of inflation and deflation produced by inspiration and expiration receives the largest unit volume of blood with the least cardiac effort. Under such conditions the vessels are neither elongated nor tortuous and peripheralresistance is lowest. It has also been found that under these most favorable conditions pleuropulmonary resistance to infection is highest, and the rate of repair and functional rehabilitation is well-nigh doubled.

The effects of restricted parietal motion on the visceral and parietal pleura are significant since pleurisy is commonly the cause and frequently the danger of treatment. The parietal pleura receives its blood from the same vessels that supply the adjacent parietes so it is helped by restricted parietal motion. The visceral pleura is more effective in defense against pleural irritations than the parietal because it has a richer blood supply and a larger expanse. The gravity of serositis in general is determined by the excess rate of production of effusions over the rate of their absorption. Resistance of serous cavities is commensurate with their ability to maintain visceral and parietal reflections of serosve in apposition by absorption of exudates. Absorption from the pleural cavity is notoriously slow and slower still if visceral pleura is incapacitated by compression of exudates. Lung in the mean position between inflation and deflation provides its pleural surface with its richest blood supply. Hence, through diminished parietal motion, Nature protects the welfare of parietes, viscera, and pleurae.

Besides the foregoing adaptations in the breathing apparatus, there are responses in the circulatory unit. Satisfactory determinations of rates of flow and pressure of blood within the pulmonary circulation are yet to be made. Blood is forced through the lesser circuit in one-fifth the time under one-sixth the pressure required to drive it through the systemic. Fluctuations in blood pressure occur, perhaps synchronously with variations in systemic pressure, but are less in extent. Presumably, the margin of safety in the right heart


is as wide as in the left. However, nothing definite is known of the persistence and effects of pulmonary arterial hypertension.

Shunting of blood from one part of a lung to another or to the contralateral lung is attributable in the main to hydrodynamic influences. Anatomists have found nerve cells indicating possible vasomotor controls, but physiologists have not demonstrated corresponding functions. However, two peculiar reactions occur that may be caused by such influences. Massive collapse of an entire lung has followed injury to the opposite side of the chest. Such lung must be deflated, and deflation can result only from failure of air to enter the bronchi or from interference with delivery of blood through the pulmonary arteries. Massive collapse might be explained by a unilateral temporary occlusion near the bifurcation of the trachea were it not for the second reaction. Surgeons who have performed open thoracotomy without the protections from differential pressures have noted a sudden lateral shifting of the mediastinum which has been called fluttering or epilepsy of the mediastinum. This has been found experimentally to be associated with equally sudden variations in intrapleural negative pressures, and, in the absence of tracheal obstruction, was attributed to shunting of blood from one pulmonary artery to the other. Conceivably, such shunting might be explained by alternating kinking and unkinking of the pulmonary arteries, but alternating vasometer spasms seem more plausible.


The object of treatment is to protect and to restore the function of external respiration, which is estimated by the vital capacity. The means are to assure the integrity of both the circulatory and the breathing units. Available methods are few.


Abnormalities occur in the amount and distribution of blood in an organism. If the amount of blood has been reduced by hemorrhage or as the result of increased blood destruction or decreased hematopoiesis, prompt relief is attainable with transfusions. If the amount of blood in circulation is reduced because of exemia or the escape of plasma into the tissues, benefits follow intravenous administration of hypertonic glucose which may be given with and without insulin or with and without gum acacia. Each has its indications.

The importance of overcoming shock or states bordering on shock can be illustrated experimentally. If a robust dog is bled from a femoral artery until the heart stops, and an open thorax is then created promptly, the heart beats again for a considerable interval and more blood escapes from the artery. If, however, a dog in a state of prolonged and profound shock is similarly treated. the heart is not reactivated. In other words, little or no reserve blood is stored in pulmonary vessels during shock. Immediate compensatory or adaptive responses are impossible. Additional demands for oxygen can not be met because an important cog in the mechanism of external respiration is not working; hence, the need for providing an ample blood supply before attempting an operation and avoiding pulmonary compression, so far as possible, if that operation is intrathoracic.



Means to counteract lesions in the breathing apparatus are extrapleural and intrapleural.

Extrapleural procedures can restrict parietal movements, increase pulmonary deflation or produce compression.

Lesser degrees of restricted movements of the parietes can be obtained by bandages, swathes, and adhesive plaster, particularly if it is remembered that adhesive plaster splints are fully effective only when snugly applied, corsetlike, around the entire chest. Frequently, desirable benefits can be secured by further increasing the spontaneous restrictions to motion, particularly if the lung is placed simultaneously in that stage of deflation wherein the blood supply is most favorable. Then, benefits may be provided readily by section of the cervical portion of the phrenic nerve, if permanence is required, or by injecting it with cocain or with appropriate dilutions of alcohol should subsequent regeneration be desirable.

Extrapleural thoracoplastic operations are effective in producing permanent compression to aid in obliteration of cavities, particularly in tuberculosis and some forms of bronchiectasis. Failures of these operations, if patients are properly selected, can be attributed to insufficient costal resections or to having attempted too much at one stage.

The operations are dangerous because the patients are usually weakened by a losing struggle against an affection that is mainly unilateral, and during operation they must lie on the sounder side. A more serious handicap is cardiac incompetence. Increased peripheral resistance in the pulmonary circuit has resulted from the internal compression caused by the lesions. It is usually of long duration and has sufficed to reduce the margin of safety especially in the right heart. Fever, intoxication and anemia add burdens by causing myocardial degeneration. It is obvious why excisions of parts of only a few ribs can so alter distribution of blood in the pulmonary circulation that the greater labor required to effect compensation can lead through progressively increasing tachyeardia to lethal myocardial exhaustion.

Rest, digitalis, transfusions and intravenous administrations of glucose can be extremely valuable in surgical preparation. The greatest assurance of safety lies in performing these operations in stages with sufficient intervals to permit orderly readjustments in the circulatory apparatus. One stage too many is preferable to one stage too few. An incompleted operation can provide the dangers and distresses of a finished procedure without affording the benefits.

Intrapleural procedures are called closed when free pleural surfaces are not exposed to the air and open when they are so exposed.

Closed methods are drainage afforded by single or repeated aspirations or by the continuous removal of exudates accomplished by air-tight, intercostal tube drains, artificial pneumothorax and thoracotomies performed through preformed adhesions. The same sources of danger are present in all. One is the sudden entrance of air into free pleural space. The other is caused by rapid and forceful operating. Both have the same effect, abrupt changes in


intrathoracic pressures which require immediate readjustments in the pulmonary circulation. Likewise, the same precautions are to be observed. Aspirations and drainage through tubes should be gradual. Artificial pneumothorax should be induced slowly and at repeated sittings so/that intrapleural pressures are progressively reduced before positive pressures are created. Accuratehemostasis and gentleness are essential in closed thoracotomies.

Empyema will continue to be an important part of intrathoracic therapy. The best treatment is prevention, which means early diagnosis and immediate aspiration as first advised by Bowditch. 7 Fluoroscopy, essential to both, has * been insufficiently practiced and is more generally available since bedside roentgenray units have been made easily portable.

Early aspiration, perhaps closed drainage, even if it fail to abort empyema, can minimize its severity, extent and the necessity for rib resection. More important still, it reduces the persistence of diaphragmatic paralysis, which Pryor 8 showed to be so common and Middleton 9 found to be commensurate with reduced vital capacity and degrees of disability. Moreover, pulmonary deflation with its added burdens on the pulmonary circulation are minimized during aggressive phases of pneumonias when a slight shift in the patient's favor can obviate a fatality.            

Rib resections impair parietal mobility and are to be avoided whenever possible. On the other hand, it is usually unwise to delay performing this operation after the application of ordinary surgical principles indicates open drainage. Much may be accomplished by the use of surgical solution of chlorinated soda (Dakin's solution) to dissolve fibrin and by employing gentian violet as advocated by Keller,10 to disintegrate fibrous tissue.

After-care has been too often neglected. As soon as the acute process permits, breathing exercises are needed to stretch adhesions, to reactivate the diaphragm and to develop compensatory physiologic emphysema in order to maintain a high vital capacity.

Discussion of open methods for treating intrathoracic lesions is limited to consideration of ways of performing thoracotomy in the absence of pleural adhesions. Incisions should be designed to assure permanent air-tight healing, else open pyothorax is inevitable. It is the principal cause of postoperative disabilities and deaths. Differential pressure is required for safetv. Analgesia suffices. Both can be obtained by the proper use of gas-oxygen as developed by Gwathmey.11 It is wise to inject the phrenic nerve with 1 per cent cocaine. The immediate paralysis of the diaphragm makes operation easier and hastens recovery. There is less postoperative discomfort as the effects of cocaine last for four or more days. Effusions are more rapidly absorbed because the pulmonary blood supply is richest under these conditions and healing of lung tissue is favored. Operators differ in respect to drainage. Postoperative pleuritic effusions are constant, commonly of considerable volume, are slowly absorbed and offer a favorable medium for bacterial growth. They cause pulmonary compression and its train of evil influences. Air-tight drainage is possible, is effective, and is seldom to be regretted. Omission of drainage can be disastrous. While suturing the wound at the end of a thoracotomy,


it is important to obtain air-tight closure with lungs in full inflation and to obtain it in such a manner that the stitches mav not cause tension necroses.

After-care, if otherwise it can be called such, includes breathing and body exercises to restore parietal movements, pulmonary elasticity, and intrapleural negative pressures as steps in the rehabilitation of external respiration which can be measured by frequent estimations of vital capacity.


Life and activities are made possible by external respiration which is provided through coordinated actions of the breathing and circulatory units.

Coordinated actions of breathing and circulatory units maintain sufficient volumes of ventilated air in contact with a suitable area of alveolar epithelium to assure such interchange of gases through a similar expanse of capillary endothelium as will aerate equivalent amounts of blood. The interrelationships between the volumes of ventilated air, the area of alveolar epithelium, the expanse of capillary endothelium and the amounts of blood needing aeration must remain constant although they are constantly fluctuating with each breathing cycle and with variations in activities of the individual.

This constancy is assured under normal conditions by an arrangement of an air cell-capillary structure and function whereby fluctuations in degrees of inflation produce equivalent fluctuations in amounts of blood, and vice versa.

Under abnormal conditions, the relationship is modified through the air cell-capillary gear, so that additional areas of alveolar epithelium and of capillary endothelium are provided to compensate for such as may be temporarily or permanently inactivated.

External respiration in a normal individual is undisturbed whether lie is at rest or exercising full mental and physical powers. Disability is nil, and the limits of activities, including defense and repair, are measured by normal vital capacity.
External respiration in persons affected with incompetent breathing or circulatory units is impaired. This impairment restricts the development of mental and physical powers to the level at which the relationships between ventilated air, areas of alveolar epithelium, expanse of capillary endothelium and amounts of blood can no longer be kept constant. There is disability corresponding to the level. Activities including defense and repair are proportionately restricted, and the restriction is measured by the reduction in vital capacity. Preservation of vital capacity during the acute phases of diseases affecting the thorax and its contents and restoration of vital capacity thereafter are the therapeutic objectives.

Vital capacity measures external respiration which is largely under control of the air cell-capillary gear. Treatment seeks to renew parietal mobility. to reestablish normal intrapleural negative pressures and to restore pulmonary elasticity, at the same time providing for deliveries of suitable amounts of good blood by the right heart and preventing obstruction to the pulmonary circulation through incompetence of the left heart. The specific aim is to rehabilitate the air cell-capillary gear in order to secure the largest measure of external respiration. Achievement is measured by vital capacity.



The principal effects of intrathoracic diseases are the malfunctions due to lesions of heart and lungs. Both heart and lungs are possessed of remarkable powers of compensation and repair but, when seriously affected, cause enormous totals of distress, disability and death. Many lesions of heart and lungs from which spontaneous recovery is impossible can be remedied surgically by procedures already devised.

A wider application of known methods and the introduction of new procedures can be hastened by more exact knowledge of intrathoracic structures and functions.

At present, vital capacity is significant in estimating latent powers including defense and repair, in determining therapeutic procedures and in measuring results of treatment.

Vital capacity is regulated by the air cell-capillary gear, the weaker part of which is the circulatory segment. Progress in intrathoracic surgery depends largely on the realization of more effective measures to promote and to conserve the integrity of the pulmonary circulation.


The actions, reactions, and adaptation of the breathing and circulatory units above described will be discussed in reference to thoracic injuries in particular. Repetition, even though fatiguing, will emphasize the biologic principles underlying treatment, which, if correct, will be permanent however much methods may be improved.

Thoracic injuries affect the parietes alone, the viscera alone, or, most frequently, both parietal and visceral lesions are produced. Hemorrhage and the exposure of lacerated tissue to air and to infection, which occurs with all wounds, are of special significance because of the untoward effects of hemopneumopyothorax.

Chest injuries, severe enough to need treatment and not causing prompt death, produce an immediate defense reaction, restricted motion, particularly of the injured side or the side of greater injury.

The virtues of restricted motion in respect to defense and repair are two-fold. The blood supply exceeds deliveries made when there is immobilization. Continued limited motion, an incomplete interference with function, assures opportunities for the earliest and most complete restitution of function.

Thoracic parietes are benefited by a narrowing of intercostal spaces. Tension upon soft tissues is reduced, a richer blood supply is favored and the size of tissue defects is minimized. Simultaneously, diaphragmatic excursions are curtailed, first, through spasm; later, through relaxation due to fatigue, which, in the presence of intra-abdominal pressure, carries the affected side to an unusually high position. Some motion, even when there is paralysis of one side of the diaphragm, is provided by a contralateral tug with each inspiration so that here, too, conditions favorable to healing are provided.

Diminished motion of the parietes, and particularly of the relaxed diaphragm, not only decreases pulmonary excursions but also lowers the mean


pulmonary inflation. Hence there is a diminished volume of lung subjected to less motion. Observations made by Steiner12 upon patients suffering from irritations to lung and to the visceral or parietal pleurae, even before pneumonitis or pleuritis is demonstrable and though they may never develop, prove this reduction in size and motion of lung to be a natural defense reaction. Experimental evidence (Cloetta)13 Shows that these conditions provide for delivery of the richest blood supply to a unit volume of lung, and, quite as important, with the least expenditure of cardiac energy.

Pleural defense and repair is obviously affected by the same conditions since the blood supply to the parietal reflection is derived from the vessels supplying the chest wall and the diaphragm and the visceral layer is nourished (in man) chiefly by the bronchial arteries. Pulmonary inflation is also affected through injuries to the circulatory mechanism by hemothorax, by intrapulmonary hemorrhage and by variations in the volume and force of the blood delivered to the lung.

Hemothorax, usually accompanied by some pneumothorax, is the commonest complication. Blood, escaping from either visceral or parietal lacerations into a pleural cavity free from adhesions, tends to settle in the costophrenic angles so that intrapleural negative pressures are reduced and progressive degrees of collapse occur before the lung volume is affected by direct pressure. As the level of the extravasated fluids rises, pressure is exerted chiefly upon the contiguous lung which is also elevated. When the extrapulmonary pressure exceeds the tension within the pulmonary arteries, that portion of the lung is deflated. At the same time the blood which should have been delivered to the zone of lung under compression is shunted to the superadjacent lung, wherein, because of the air cell-capillary gear, a margin of compensator emphysema is produced. This emphysema produces the Skodaic resonance constantly recognizable above the level of pleural effusions. Should the hemorrhagic extravasation continue to increase, the same compensatory changes are augmented so long as the circulation remains competent. However, a point will be reached at which the excess blood diverted from the compressed lung can no longer be distributed only to homolateral lung, but will in part be shunted to the lung in the opposite side of the chest. Here will be produced by the same air cell-capillary action the contralateral emphysema so commonly noted with large pleural effusions and massive pulmonary consolidations.

When unilateral negative intrapleural pressures are nullified--i. e., equal to the atmospheric pressure--the lung is in a position of collapse or as it would be with an open thorax. Intrapleural pressures in excess of the atmospheric produce degrees of pulmonary compression which vary directly with the amount of positive pressure exerted upon the lung and indirectly with intrapulmonary intravascular tensions. The final state, complete, rapidly produced unilateral compression, is seldom compatible with life and therefore rarely encountered.

From a practical standpoint it is wise to examine these adaptations more closely. However they may be produced, by variable proportions of hemothorax, pneumothorax and pleuritic effusions, as unilateral intrapleural pressures grow progressively more positive, they are met, as has been shown experimentally,3 by increments in systemic blood pressures within the lesser circulation.


So long as intravascular tension within a lung exceeds the extrapulmonary pressure the unit volumes of blood delivered will remain approximately normal, but such deliveries are secured by a greater expenditure of cardiac energy. When the extrapulmonary pressure exceeds the intrapulmonary vascular tension, there occurs the inevitable shunting of blood from zones of greater to the nearest zones of lesser compression. The compensatory emphysema inevitable with increased blood deliveries assures proper aeration of excess blood, but it is obtained at the cost of increased cardiac labor which is proportionate to the adaptations demanded and is effective so long as circulatory competence persists.

Possibilities for compensation vary within wide limits. Our observations have shown that a competent circulation in man will adjust itself to an abrupt interruption to the blood supply of an entire lung without recognized embarrassment. A healthy monkey will tolerate an increase in unilateral intrathoracic pressure exceeding twice the atmospheric pressure without giving indications of immediate distress. On the other hand, dogs, in which neither unilateral pneumothorax nor unilateral hydrothorax can be maintained, have not such powers of adjustment. They can barely tolerate atmospheric intrapleural pressure and then the better when slowly induced so as to provide opportunity and time for a compensatory rise in blood pressures. When their heart muscle is fatigued by meeting repeated demands for compensation, dogs can tolerate progressively less reduction in negative pressure. Similarly, a fatigued monkey or a man handicapped by myocardial deficiency, hemorrhage or shock, may be unable to withstand even atmospheric pressure, although gradually induced.

Another complication due to lesions in the circulatory apparatus is intrapulmonary hemorrhage, which may arise from lacerated pulmonary arteries or veins or from severed bronchial arteries. Resultant interstitial extravasations reduce pulmonary elasticity and affect pulmonary inflation. The results vary from localized hematomas to a diffuse brawny infiltration or splenization that may involve an entire lobe and is in effect an infarction.

Intrapulmonary hemorrhages cause intrapulmonarv compression, and, like extrapulmonary compression, compel a shunting of blood from foci wherein intravascular tension is exceeded to surrounding lung wherein it is not exceeded, so that the same factors are active in causing compensatory emphysema.

Incompetence is the most significant of all injuries to the circulatory unit whether produced by myocardial deficiency, anemia or exemia alone or in combination, because of the limitations to compensations that make external respiration possible. So important is this phase of the problem that it is fair by way of emphasis to anticipate therapeutic discussion and state that those procedures offering the greatest immediate protection to the lesser circulation and the largest possibilities for its rehabilitation, both measurable by vital capacity, are most effective in intrathoracic therapy.


The aims of civil and of military surgery are identical, to provide the largest opportunities for functional recovery with the least danger, delay, and distress. Handicaps imposed by warfare too frequently restrict what


should be done to what can be done. Makeshift methods need not be discussed as they are born of unfavorable conditions and individual ingenuity. Interest centers in means to a satisfactory performance of thoracotomy under field conditions, for thereby alone can many of the severely wounded be saved from death and a very large proportion of all the wounded be returned promptly to duty.

Surgeons have been enabled to invade body cavities effectively by first learning to control subsequent inflammation of the lining membranes of those cavities. This is particularly true of the thorax. Appreciation of the need to control pleuritis by restricting pleural irritation and by conserving the defensive and reparative powers of the pleura is simultaneous with the realization that the great majority of deaths not immediately due to injury and much of the late disability are attributable to acute and chronic pleural inflammation.

The explanation is simple. Chest wounds provide all of the conditions favorable to the development of pleuritis, tissue laceration, unyielding costal parietes, hypertension due to hemothorax, exudation more rapid than absorption, foreign bodies, and the presence of bacteria. Acute pyothorax added to burdens of recent wounds and exposure is an extremely serious complication, particularly when, as often occurs, circumstances prevent adequate personal attention to the sick. Chronic pyothorax, even though treated as superbly as has been done by Keller,10 inevitably causes material disability.

Methods for controlling pleural irritation should cooperate with natural defense reactions. Mesothelium lining serous cavities, compared to other tissues, is exceptionally resistant to irritants and possessed of correspondingly high powers of regeneration, provided its blood supply is adequate.

The extent and character of serositis is determined by the intensity of irritations, their dissemination through motion, capillarity. and gravity, by the hypertension produced within the cavity and the subserous reactions which curtail the supply of blood. In other words, resistance of serous activities is commensurate with their ability to maintain their mesothelial surfaces in approximation. This depends upon a greater rate of absorption than of exudation and upon flexible parietes.

Pleural resistance is relatively lower than peritoneal because the rate of absorption of pleural effusions is less rapid than the rate of exudation and because parietal adaptability is virtually limited to the various positions which may be taken by the diaphragm.

A  major portion of pleural resistance is borne by the visceral reflection and this is subject to material reductions by the interferences with the circulations in the lung consequent upon injury. As already stated, the total volume of blood delivered to a normal lung is commensurate with activity of respiration and with circulatory competence. If, as the result of extrapulmonarv pressures (air, blood, exudate) or intrapulmonary pressures (interstitial hemorrhage), the pulmonary arterial pressure is surpassed, deflation occurs which amounts eventually to atelectasis when blood flow is inhibited. By the same token, if a large branch of the pulmonary artery is occluded, permanent atrophy of the lung supplied thereby is the result (figs. 181, 182). Presumably, in areas


of partial or complete deflation and in atrophy the bronchial arterial blood supply is proportionately reduced. If the bronchial supply is withheld, necrosis follows. Although Karsner and Ghoreyeb14 have shown that when the pressure in either the pulmonary or bronchial circulation is reduced to zero, blood may pass over from one to the other, it was not evident in wounded human lungs that this interchange was free enough to be effective. Therefore, in man, because the visceral pleura is supplied by the broncial arteries, the basic need is to maintain a degree of pulmonary inflation compatible with function because the bronchial arterial blood supply varies directly with functional activity.

Natural defense reactions indicate that the limited degree of inflation determined by reduced costal excursions and the high position of a relaxed diaphragm is the most propitious. According to Cloetta,13 the pulmonary vessels in at lung thus inflated are neither tortuous, as in greater deflation, nor elongated, as ill great inflation. In consequence, lungs thus inflated contain the maximum unit volume of blood delivered with the least cardiac effort.

Rapidity and extent of dissemination of intrapleural irritants are well illustrated by a simple hemothorax. As first demonstrated experimentally by Denny and Minot,15 and confirmed by the observations of Elliott and Henry16 upon the wounded, intrathoracic movements are sufficiently active to defibrinate the blood in the pleural cavity which does not coagulate promptly and to spread the exudate over the entire pleura free from previous adhesions. Delrez17 and Middleton18 showed that blood is so irritating to joint and chest serosa as to produce a serofibrinous serositis. Middleton18 found the chief irritants to be fibrin and fibrinoplastic substances. Not all of the latter are removed when coagulation occurs so that serum remains an active irritant.

These facts explain the diffused pleural reactions, varying from a serofibrinous to an organized exudate, noted when a thoracotomy is performed hours, days, or weeks after hemothorax has resulted from injury. They also help to explain why pleuritis must be combatted otherwise than peritonitis. Generalized pleuritis is the rule; generalized peritonitis, the exception. Thoracic parietal adaptability is restricted to the diaphragm; abdominal parietal adaptability is only restricted beneath the costal angles and in the pelvis. In both healing occurs by adhesions between irritated surfaces which must be in apposition. Hence routine drainage of the general pleural cavity, which is possible if properly done, is indicated, and attempted drainage of the general peritoneal cavity. which is physically and physiologically impossible, is contraindicated. Irritation of serosa provokes a very rapid serous effusion which occurs promptly with hemothorax and soon exceeds the amount of blood originally present. High position of the diaphragm is a constant accompaniment of hemothorax (Bradford,19 Elliott,'s Soltau20 ) and is due at first to inhibition of contractions and later to paresis. Steiner's12 observations upon the early effects and Pryor's 8 studies of the late effects of pleuritis show that the early upward (displacement of the diaphragm is almost constant and tends to persist rather than to recover unless specially treated.

The effects of pleuritis above mentioned are intensified by the presence of bacteria and their toxins. When the pleuritis is intense it provokes a cortical


subserous pneumonitis unless the lung be so devoid of circulation that reaction is impossible, and then, like a foreign body, it is an added burden to defense. Moreover, even when a lung is but partially compressed its circulation is inadequate for defense because it is only under such conditions, according to Karsner and Ash,21 that pulmonary embolism causes infarction.

The effects of collapse and compression of the lung upon the circulation, particularly the burdens thereby adcled to the heart, have been described. It is only necessary to add here that heart muscle, recently compelled to overexertion, is particularly susceptible to intoxication. Hearts, already fatigued and still compelled to work disadvantageously, tolerate so poorly even the limited absorption from pleuritic effusion that the margin of safety is widened if their physical load is reduced by pulmonary reinflation which decreases peripheral resistance notwithstanding greater absorption of toxins from the pleura through the improved circulation. Moreover, the improved circulation also raises local resistance and favors repair.


The capability of maintaining its lining membranes in apposition is a gauge of its powers of absorption and roughly measures resistance of a serous cavity. Accumulation of fluid or air means dead space, hypertension within the cavity, diminished blood supply and decreased resistance, and in the chest, added cardiac labor. Methods of treatment, which prevent separation of serous surfaces, or, after separation has occurred, reestablish and maintain sur- face contact, provide the best opportunities for immediate recovery, of course assuming that circulatory incompetence is given proper attention.

The degree of ultimate recovery is determined by the promptness and completeness of restoration of function. Since the identical factors controlling respiration affect the circulation, measures must be adapted to restore mobility of the parietes, especially the diaphragm, to reestablish normal intrapleural negative pressures and pulmonary elasticity.

The most important part of treatment is restriction of pleuritis; the next most important is overcoming the effects of pleuritis. If these demands are given due consideration, non essential detail will be omitted. The specific objectives are four:

(1) Parietal healing.- Permanent air-tight parietal closure is imperative, because the worst complication is open pyothorax, particularly if it occurs early and before the formation of limiting adhesions. Smooth healing of parietal pleura is especially advantageous. It limits pleural effusions and offers the most effective barrier to the extension of inflammation, the usual antecedent to open pyothorax, either from within the chest to the parietes, or vice versa. Also it is the best protection against persistent or recurrent empyema.
(2) Restriction of pleural effusions.- Pleuritic effusions interfere with all of the factors which together make normal respiration possible and they likewise decrease pleural resistance. They can be restricted by reducing pleural irritation but not enough to be absorbed as rapidly as they form. Their removal is


always desirable and often iniperative and may be accomplished without ill effects by closed (air-tight) drainage.
(3) Pulmonary inflation.- Underinflated or overinflated lung adds to cardiac labor and narrows the margin of safety. Lung tissue that can not be inflated, or, if inflated, can not remain inflated, is without function and therefore without the blood supply essential to defense and repair. In general, such tissue is a direct menace and shorld be removed.
(4) Pleural adhesions.- Early fibrinous adhesions are inevitable and desirable. They limit the progress and extensions of pleuritis. Experience in com-bating peritoneal adhesions and the work of Delrez17 and of Willems 22 in overcoming adhesions following arthritis have proved that the principle of inducing active motion as early as the infectious process permits and of continuing it to the limit of pain inhibition gives the best functional results. The same principle applied in the treatment of pleurisy is equally efficacious. The anatomic explanation for the return of function is the persistence of some more or less isolated mesothelial cells of the serosa beneath exudates, even after organization has occurred. If motion disrupts adhesions gradually and without causing hemorrhage, the mesothelial cells which have been exposed by the disruption can, because of their extraordinary regenerative powers, overgrow the adjacent tissue defects. Thus is made possible restoration of a pleural cavity after pleuritis which can be so perfect as to permit of normal function.

If these four points are kept constantly in mind, the details of treatment to be described will appear less petty. Attention to details secured the tissue repair upon which recovery of function depends, and when neglected, as was at times imperative during periods of stress, the average of results obtained was less satisfactory.


Injuries to the thorax affect the parietes alone, both the parietes and viscera, or, more infreqentlv, the viscera alone. The last are due to indirect violence, or to sudden and considerable changes in atmospheric pressure result- ing from near-by explosions. Wounds, perforating, penetrating, or tangential, are similar to those involving other structures, but are distinguishled by one peculiarity-their liability to produce an open thorax. They are also coin- plicated by wounds of the abdomen and spine more commonly than are wounds elsewhere.

In general, shell fragments cause greater tissue injury than bullets, and more commonly carry other foreign bodies and bacteria with them into the tissues. The physical injuries of the thorax are of the same type as injuries elsewhere; they include a central zone of tissue destruction, an intermediate zone of tissue injury, and  a surrounding zone of hemorrhagic infiltration, which merges into an outer margin of active hyperemia in the tissues still capable of defensive reaction. They are subject to the same bacterial contamination, the same implantation of foreign bodies, as are other wounds.


These comprise about 10 percent of chest wounds, and consist of contusions, lacerations, and punctures, with or without simple or compound fractures


 of the ribs, scapula, clavicle, or sternum. They are caused by crushing injuries, by the direct impact of spent missiles, and most commonly by tangential injuries from various types of projectiles.

In general such injuries require the same treatment as similar wounds elsewhere-thorough cleansing and at resection of the necrotic and dangerously devitalized tissues. The most important details are to avoid opening the pleural cavity and to protect the parietal pleura from infection. Contused skin and subcutaneous fat do not heal well, particularly if suture is necessary. As a rule wide cutaneous resections are avoidable, but when indicated some plastic flap closure can be made.

Except in the muscles of the erector spina, group and the muscles attached to the scapula, there is relatively little danger of gas gangrene, because chest muscles are not surrounded by unyielding structures, and are less liable to the pressure anemia caused by swelling after injury or by the constriction of bandages. As a consequence less radical excisions of injured portions of the chest musculature are required than in similar injuries elsewhere.

Fractures are treated upon general principles although in compound fractures of the ribs particular care is necessary to resect damaged tissues widely and to avoid additional injury to intercostal structures. Injuries to nerves and blood vessels about the clavicle are repaired as accurately as possible.

Perforating wounds of the parietes occurring so low in the lateral aspects of the chest as to affect only the costophrenic sinuses may occasion negligibly slight intrathoracic disturbances, and yet they may require thoracotomy to effect deep transpleural repair.

Injuries of the spine without cord lesions require particularly careful excision because of the liability to infection due to muscle injury and the presence of bone fragments. Lesions of the cord may demand laminectomy. When paralysis is complete, efforts to obtain primary healing are particularly desirable in order to reduce subsequent distress.


In 10 percent the visceral injuries are due to force transmitted from the parietal injury without penetration of the parietal pleura. In the remaining 80 percent the visceral lesion is caused by projectiles or indriven rib fragments. In these it is important that an air-tight closure of the pleura be secured whether it has been opened by the primary injury or at operation.

Variations in methods of operating upon wounds of the chest wall, from those outlined above, are determined by the extent of intrathoracic injury, the necessity for performing thoracotomy, and the general condition of the patient. Whether the parietal pleura is opened by the primary injury or at subsequent operation, a permanent air-tight closure is essential.

Ideal repair of defects in serous membranes is obtained by an accurate approximation of serous surfaces without undue tension, because anything foreign to serous surfaces, including other living tissues, is an irritant. When the defects in the parietal pleura can not be closed by approximation, as frequently occurs with multiple rib injuries, makeshifts must be provided. The best substitute for parietal pleura is visceral pleura. Distended lung is easily


stitched to the margins of the defect in the parietal pleura. Injuries low in the flanks quite often require this type of closure; frequently the diaphragm must be utilized, either alone or in conjunction with the visceral pleura. Adhesions about an injured area of serosa are unavoidable and their immediate establishment by suture is sometimes desirable to reeduce subsequent intra-pleural effusion. The next best substitute for parietal pleura is accurate apposition of the smoothest obtainable muscle surfaces. Failing these a covering can be obtained by using a pedicled skin flap.

Wounds of the diaphragin are always serious because of added intraabdominal complications and occasional pericardial involvement. Those on the right side are commonly associated with a flow of bile from the injured liver into the thorax, which is generally followed by a severe empyema. On the left side there is liability to diaphragmatic hernia. On both sides there may occur a back flow of urine from injured kidneys.

Diaphragmatic defects are best closed so as to effect a smooth approximation of intrathoracic serous surfaces. On the right side the edges of diaphragm everted against the lacerated liver aid in hemostasis; on the left, the irritation from exposed muscle is better tolerated by the peritoneum than by the pleura. Excision of lacerated edges of wounds in the diaphragm need not as a rule be extensive and should be as limited as conditions permit. Lockwood and Nixon23 believe that repair of these wounds is more important than that of any hollow or solid abdominal organ. Transpleural repair of the diaphragm is usually simple and accurate; transperitoneal repair may be impossible.

Considerations of injuries to thoracic contents may be almost restricted to wounds of the lung. Cardiac, intramediastinal and vascular injuries seldom come to operation. The effects of traumatism, direct and indirect, upon the lung are intrinsic and extrinsic, as they occasion pressure upon and irritation to the surface bv hemothorax and inflammatory exudates, or pressure within the parenchyma through interstitial infiltration with blood or transudate and by pneumonia. Tissue destruction, injury, and infection of the lung are of relatively little importance because of the natural resistance of lung tissue and its remarkable powers of repair, provided it has an adequate blood supply-a factor letermined principally by the degree of compression. Operations upon the lung are required to protect the pleural cavity and to make possible an immediate resumption of respiration, rather than to prevent complications in the lung itself.

Injury to the lung, whether caused directly by a projectile or an in driven rib fragment, or indirectly by the impact of a high-velocity projectile, or by sudden variations in atmospheric pressure, is evidenced by parenchymatous bleeding. If the parenchymatous injury is superficial and extends through the pleura a hemothorax results. Bradford I showed that the injured lung is the principal source of hemothorax.

Though some hemoptysis is constant, intrapulmonary hemorrhage due to laceration of the pulmonary artery is mainly interstitial and usually is limited, since lung tissue is capable of a high degree of spontaneous hemostasis for bleeding from pulmonary vessels. Hemorrhage from a freely opened bronchial artery, carrying six times the pressure of the pulmonary circulation, produces


a very much more extensive infiltration before the extravascular tension suffices to check the bleeding.

Interstitial hemorrhage may vary from a slight, limited, plum-colored effusion, which barely interferes with normal aeration, to a true, bright red, traumatic infarction or splenization. The danger from splenization arises from its interference with aeration and with pleural blood supply. Experimental evidence in this respect (Plate IV) confirms clinical observations. If this brawny hemorrhagic infiltration extends to the pleura, an intense reaction results. If such an area can not be aerated with the aid of increased intratracheal pressure, the affected portion of the lung is probably permanently injured, even if it be viable. The immediate danger is the production of empyema, a danger sufficient to warrant excision of the affected lung if recovery is doubtful.

Wounds of the large bronchi are seldom seen, as they are usually associated with rapidly fatal injuries. Such wounds when not immediately fatal and those in which lacerated tissue either in lung or parietes has a valve-like action and permits air to be forced or aspirated into the pleural cavity with each inspiration and not expelled at expiration produce the rare examples of extensive pneumothorax. A limited pneumothorax constantly accompanies lung wounds. Occlusion of a large bronchus causes not only a complete atelectasis of the corresponding lung, but leads to a fatal bronchopneumonia and purulent pleurisy by damming back secretions in the affected lung structures. Occlusion of a smaller bronchus causes only an atelectasis in the por- tion of lung supplied by that bronchus, which is evanescent if the obstruction is not permanent. Compression of bronchi, possibly in part due to muscle spasm, aided by the presence of intrabronchial mucus and blood, leads to a peculiar type of collapse, particularly where there is added some degree of external pulmonary compression. A part of a lobe or an entire lobe may be affected. The lung is flat and quite bloodless, the pleural surface is in puckered ridges, and it feels like meat. Reinflation is impossible. There is, as a rule, an associated crushing injury of the overlying parietes without direct injury of the affected lung. This type of injury, carnification, demands excision. Infection of such tissue with anaerobes is the usual cause of the stinking types of pulmonary gangrene.

Wounds of the pericardium and heart and of the vessels within the mediastinum are rarely seen at operation. Hemopericardium, in addition to causing serositis, diminishes the effectiveness of the heart action by the pressure exerted. This is demonstrable experimentally, leads to reduction in blood pressures, and is in effect identical with the cardiac tamponade of pericardial effusion.


These are limited to pulmonary collapse and intrapulmonary hemorrhages, with or without a complicating hemothorax. Their treatment by conservative or radical methods is determined by their severity.





Of paramount importance in determining the approximate extent and nature of intrathoracic injury from a missile or a bayonet thrust is an appreciation of the anatomic relationships with particular reference to surface landmarks and the vital underlying structures. Anatomic models or charts will eliminate much uncertainty in plotting the course of a missile between known points of entrance and exit. Deflection by bony structures and alterations in the relationship of intrathoracic organs occasioned by changes in intrapleural pressure must enter into the proper evaluation of this determination.

Alterations in the contour of the chest and displacement of the cardiac impulse are common results of intrapleural accumulations. Obliteration of interspaces may arise under the same circumstances whether the accumulation be fluid or air. The respiratory movements should be carefully analyzed in their component factors, costal and diaphragmatic. Aside from the definite changes which would arise from alterations in the position of the diaphragm, the inhibitory influence of trauma on the movements of the chest, either directly or remotely applied, must be borne in mind.

In general, the remaining physical signs are dependent upon air content and sound transmission. Air content, elicited by percussion, may be either increased or decreased in chest wounds. Increase in air content is denoted by a lower pitched (hyperresonant or tympanitic) note and, resulting from a thoracic injury, indicates emphysema or pneumothorax. Reduction in air content leads to a higher pitched (impaired to flat) note and under the present consideration suggests a pleural accumulation of fluid (blood or chyle), intrapulmonary hemorrhage, or pulmonary collapse. In differentiating these conditions the position of the heart is helpful. Pneumothorax and fluid accumulations in the pleural cavity lead to definite cardiac displacement toward the opposite side. Displacement of the heart toward the side of the altered note suggests massive pulmonary collapse. Cardiac displacement is unusual in emphysema and uncomplicated intrapulmonary hemorrhage.

The final consideration in the diagnosis of intrathoracic injuries devolves on the question of the transmission of sound vibrations as determined by the palpating hand and listening ear. Alterations in tactile fremitus, breath sounds and vocal resonance are dependent upon changes in the conducting media between the larynx and the parietes. All conditions introducing new media of conduction or interfering with the normal channels of transmission, viz, the bronchial tree, tend to reduce or obliterate fremitus, breath and voice sounds. Thus fluid or air in the pleural cavity will decrease sound transmission by the introduction of an aldded medium, while intrapulmonary hemorrhage leads to the same result usually by obstruction of the bronchi. An apparent departure from this rule will be discussed under the reversed circumstance, namely, reduction of the media of conduction. Obviously, when the media of sound transmission are reduced, fremitus, breath and voice sounds are increased. For the sake of clarity, an approach of breath and voice sounds to the basic type in the larynx or trachea is considered an increase irrespective of pitch. Unification of media, such as occurs in pneumothorax with a patent


bronchus, is attended bv a great increase in tactile fremitus, breath sounds (cavernous) and voice sounds (pectoriloquy). Furthermore, certain cases of pulmonary infarction and of massive collapse may show by reason of a unified medium either increase or decrease in these signs of sound conduction dependent on patency or occlusion of the communicating bronchi.

Further development, of the physical signs in chest wounds from these fundamentals are suggested by an appreciation of the physics and the pathologic physiology of fluid and air accumulations in the pleural cavity. Skodaic resonance, movable dullness, succussion splash, egophony, whispering pectoriloquy, metallic tinkle, and bell tympany are to be considered in this relation. The peculiar crackle of mediastinal emphysema was first described in the chest wounds of the Great War.


The problem may be reduced to two fundamental questions: Does the injury warrant intervention, particularly intrathoracic intervention? If so, what methods are indicated?

Immediately arise the questions of diagnosis and of prognosis. The nature of intrathoracic injuries can be more accurately and rapidly determined by fluoroscopy than by any other means. The prognosis may be established at the same time.

Injuries involving the chest wall alone usually cause a reduction of motion upon the affected side, and can easily be mistaken for penetrating or perforating wounds if the patient is not examined fluoroscopically. The danger in injuries of the thoracic walls lies in the liabiliy to subsequent pleurisy by extension. if infection with a virulent organism develops. This danger can be reduced by immediate excision, which is rarely to be regretted and is too often at source of serious consequences if omitted. Exploration of the tracks of rifle bullets showing clean wounds of entrance and exit frequently reveals severe tissue lacerations and unsuspected rib fractures, and occasionally also discloses foreign bodies in addition to the tar drop blood clots constantly present.

Wounds of the parietes associated with intrathoracie injuries, even more than those involving only the parietes, demand accurate excision and primary suture to prevent the development of empyema. The larger the wound and the greater the destruction, the greater the urgency for intervention. Differences of opinion arise over the treatment of lesser injuries. Injuries to the lower margins of the ribs often cause hemorrhage from the intercostal vessels with bleeding into the chest. Penetrating or perforating w-un1s caused by small shell fragments and bullets not infrequently cause incomplete fractures of the ribs, with the result that bone fragments are driven through the pleura. perhaps into the lung. Even at the wound of exit from the chest bone fragments may project into the pleural cavity, and the intercostal vessels be injured. These slight rib injuries can not be recognized under the fluoroscope. and frequently escape detection whenl plates are made. Their significance lies in the fact that bone fragments are the most dangerous foreign bodies. From these considerations it is evident that virtually all wounds of the chest wall should be treated bv radical operative methods. though not necessarilv by thoracotomy.

Indications for performing thoracotomy to cope with intrathoracic lesions are three: (1)Those based upon preoperative findings; (2) those based upon operative findings; (3) uncomplicated hemothorax.


This group includes retained foreign bodies, unless very small, particularly if they are free in the pleural cavity; extensive splenization (Duval)24 or serious hemorrhage appearing externally; suspected wounds of the diaphragm (Lockwood and Nixon);23 and acute infection (Gask) 25 which can be assumed when treatment is delayed.


Gask 23 in particular advocated proceeding withoperation within the chest, when parietal operation disclosed serious internal lesions. Duval’s axiom, "The general rules of surgery must be applied to wounds of the lung," gained wider recognition as benefits attainable by operation were realized.
Hemothorax.- If hemorrhage may be called a wound complication, hemorrhage is the commonest and most important complication of chest injuries. Reports upon thoracic injuries by internists center about hemothorax, its physical signs and symptoms, the best time and method of aspiration, and the late complications. Lockwood and Nixon 23 saw its dangers, although they felt the dangers were insufficient to warrant thoracotomy. Gask 26 in a post-bellum report was unprepared to say that simple hemothorax of moderate size should be treated by operation. At the close of the war there was a growing sentiment in favor of open operation in the treatment of extensive hemothorax, particularly if massive clotting hadsoccurred. Aspiration remained the method of choice in treating limited hemothorax as it led to early recovery and return to duty.

The arguments against operative treatment of hemothorax are that conservative treatment, early aspiration with air replacement as practiced by Bastianelli,27 or aspiration after 7 to 10 days, with or without oxygen replacement, gives better results, particularly a lower immediate mortality rate, and reduces the pressure upon surgical facilities during periods of active fighting. The indications from the standpoint of morbid physiology are definitely for the earliest elimination of pleural irritation and pulmonary compression, the inevitable consequences of hemothorax and pneumothorax.

Hemothorax is of itself seldom sufficient to cause death either through pressure or acute anemia. The large amount of bloody fluid which is present after a few hours, its dilution as compared to normal blood, its liability to increase after active hemorrhage has ceased, prove that a pure hemothorax has a transient existence, and that it is only a few minutes before a secondary pleuritic effusion is added. In the presence of clots there is still more intense irritation, a fact which led Elliott,16 Davies28 and others to advocate open operation for their removal.

Later results of hemothorax are equally serious. Bradford 29 noted a decreased absorption rate of fluid and gas from the pleura attributable to the fibrinous exudate. Delayed absorption of hemothorax has been repeatedly mentioned, and, according to Davies,23 it is far from being the exception.


Absorption is not always promoted by aspiration, as evidenced by the example reported lbv Tuffier30 of a hemotliorax which persisted for 15 months in spite of 27 aspirations.

As the result of pulmonary compression, vicious adhesions, thickened pleura, and an immobile diaphragm, there is deficient expansion, shoulder drop, scoliosis, parenchymatous sclerosis (Tuflier),30 and incapacitating dyspnea. The dyspnea may last for months or even years, and, according to Leslie,31 is largely dependent upon the degree of permanent collapse of lung tissue.

The possibility of secondary infection of a hemotlhorax adds another and more dangerous complication. Gask 25 found that nearly all war wounds are contaminated. Soltau,32 in his bacteriologic studies of empyema, found gas- producing organisms in 48 percent, streptococci in 40 percent, and organisms from the lungs in 12 percent. According to Elliott,16 one-fourth of the hemothoraces became septic no matter what type of missile caused the wound; of these one-half died, and one-third of those who recovered were invalided out of the service, making a total of 16 percent in which death or incapacity was due to infected hemothorax. Moreover, it is difficult to recognize the presence of infection in a hemothorax since a simple hemothorax usually runs a febrile course and because the bacteria, as pointed out by Leslie,31 may be in the clotted portion of the exudate, so that the fluid portion obtained by early aspiration may be sterile. Open drainage was generally held to offer the best chance of recovery from this type of pleuritis though closed drainage would have been more efficacious. Gask 25 recommended thoracotomy with mechanical pleural cleansing and immediate closure. Statistics are not available to indicate the proportion of the 75 percent of hemothoraces which did not become infected and yet contributed to death and incapacity through pleural irritation and pressure. The total loss in man power due to hemothorax alone can be reasonably estimated at 25 to 35 percent of those receiving intrathoracic injuries.

If it be granted that thoracotomy has become a safe operation, then cumulative evidence indicates that immediate open operation is the sane treatment for all but the limited degrees of hemothorax, even if restricted to the removal of the liquid and coagula. In addition the lung injuries can be accurately repaired, foreign bodies removed, the inside of the thorax inspected and needed attention given to additional injuries not otherwise recognizable. It is still more significant that after these precautions have been taken the lung may be inflated to normal limits and the chest closed, with an immediate return of nearly normal respiration and circulation, factors which provide the most favorable conditions for defense in a contaminated pleural cavity and assure the earliest resumption of activity.

The splendid results reported by Bastianelli 27 in treating chest wounds by immediate aspiration and air replacement, reinforced by the wide acceptance of the value of artificial pneumothorax in the treatment of certain types of pulmonary tuberculosis, demand a reasonable answer.

According to Bastianelli, the cushion of air separating the viscerae and parietal surfaces prevents the formation of pleural adhesions, and the positive intrapleural pressure controls hemorrhage from the wounded lung. Recovery


is prompt because vicious pleural adhesions are prevented, and the repair of the lacerated lung is fostered by immobility and collapse.

Opposed to these theories are definite facts. Air is a serous membrane irritant and elimination of irritation is a basic principle in the prevention as well as the treatment of pleuritis. Positive intrapleural pressure, high enough to stop hemorrhage by pulmonary compression, compresses the homolateral lung, produces contralateral emphysema, and interferes with pulmonary circulation. Compressed lung means contracted lung which is often difficult or impossible of reinflation. If reinflation occurs, a universal adhesive pleuritis is certain. The exact conditions for minimizing the resistance to the flow of blood through the lung, as established by Cloetta,13 a reduction of mean pulmonary inflation, are met by diaphragmatic paralysis. Diaphragmatic paralysis lasting four or five days can be established by infiltrating the phrenic nerve trunk with 1 percent cocaine. If this procedure be coupled with closure of the chest without residual pneumothorax, the best conditions for repair and defense have been provided for both lung and pleura.



An individual suffering from a recent chest wound is in need of immediate physical and psychic rest. This can be provided by the administration of a full physiological dose of morphine to reduce coughing and to minimize respiratory effort. Care should be taken in examining and dressing the wound to permit the most comfortable position to be assumed and to avoid exertion on the part of the patient. First-aid should be restricted to hemostasis in the parietal wound and temporary closure of an open thorax. The latter is best accomplished by strapping a firm, thick broad pad over the wound with adhesive plaster, and then fixing a tight swathe about the chest. Provisional suturing of these wounds often leads to tissue emphysema, which interferes with the circulation and materially reduces resistance.

A second dose of morphine is given one-half hour before operation. If there is no question of an abdominal wound, water or hot drinks should be given freely, even just before anesthesia begins. Smoking is a great solace and does no harm if coughing has ceased. Anything to reduce distress is desirable.

Uncomplicated chest wounds are not prone to produce shock, but when there is considerable pulmonary compression they cause a high venous pressure due to interference with pulmonary circulation. If intravenous medication is indicated, it must be given very cautiously in order to protect an overburdened right heart. It is often wiser to give a transfusion of blood after reinflation has been established at operation. A grave anemia is often masked by an unusual distribution of blood because of the reduced content of the compressed pulmonary vessels. A blood transfusion after the evacuation of a large hemothorax and after satisfactory hemostasis has been effected will materially reduce danger.
A diagnosis of the probable nature and extent of intrathoracic involvement is furnished by physical and fluoroscopic examinations. One should


determine if possible the extent of hemothorax and pneumothorax, the presence of extensive intrapulmonary hemorrhage, the presence of extensive clots, the evidence of contralateral injury, the size, position, and motility of retained foreign bodies, the presence of bone injuries, the position and mobility of the diaphragm, the size and position of the heart, and the probable path of the missile.

A moderate-sized hemothorax where there are no adhesions darkens the whole side of a chest as these patients are examined in a recumbent position. Clots when visible are apt to be seen low along the spinel and do not move. Intrapulmonary hemorrhages can be seen only when the hemothorax is limited. The shadow corresponds to a lobe and moves with respiratory effort. Contralateral involvement, except for mediastinal displacement, is uncommon. Gask25 warns against operation when there is evidence of contralateral collapse, but the danger of this complication is avoided when positive-pressure anesthesia is used. The location of foreign bodies is important when theyare small and show by transmitted motion thatthey are close to important vessels or the heart. Otherwise there may be a temptation to omit removing them. Their position usually changes when the chest is opened, but as a rule rule foreign bodies are easily found. Positive evidence of bone injury is valuable; negative evidence is worthless. Its existence must be assumed until disproved by actual exposure at operation. The position and movement of the diaphragm help to determine the presence of injury of that structure and the level of operative incisions.

The heart's position is an excellent index of mediastinal deviation: the size of its shadow may tell of cardiac or pericardial injury. Projection of lines between the wounds of entrance and exit in perforating injuries or between the wound of entrance and the missile in penetrating injuries indicates the structures probably involved. The character of the injuries may be guessed from the size and shape of a retained shell fragment, or, if a bullet, from its shape and path.


It is usually held that the use of differential pressure is unnecessary. The writer does not hesitate to assert that positive-pressure anesthesia, or. better, analgesia, is the most important single factor in successful thoracotomy under war conditions. A full preoperative dose of morphine permits induction of an even, deep analgesia with nitrous oxide and oxygen without increasing the gas-oxygen ration above 3 :1. This was proved by Cannon's 33 observations to be the limit of safety in the presence of shock. After the induction stage the proportion of nitrous oxide can be reduced occasionally to less than 50 percent.

The American Red Cross nitrous oxide-oxygen apparatus devised by Gwathmev fulfilled every requirement. A rough gauge to measure the flow of gases, a mixing bag to serve as a pressure indicator, and a close-fitting face-piece are the only essentials. Pressure is furnished from the gas and oxygen tanks, controlled by regulating the flow of the gases, and delivered to the trachea by holding the facepiece snugly in position. The pressure can be 


varied as gradually or as suddenly as desired. The physical effort required is not great.

The dangers are real. Anesthetics delivered under pressure are more rapidly taken up by the blood. The more rapid absorption is due to increase in the volume of air containing the anesthetic, in the area of air cell epithelium, in the expanse of capillary endothelium, and in the amount of intracapillary blood due to greater inflation from intratracheal pressure. Anesthetics, when given under pressure in undue concentration, produce toxic effects which can appear suddenly and be rapidly fatal. A second danger is too great pressure. Normal heart and lungs can tolerate an excess pressure, but when the heart muscle and pulmonary circulation arc both impaired the margin of safety is much narrower. In general, positive pressure up to 16 mm. of mercury is safe and suffices for all purposes. Beyond that point compression of the pulmonary capillary bed can provoke an acute dilatation of the right heart. A few control observations with a manometer in the circuit show the degree of distention of the mixing bag which corresponds to a pressure of 16 mm. of mercury. It is rarely expedient to exceed this limit.

In practice the analgesia and hypertension are gradually induced with the patient in position for operation (fig.183). During the induction stage nothing is done to the wound. When deep analgesia has been reached, dressings are cut away and the skin prepared by dry shaving and scrubbing with ether. By this time inflation has been accomplished, so that uncovering a sucking wound occasions little increased respiratory distress.

Various stages of operation can be carried out under the same light anesthesia and the pressure varied as required. Repeated gradual inflation and deflation do not cause variations in the systemic blood pressure which can be detected by clinical methods. Both sides of the chest can be opened without evident distress to the patient. Graham and Bell 5 showed that if an opening in the chest wall of dogs exceeds certain dimensions, the aspiration of air into the lungs ceases. The size of the opening compatible with life in dogs holds relatively true for man provided man's circulation is deficient, c. g., with shock or acute anemia, when the powers of circulatory compensation are materially reduced. The use of positive pressure obviates this complication, as well as most of the coughing and dyspnea which occur when a chest is opened, particularly when a considerable hemothorax is present.

Another important point is the avoidance of traction upon the mediastinum. Traction upon a lung can cause in man a rapid drop of blood pressure of 40 mm. or even more. A deflated lung can not be brought outside of the chest from a position of collapse without traction, but a deflated lobe can be guided toward the thoracotomy opening and delivered without the least traction as inflation is produced.

Clinical results are confirmatory. Patients anesthetized under positive pressure as a rule left the operating room in better condition than at entrance, and the improvement was not evanescent. Operations need not be hurried. After the wound is closed the nitrous oxide is stopped and oxygen under decreasing pressure is given for the few minutes required for a return of consciousness.


Contrary opinions were held by other surgeons, notably Gask,25 Duval, 24 Depage, 34 and Lockwood, 23 who advocated open anesthesia or local anesthesia. Surgeons using their methods worked in the same units with others who use positive-pressure gas and oxygen analgesia. According to medical officers in charge of chest wards, patients of the former were less comfortable and had a more protracted convalescence.


Among the objections to routine excision of all chest wounds are the dangers of anesthesia and the time and extra nursing required. The use of gas as described is a satisfactory answer. Positive pressure is a cure and preventative of pulmonary collapse. Pneumonia after its use is infrequent. Excisions can be done more thoroughly, rapidly, and without the distress of local anesthesia. Accidental opening of the pleural cavity is of little moment, as deflation does not occur and the pleural rent can be accurately repaired.

Disclosure of unsuspected deep injuries is a common experience when parietal wounds are excised as a routine. Frequently a sucking wound is hidden by change of posture so that an open thorax may be suddenly encountered. Parietal excision should always be undertaken with a view to primary closure, because this becomes obligatory if a penetrating wound is encountered.

A satisfactory method of muscle excision consists in splitting between muscle bundles on either side of the injured area and carrying this splitting for some distance in both directions before dividing transversely. This block resection permits the later approximation of smooth surfaces at the point of injury, removes the cut fibers farthest from this point, and interferes least with blood and nerve supply. Excision done in this manner, often so restricted that bruised muscle was retained in order to make closure possible, was not followed in our experience by gas gangrene in chest wounds in a single instance, though gas gangrene occurred in other wounds in several patients who had multiple injuries.

Fractured ribs should be widely resected, and, when possible, the cut ends should be snugly covered with soft tissue, periosteum, or muscle, to reduce the chances of bone infection. Hemostasis may require the use of bone wax. Fractures of the scapula, particularly those caused by shell fragments, are less easily managed. Wide resection is indicated, but the bony framework and periosteum should, if possible, be preserved to make regeneration possible. Denuded bone must be removed; pockets should be collapsed to obliterated dead space. Injuries to the sternum are easily cleansed with gouge or rongeur. Compound fractures of the spine require adequate exposure, radical resections, and great care in removing loose fragments. Bone fragments are the most dangerous foreign bodies in the presence of infection. Excision of skin should secure a clean margin that bleeds freely and nothing more. It is safer to take chances with slightly traumatized skin than to run the risk of necrosis from excessive suture tension.


Positive indications for thoracotomy will generally be recognized before operation. Occasionally a decision will rest upon operative findings. A bullet


may perforate a rib without causing a complete fracture and carry with it splinters of the compact, inner layer into the lung; or, there may be an unrecognized comminuted fracture of the ribs associated with pleural and pulmonary lacerations and only a moderate hemothorax. Similar and even more extensive injuries may be found near the wound of exit. Usually the question is how to proceed rather than whether or not to operate. Three possibilities are offered: First, to do a limited thoracotomy at the entrance and exit wounds, a compromise between radical operation and aspiration; second, to proceed through the wound and resect the ribs widely enough to give the approach needed for deep repair; third, to repair the entrance and exit wounds and then to open the chest through uninjured tissues at a point of election.

By limited thoracotomy is meant: Intrathoracic operations performed through the wounds of entrance or exit with little or no additional enlargement, never enough to permit exploration by the pleural cavity. It is indicated when wounds are small, when destruction is relatively slight, and when the lung can be reinflated. In the absence of or after removal of hemothorax inflation of a deflated lung causes the wounds of entrance and exit in the visceral pleura to approach the corresponding wounds in the parietal pleura. A metal nozzle (fig. 187) attached to a Potain aspirator is introduced through the opening in the parietal pleura, and the fluid blood removed by directing the tip into various dependent pockets while the lung is partly deflated. When inflation is produced the visceral injury floats up to the parietal. If it is small and dry, it need not be repaired. If it is bleeding or if air is escaping, the lung can be secured by forceps at the margin of the wound, and a cone of lung with the wound at the apex can be gently brought out through the opening in the parietal pleura. A purse string suture closure, as in intestinal perforation, suffices. If there has been some rib injury, and bone fragments have been driven into the lung, a small incision through the posterior layer of periosteum exposed by the rib resection will permit excision of the lacerated lung tissue and removal of the bone fragments. After the lung is repaired it is wise to suture the lung wound into the defect of the parietal pleura. Adhesions between these areas are certain under any condition and are restricted by this fixation. This simple method is particularly useful in through-and-through wounds, but less so in penetrating wounds. It can be applied in about 40 percent of all chest injuries, the proportion varying with the type of fighting. The defects of this method lie in the possibility of leaving clots behind and of overlooking rib fragments.

The majority of more serious chest injuries permit no latitude in selecting the most desirable route for surgical approach. Parietal excision and repair alone are a burden to a man in weakened condition, so that the intrathoracic repair must be attempted by enlarging the wound of entry or exit. This avenue of approach gives a more or less unsatisfactory deep exposure, usually better for diaphragmatic injuries then for pulmonary because a retracted, compressed, and uninflatable lung can not be manipulated readily. As a rule there is a tendency to do too little under these trying conditions. A desire to avoid immediate death of a patient is apt to prevent the removal of injured lung whose presence is incompatible with recovery.


Methods of gaining deep exposure are deterinitied by the necessity of getting a satisfactory parietal closure. The preservation of parietal pleura is most essential. This layer is no more amenable to plastic surgery than a wet drum head. Each injury is a separate problem, the most desirable solution of which is an approximation of the type of repair described in the elective operation.

Thoracotomy of election is designed to give free access to one side of the chest in order to permit the dislocation and manipulation of each lobe of the lung and repair of the diaphragm. It may be performed before or after the plarietal excision, depending upon the type of injurv. The chest should be opened on a level with the root of the lung and in a manner favorable to closure. Intercostal incisions are easiest to make but give less satisfactory exposure, and their healing is less certain. Cowell 35 designed a transcostal incision which has much to recommend it when deflation is present, but which is not applicable when inflation is relatively normal. Rib resection can be done very quickly and safely, gives a maximum exposure with minimum damage, and provides the best opportunities for repair. A piece of the fourth or fifth rib longer than the width of the operator's hand is removed so that the mid-point

FIG. 183.- Patient in position for operation. Line of incision for a thoracotomy of election

of the defect is slightly nearer the angle of the rib than the sternum. In this area one may secure the maximum effects of rib rotation with the greatest separation for operation and the greatest approximation for closure.

The skin is incised directly over the rib selected (fig. 183) and the rib exposed. A short cut in the lower margin of the pectoralis major permits this muscle to be retracted medially. The latissimus dorsi is easily freed and retracted laterally. The long thoracic nerve can usually be saved. Incision of the periosteum along the midline of the rib is continued posteriorly between the fibers of the serratus magnus (fig. 184). The periosteum is separated with care to avoid injuring the intercostal vessels and nerve. After rib resection the posteriol periosteum and parietal pleura are incised along the midline, the incision extending to within a half inch of the rib ends (fig. 187). The release of tension by this incision permits the intercostal muscles to retract and allows the ribs next above and below the incision to be rotated upon their axes and increase the exposure. A wider opening can be obtained by increasing this rotation by outward traction. Exposure thus obtained is quite as good as that provided by a mechanical rib spreader (fig. 186, a.), which crushes the intercostal muscles and thus interferes with subsequent healing. Lockwood 23 believes that the use of rib spreaders adds to shock. A headlight makes it possible to continue most of the intrathoracic work under visual control.


FIG. 184.- Method of exposing a rib for resection. The serratus is excised parallel to its fibers; the long thoracic nerve is retracted posteriorly

FIG. 185.- A, Simple type of rib shears; B,.bone biting forceps or chest surgery. The severed bone fragments are held in the bite of he forceps, and can not be sucked into the pleural cavity.

FIG. 186.- A, Tuffier's rib spreader; B, thin-bladed clamp used to secure hemnostasis, and as a tractor


It must be borne in mind that this operation is undertaken in addition to excision and suture of the parietal wound or wounds. Its advantages are obvious. The disadvantages are the longer time required and the creation of another wound in the parietal pleura opposite to normal visceral pleura, which assures the formation of dense adhesions in an area where otherwise only slight adhesions would develop (fig. 182).


Removal of the uncoagulated blood and exudate is effected without delay as clotting occurs very rapidly after operation is begun. A curved metal or glass nozzle (fig. 187) fitted to a Potain aspirator, which is long enough to reach

FIG. 187.- Thoracotomy of election. A rib has been resected; the tip of an aspirating nozzle is being introduced into the pleural cavity

all parts of the chest, is satisfactory if the distal end is flanged and protected by a coarse-meshed sieve to prevent the aspiration of large clots which would plug the tube and to prevent injury of the pleura.

The first aspiration is done with the lung deflated in order to facilitate introducing the tip of the nozzle into the various depressions where fluid may accumulate. During the period of deflation it is wise to remove clots which may be covering injuries or foreign bodies and may be less accessible later. Dense clots can be lifted out with forceps or manually. The balance must be mopped out with moist gauze. However gently this is done, an undesirable irritation is produced.

An important part of cleansing is the prevention of soiling. Skin edges should be covered with moist gauze. As soon as the pleura is opened the


parietal wound should lce covered with more gauze, which is held in place by the retractors and by clips. The field of operation should be isolated by gauze pads even more carefully than in abdominal surgery.

Thoracotomy, however simple, is followed by the production of excess serous exudate. Every effort should be directed to the limitation of this effusion and to protect it from infection. The chest should be made clean and dry before closure, and no irritants, such as bone fragments, clots, bare ribs, or exposed muscles, should be overlooked. Washing out the pleural cavity is a temptation to be resisted as the subsequent healing is poor. Mopping gently with moist gauze gives the best results.


Natural methods of defense have proved that immobility of the diaphragm provides conditions essential to pleural resistance. These consist in providing an adequate reduction of motion and in assuring the optimum circulation of blood in the lung. Cloetta's36 experiments show that the least resistance to pulmonary circulation is present when the maximum negative intrapleural pressure is reduced to 2 to 3 mm. of mercury, and that at this pressure the lung contains the largest volume of blood. Middleton18 found that in dogs an immobilized diaphragm provides this reduction. Experimental and clinical observations prove that recovery from thoracotomy when the phrenic nerve is blocked is more rapid, more certain and less painful than when it is not blocked. Operation is less difficult in the presence of an immobile diaphragm, particularly because acute pneumothorax causes increased re-

FIG. 188.- A, Cow horn rib stripper; B, Tuffier's lung forceps; C, periosteal elevator

One percent cocaine injected in the trunk of the phrenic nerve causes a paralysis for four or five days. It materially reduces distress by blocking the sensory fibers described by Capps.37 Novocaine, quinine and urea hydro- chloride, and magnesium sulphate are unsatisfactory for this purpose.

When the phrenic trunk can be reached through the thoracotomy exposure, it should be injected before intrathoracic operation is attempted. If it is suspected before operation that intrathoracie injection will be impossible the nerve may be injected in the neck. Cervical exposure of the nerve can usually be


FIG. 189.- Incision for exposure of the phrenic nerve

FIG. 190.- Exposure of the phrenic nerve


accomplished with little distress under local anesthesia. The illustrations (figs.189 and 190) show a larger exposure than required. It may be done largely by blunt dissection with little hemorrhage. Relief of much of the intrathoracic pain following injection of the nerve suggests that blocking of the phrenic nerve should be considered an important step in postoperative treatment, if it has been neglected at operation and convalescence is delayed.


The indications for blood transfusion are commonly those of acute anemia, so that intravenous administrations of hypertonic solutions of glucose, with and without gum acacia, tincture of digitalis and insulin are to be considered if suitable donors are unavailable and when patients are suffering from the exemia of shock or need additional protection to their circulatory units subsequent to operation. Transfusion should be started as soon as hemostasis is complete and inflation can be maintained. The blood must be warm and be introduced slowly. No harm comes from transfusion given in this way. Many lives can be saved.

The large amounts of blood and colored fluid removed when the hemothorax is extensive suggest the possibility of aspirating it into sterile bottles containing sodium citrate, filtering, and giving it as an autotransfusion. Tests, however, showed that the attendant risks are prohibitive.


Operations upon the lung are undertaken primarily to protect the pleura. Accordingly accurate hemostasis and surgical repair must be obtained in such a fashion that immediate inflation and resumption of respiration may be established in order to insure reapposition of pleural surfaces and an unimpaired blood supply. These conditions require freedom from internal and external compression. In general, lung tissue which can not be inflated or which is devoid of blood supply is actually or potentially a foreign body whose presence is virtually certain to produce or to intensify an empyema. Excision of such lung, whether or not it has been directly injured, must be considered the ideal treatment, and is demanded whenever the location and extent of the affected area indicate the probability that the consequent empyema will be fatal. Our personal experience is the basis for this radical statement. No individual recovered so far as is known where there was failure to excise permanently injured lung. Not all individuals recovered where this excision was done. Post-mortem observations indicated that the excised lung had contributed to the fatality, but not that excision per se could be blamed for the death.

Perforating wounds caused by rifle bullets and small shell fragments require closure of the pleural defects only when there is bleeding or escape of air after reinflation. Thev can be inverted with a purse-string suture, and should be so inverted if they are accessible. Penetrating wounds are similarly treated. The missile is usually found without difficulty and frequently lies close to the surface. A small pleural incision and a little blunt dissection with


a pair of forceps makes removal easy. Small shell fragments close to big vessels should be carefully removed to obviate the dangers of late hemorrhage.

Large lung wounds are usually caused by tangential violence and are surrounded by more hemorrhagic infiltration than the smaller wounds resulting from more direct impact. The amount of excision required is determined by the degree of internal compression (infiltration). If the lung reinflates completely, trimming the shaggy margins of the defect with scissors and wiping them off with gauze is sufficient. If there is a zone of splenization about the wound which does not inflate or bleed freely, resection must be carried back to margins which are comparatively normal. It is not merely the repair of lung parenchyma which is to be considered. The integrity of the corresponding visceral pleura is of far greater immediate significance and is assured only if the circulation and, therefore, the function of the subjacent lung is restored.

Lung tissue which is not obviously injured and which is nevertheless dangerous because of impaired circulation is characterized by limited inflation, collapse or compression, although intrapleural negative pressures are at least normal, or in spite of positive intratracheal pressures. Abnormal deflation, in the absence of reduced pulmonary elasticity, is due to the failure of either air or blood to enter the involved portion.

Massive contralateral collapse of the lung, noted occasionally with unilateral injury or disease, is attributable to spasm of bronchiolar musculature producing unilateral atelectasis or to a shunting of blood from the pulmonary artery of the sound side to that of the affected side, thus temporarily incoordinating the air cell capillary mechanism. The latter explanation, the shunting of the pulmonary arterial blood, is the more probable and can be traced to unusual responses to distortions in intrapleural pressures. Our experiments show that fluttering of the mediastinum, noted occasionally when thoracotomy is performed without differential pressure anesthesia, is due to sudden and rapidly repeated shunting of blood back and forth from one to the other lung. Neither of these complications occurs if positive pressure analgesia is used.

An atelectatic area of lung is airless because ingress of air is prevented and air previously present has been absorbed. Atelectatic lung after reinflation would have a normal circulation. Occasionally, and particularly after crushing parietal injuries, atelectasis is complicated by destruction of the blood supply and is in effect a traumatic anemic infarction, called carnification. Intratracheal positive pressure anesthesia makes a positive diagnosis of these lesions. Atelectatic lung, noninflatable, is overdangerous.

The more severe intrapulmonary lesions are well shown in a drawing (Plate V) of a right lung removed from a man who arrived dead at a field hospital. A jagged tangential wound was present in the inferior margin of the lower lobe. There was splenization of nearly the whole lobe. The middle lobe, except for one irregular area of hemorrhagic infiltration, and the entire upper lobe were emphysematous. The upper part of the lower lobe was atelectatic, brownish in color and carnified. Plate IV shows the effect of experimental splenization. The marked fibrinous exudate upon the pleura over the area of hemorrhagic infiltration is proof of a defensive reaction. The limitation of




the exudate to the damaged lung is evidence that such lung is especially dangerous in the presence of infection.

Resection of a part of a lobe should be roughly pyramidal or wedge-shaped with the narrow aspect of the wedge toward the hilum, in order to remove only the parenchyma supplied by a bronchus and its associated vessels. It may be compared to cutting a dead branch out of a bush. Positive pressure makes this possible by clearly outlining the limits of normal lung. Resection is difficult only when exposure is unsatisfactory. In the thoracotomy of necessity dislocation of a lung is difficult because these operations are usually done at a lower level and since the lung is generally less pliable because of more severe injury. Under these conditions some traction is unavoidable.

FIG. 191.-Thoracotomy of election. The margin of the deflated lung being guided toward the wound with Tuffier's forceps

It may be secured by a loop of gauze placed around the root of the lobe to be dislocated and caught with a clamp; or a pair of special forceps (fig. 186, b), like an enterostomy clamp, may be placed proximal to the margin to be removed and the resection carried out distal to the clamp. This is a makeshift procedure, but makes fairly accurate work possible.

Thoracotomy of election presents the most satisfactory exposure of the lung. After the chest has been opened as described above and the lung injuries have been determined, the lobes can be delivered into the wound for operation without traction. During deflation a free margin of a lobe is grasped gently with forceps of the Tuffier type (fig. 188, b) and guided toward the parietal opening. As the anesthetist increases the positive pressure this lobe follows the forceps into the wound (fig. 191). Manipulation with suitable variations in pressure allows the selection of the most satisfactory position for operation.


Little traction is required to maintain this position. which is preserved by the requisite inflation and the gauze packed between the lobe afnd the superficial wound.

A wedged-shaped resection is made through the mnargins of normal or relativelv normal lung. Hot compresses to the cut surfaces check excessive bleeding while clamps are applied. Severed bronchi that can be seen or that provoke active bubbling must be secured. All branches of the bronchial artery spurring red blood under high pressure must be clamped. It is necessary to secure only the more actively bleeding branches of the pulmonary artery and but few branches of the pulmonary veins.

Pulmonary inflation makes possible the accurate clamping of these various structures, which would be difficult or impossible to find on the cut surfaces of a lung in deflation. Mass sutures inserted and tied while the lung is inflated would become too loose in deflation; if tied in deflation, they would be too tight during inflation. Consequently, multiple fine sutures must be used to approximate cut surfaces.

FIG. 192.- Methods of reducing the number of stitches used in repair after a resection. At the bottom is a simple ligature; in the middle, a suture on one surface is combined with a ligature on the opposite side; at the top, a double ligature is being made to serve as a suture.

The number of sutures required may be reduced number of stitches used in repair if the vessels which have to be tied are ligated in conjunction with a suture (fig. 192). The pleural repair should become an exaggerated serosa-to-serosa approximation as a safeguard against air leakage. A fine catgut stitch of the Cushing type introduced one-half inch from the margin of the incision accomplished this satisfactorily. Little puckering of the lung need result and only one knot should appear upon the pleural surface (fig. 193). When resection includes two surfaces of a lung it is easier to suture the dependent pleural layer first, when it can be done more rapidly and from the inside. and then to reconstruct the lung upon this plane.

Lung repaired by this method heals splendidly and with the minimum of scar tissue. Duval 24 states that the scar may be impossible of detection microscopically. There is in consequence little influence upon pulmonary elasticity. Vicious union is prevented so that the functional result is nearly perfect..
FIG. 193.- Closure of the visceral pleura with an exaggerated Cushing stitch

 Complete lobectomy is indicated when an entire lobe is carnified The possibilities of recovery are slight, not because of the operation but on account of the other serious injuries which are usually associated with military surgery because, aside from the injury, tissues are normal and individtlals healthy. Crushing with forceps and ligation seem to be satisfactory if an ample flap covering is provided.


The unusual wounds of large bronchi should be sutured directly, and the suture line reinforced by a pleural graft to give additional protection against empyema from leakage.


Early operations upon the heart and pericardium have been few because wounds affecting  them are either promlptly fatal or are treated expectantly. Such injuries as are disclosed by operation are easily remedied as they require little more than simple suturing.

Wounds of the mediastinum are always serious, particularly because of liability to infection. Bleeding from mediastinal vessels, even from the azygos veins, is not easily checked. Foreign bodies should be removed with the least traumatism and the resulting defect carefully repaired. Wounds of the thoracic duct have usually been fatal in spite of ligation.


When excisions are necessary they should be as limited as safety permits and adapted to spare the branches of the phrenic nerve. Usually all serious diaphragmatic injuries are well exposed by a thoracotomy of necessity. In the repair of defects of the diapliragmn and in suturing it to parietal defects care should be taken to avoid including nerve branches in the sutures. Adhesions need not eause persistent immobility, but diaphragmatic immobility due to persistent paralysis commonly assures permanent adhesions.

When abdominothoracic injuries are suspected it is easier to attend to the chest injury first, as this operation is better tolerated and will establish more favorable conditions for laparotoiny. Occasionally the entire abdominal repair can be completed through the diaphragm.


Next to positive pressure analgesia with nitrous oxide and oxygen, accurate closure of the chest wall, and particularly the repair of the parietal pleura, is the most important plase of the entire operation. No matter how excellent the rest, of the care may be a failure to obtain air-tight healing of the parietes results in an open pyothorax. The parietal pleura heals more rapidly than the more superficial tissues and offers greater resistance to the extension of inflammation either from within or without. Failure to secure smooth healing of this layer may lead to a persistent and severe empyema even though the more superficial wound heals firmly.

An illustration of the result of imperfect healing of the parietal pleura is found in Figure 194. The pleural sutures had caused tension necrosis so that, separation occurred with exposure of the muscle and rib endls to the pleural cavity. This accident, occurred in a dog but is exactly comparable to conditions seen repeatedly at human necropsies.

A well-controlled series of clinical and experimental observations show that smooth parietal healing is assured only when there is accurate pleural approximation in the absence of suture tension. A closure of this type is possible when the ribs next above and below the defeet are brought abnormally close together.

The pleural approximation must extend beyond the rib ends at either angle of the incision in order to make certain that no denuded rib be exposed or become exposed within the pleural cavity.

The method of suture has been illustrated diagrammatically to show the steps in the closure of a thoracotomy of election. During the period of closure the anesthetist gradually increases the positive pressure to produce a slight degree of hyperinflation, so that when the parietal pleura is hermetically closed there shall be the least residual pneumothorax. Under these conditions approximately normal negative pressure is immediately reestablished.

Mattress sutures are preferable for reuniting the parietal pleura, and usually can be inserted through both margins of the cut periosteum, so as to gain double support (fig. 195). When this is impracticable the sutures should be placed far enough below the lower margin of the incision to avoid the intercostal vessels. Single mattress sutures are placed at each angle and in the middle of the incision. The spaces between these are closed with continuous mattress stitches. As indicated in the diagram the outer loop of each lateral stitch is inserted behind and lateral to the cut rib end. When the stitch is tied, pleural approximation external to the rib end is secured and the denuded rib excluded from the pleural cavity.

FIG. 194.- Inner aspect of the chest wall, obtained after death from purulent pleurisy without open pyothorax. Suture tension necrosis has permitted separation of the margins of the parietal pleura with exposure above and below results of the muscle and bone

 An attempt to tie these sutures without overapproximation of the ribs next above and below results either in the stitches pulling out immediately or sloughing out later. The ribs are pulled together by means of a wire stay passed about them, as indicated in Figure 196, from which for the sake of clearness, the mattress sutures have been omitted. One loop is passed beneath the rib, and, if possible, kept outside of the pleura by hugging the rib with a needle, the point of which has been made slightly blunt. As this loop of the stitch crosses the pleural incision it takes in the margins of the parietal pleura and intercostal muscles to promote their subsequent apposition. The other loop is passed over the rib and does not include the intercostal structures.

When this wire stay is drawn taut the ribs are so approximated that it becomes possible to appose pleura to pleura without tension by drawing upon the mattress sutures. The three single stitches are tied (fig. 197); the continuous stitches are pulled tight, and tied at either end and in the middle to an end of a single stitch. This closure is air-tight if the suturing is done accurately. If there is a leak, it will be revealed by escaping air and must be corrected by a stitch or two to prevent subsequent production of tissue emphysema.


FIG. 195.- Closure of the chest wall after thoracotomy. Single and continuous mattress sutures are inserted through both periosteal margins on either side. The lateral sutures have the outer loop of the stitch placed beyond the rib ends

FIG. 196.- Closure of the chest wall after thoracotomy. In inserting a wire, rib stay, the posterior loop includes the periosteum, so as to aid in approximating the parietal pleura. The anterior loop does not penetrate the intercostal tissues.


Overapproximation of the ribs likewise secures an abundance of muscle tissue for the next layer. Healing is better if there is no dead space at the rib ends. A single suture, placed near the rib ends, as indicated in Figure 198, is tied and one end is cut. The other end is now tied to the end of the lateral mattress stitch in the parietal pleura. The tissues are thus held snugly over and against the end of the rib. The balance of this layer is closed with interrupted stitches, which promote better healing than continuous stitches.

The margins of the pectoralis and latissimus dorsi muscles are approximated with mattress sutures to safeguard the layer beneath from the dangers of suppurition in the subcutaneous fat (fig. 199). The deep layer of the superficial fascia is united by interrupted stitches (fig. 200), inserted opposite to the usual manner so that when tied the knot is not exposed superficially. The skin is closed with interrupted stitches to permit the escape of serum. Superficial drainage is often desirable to prevent tension. Small, soft drains suffice, and, removed within 24 hours, do not endanger infection of the deeper tissues.

Such exact methods can be criticized as time consuming and unnecessary. In a considerable proportion of all operations these objections are valid. However, there is another proportion, too large to be neglected, which can not always be recognized at the time of operation, in which they are invalid. We observed repeatedly that when details were neglected during periods of extraordinary activities, the subsequent healing was less satisfactory and led to distress and to death which greater care might have obviated. Need to serve greater numbers can and perhaps should justify compromises with the excellent rule that what is worth doing at all is worth doing well.

The wire rib stay is particularly undesirable, but absorbable sutures do not hold and silk or linen are unreliable. The wire should be removed in from four to six weeks as it will eventually cut through the ribs, and for this reason ought to be insorted so as to facilitate subsequent exposure of the twisted ends and removal through a small incision made with local anesthesia.

Methods of closing the chest in limited thoracotomy and in thoracotomy of necessity are based upon the same principles.


The use of primary and even of early drainage of pleural cavities after operation was generally condemned because open methods were employed, and because with them collapse of the lung is inevitable. Gray 38 cited figures to show that closure was safer than open drainage. The postoperative accumulation of serum was recognized as a constant occurrence, and routine aspiration. repeated as required, was adopted as a necessary precaution against its dangers. Lockwood and Nixon 23 found that interrupted drainage by aspiration reduced the need for subsequent rib resection and open drainage.

Experience in the early and late treatment of pleurisy shows that air-tight insertion of drainage tubes provided with some device to prevent the entrance of air is the most satisfactory procedure, and that the earlier such drainage is established the better the results. It is evident from the writings of Elliott and Henry,16 Dobson,39 Mozingo,40 Blankenhorn,41 Whittemore,42 McKenna,43


FIG. 197.- Closureof the chest wall after thoracotomy. The rib stay is drawn taut and fixd so that it can easily be removed

FIG. 198.- Closure of the chest wall after thoracotomy. The layer of fat and muscle is closed over the defect caused by rib resection

FIG. 199.- Closure of the chest wall after thoracotomy. The margins of the pectoralis major and lattisimus dorsi are united with mattress sutures to protect the deep wound

FIG. 200.- Closure of the chest wall after thorarotomy. The deep layer of the superficial fascia is closed with buried sutures


and Roberts 44 and from our own experiences, that a catheter drain inserted intercostally by trocar and cannula, and provided with an automatic one-way check valve, is the method of choice. (Fig. 201.) By this means negative pressure can be maintained until full inflation is secured. Employment of fluctuations in intrathoracic pressure to provide the expulsive force is better than using suction, as it is in keeping with normal physiologic movements. It is also very much simpler, and, being automatic, is not dependent upon constant attention for its success.

Duval 24 advised against operating more than 30 hours after injury. The dangers of empyema after that time are very great but are less with operation than without, especially if primary drainage is used. We found that early operation, conducted by methods described above, gave a mortality rate of 4 percent in an unselected consecutive series of thoracotomies performed within eight hours and the men had not been chilled. When the operation was delayed beyond 24 hours or the wounded were cold or wet, the mortality rate was increased tenfold. It seems now that had primary drainage been used more as a routine many of these lives could have been saved. It is certain that primary drainage after operation for the large wounds with open thorax, foreign bodies, and lacerations of liver and diaphragm was effective in limiting disastrous pleurisy.

FIG. 201.- A, Trocar, cannula, and catheter for intercostal drainage: B, flap valve used to secure automatic oneway drainage


The use of positive pressure anesthesia is the most important factor in simplifying the postoperative care, because of the decreased distress due to the abolition of pneumothorax, the better general condition of the patient, and the elimination of complications.

Morphine in generous doses is useful in reducing pain, in limiting the rate of respiration, and in controlling coughing. The value of posture has been over-rated. Patients naturally take the most comfortable position. Other things being equal lying upon the sound side is preferable. The chief interest is in the postoperative effusions. If primary drainage has been instituted there is only need to see that the tube is free. If the chest has been closed tight one of two courses is open: Either aspiration should be done the second day and repeated often enough to prevent any considerable reaccumulation, or it should be done when signs of pressure or of infection develop, or when absorption is unduly delayed. There are objections to all methods, least of all to primary drainage.

The most important feature, next to safeguarding life, is the earliest reestablishment of full pulmonary functions. No breathing exercises should


begin until the patient is afebrile, and then if a rise in temperature should follow they must cease until the temperature is again normal. Holding the breath, slow breathing, and blowing against pressure are valuable, with particular attention given to abdominal breathing in order to reactivate the diaphragm. Patients should be encouraged to get out of bed as early as safety permits and to exercise the muscles of the thorax. Recovery is protracted as early activity is delayed. Active duty within six weeks of an injury serious enough to require resection of part of a lower lobe was achieved by one soldier, and this indicates the possibilities of treatment.


Pleurisy is to be regarded as a natural and constant sequel of an intrathoracic injury, including thoracotomy. Empyema is the commonest and most significant complication. There is no sharp line separating the two and none should be assumed. The best treatment of empyema is to limit the initial pleurisy, and this is more certainly accomplished by primary drainage than by any other method. When primary drainage or routine aspiration is impossible exploratory puncture is indicated by signs of sepsis or by signs of a sudden increase in the exudate. The probability of subsequent development of an empyema can be established by the soiling and reaction of the pleura present at the time of operation

FIG. 202.- Form of trap to be attached to a catheter drain. It provides both a receptacle for the discharge and a flap valve. If slight suction is needed, it can be provided by collapsing the trap

and and by the length of time elapsing between injury and operation.   As a rule, the more virulent the infection the more rapid the development of postoperative empyema and the less the likelihood of limiting adhesions being formed.

Open operation for empyema performed without differential pressure is dangerousif done before adhesions form, and may be futile if delayed. Catheter drainage should be the first resort and will usually be sufficient. If more radical intervention is required thoracotomy with positive pressure analgesia will permit of all necessary manipulations without the dangers of pulmonary deflation, and with the least danger of disseminating infection. The wound may be closed completely, and tube or catheter drainage established elsewhere, or a tube may be sewed in hermetically at an angle of the incision. It is understood that some form of valve is provided so that subsequent irrigations may be given without permitting air to enter the chest. Slight constant suction will often prevent coagulation of serum within a catheter. This suction and a receptacle to catch the discharge may be provided by a simple apparatus (fig. 202). Fluoroscopic control makes it easy to insert the catheter accurately and safely.


Pneumonia is usually contralateral and of the bronchial type. It is rare, as is contralateral pleurisy, when gas-oxygen pressure analgesia is used. The probable explanation is the less urgent dyspnea and consequent failure of aspiration of blood and mucus which are expelled from the bronchi of the injured side into the bronchi of the sound side.

Contralateral collapse, which has been so well described by Bradford,19 does not occur when positive pressure is used.

The presence of a foreign body may prove a serious handicap to permanent recovery. It has been said that every man harboring one foreign body suffers from two, one physical and one mental. Small shell fragments and bullets embedded in the lung may cause pain and dyspnea because they are sources of chronic traumatism, and because their removal is occasionally curative. The distress is more often due to the effects of pleurisy, restricted motion of the parietes, reduced intrapleural negative pressure, or to impaired motion of the diaphragm. Under such conditions the foreign body is an incident and its removal can change the mental attitude only. The indications for removing these intrapulmonary foreign bodies are a matter of opinion.

Three methods are available for the late removal of intrapulmonary foreign bodies: First, the use of forceps of the alligator type introduced intercostally and manipulated under fluoroscopic control. Petit de la Villeon45 introduced this procedure and reported remarkable results. Its dangers, particularly when the foreign body is at the root of the lung, are obvious. It is a blind procedure. A second method, also a compromise, consists in a short costectomy, suture of the parietal to the visceral pleura, and removal of the foreign body with forceps guided by a Hirtz compass. Like the method first described, this method is useful in the extraction of small superficial fragments which are of the least significance. When the foreign bodies are deeply placed, near the large vessels or mediastinum, an open operation should be done, the foreign body removed and the incidental damage accurately repaired. The best treatnment wolldI have been removal of the missile at the time of injury.

Foreign bodies in the pericardiumn, in the heart muscle, and in the cavities of the heart have been removed. Aside from the possibilities of civil surgery suggested by these highly dramatic operations a very practical hint has been given toward selecting the method of approach. Most operators have attempted to keep out of the pleural cavity and have generally succeeded in provoking an empyema. An open operation, such as a thoracotomni of election, if done with differential pressure, gives a very satisfactory cardiac exposure, and would make possible better surgery and smoother recoveries.

Late, persistent, or recurrent empyema suggests the possibility of some chronic irritant like a foreigni body or a costal osteomyelitis, both of which must be excluded with the least possible delay. Much has been and will be written about methods of treating open pyothorax, and various methods of irrigation and decortication. All have one object, to convert an open into a closed thorax by pulmonary inflation. The essential factor is the reapposition of visceral and parietal pleura, whether attained operatively as proposed by Roux-Berger46 or by breathing exercises. The general plan of procedure is


based upon the elimination of infection and reduction of irritation: solutions which favor disintegration of fibrin are indicated. To date nothing superior to Dakin's solution has been found.

Treatment of closed pvothorax is being advanced by the application of principles established by Bowditch.7 Instead of an immediate costectomy and the formation of an open pyothorax it is far better to introduce one or two catheters by means of a trocar and cannula. Intermittent irrigation and continuous drainage, which protect and increase pulmonary inflation, may lead to prompt recovery. If this method be partially unsuccessful, a less forbidding operation is ultimately required. Encapsulated pockets may be tapped safely if the manipulations are done with the aid of a fluoroscope. The best treatment of empyema is that proposed by Roux-Berger,46 early and complete operation after injury.

Reports by Rist, Flandrin, Bernard, Sommerville, and others,47 Pehu and Daguet,48 all show that chest wounds are unlikely to favor early development of tuberculosis. Tuffier30 noted that the disease is prone to appear on the opposite side. The reasons are clear. The injured side is apt to be incompletely expanded and is in a condition, according to Cloetta,36 to offer hypernormal resistance. The opposite side, constantly affected by compensatory emphysema, is in a state of reduced resistance.

Physical disability will persist in the wounded in proportion to the interference with the respiration and circulation. The interference may be so slight, as merely to reduce the powers of compensation and become apparent only under stress, or it may cause dyspnea, palpitation and cyanosis with slight exertion. If there is incapacity, attempts should be made to determine the cause, and such treatment, operative, medical, or physical, be employed as will improve function. It is easy to make a diagnosis of neurosis or malingering in individuals who have sustained chest injuries, and do them a grave injustice. Complete physical examination before and after exertion, controlled with it fluoroscope, and proper blood pressure determinations, will generally disclose reasons for the complaints of pain and disability.


An analysis of the histories of patients has been made to disclose the causes of deaths and of the extent and duration of disabilities in those who survived in order to determine if possible all the factors which might influence more effective care of thoracic injuries.

The relative merits of divers methods may be judged from two aspects- immediate military necessities imposed by fighting conditions, and the welfare of the individual which must always be second to winning wars. However, in a protracted struggle success is determined by the morale and strength of an army, both of which depend to a great extent upon each soldier being assured the utmost consideration before and after he is hurt. Hence, there is no reason to consider other procedures than those offering individuals the largest opportunities for undelayed recoveries.

There are several sources of error in the data presented from which deductions must be made. Our records are incomplete; observers may have been biased; and there is a natural disposition on the part of the disabled to magnify


their misfortunes, particularly when financial compensation varies with the extent of disability. On the contrary, the details that can be established suffice to indicate the facts essential to drawing conclusions. The causes of death are usually easy to recognize. The causes of disabilities are unmistakable, viz., reduced motion of the parietes, particularly of the diaphragm; diminished elasticity of lung; abnormal intrapleural pressures, and myocardial deficiencies. Not only are these lesions demonstrable, but vital capacity also may be used as a check upon the extent of their composite detriments because it measures the extent of disabilities attributable to cardiorespiratory diseases.


Attempts were made to keep notes upon each wounded man, to take photographs that would show the nature of wounds and postoperative conditions, to make routine post-mortem examinations, to trace survivors after they were evacuated from forward hospitals, and finally to ascertain the nature of their ultimate disabilities.

Some patients were treated and no notes were kept; some notes were lost. But it is improbable that these breaks in our records (not above 25 percent) are of material significance, since enough information is available to give a dependable average. Unfortunately the photographic and X-ray films were stolen from the central laboratory at Dijon. They would have shown details that can not now be described. It was impossible to trace the wounded in France, so there was no control of their later and most important postoperative treatment. It has since developed that except for necessary surgical care they got none. Although unfortunate for the individuals, this neglect is of great value in demonstrating the limitations of operative procedures alone as means to restoration of function.

Through the intercessions of the Surgeon General of the Army and the generous cooperation of the United States Public Health Service a member of the unit was enabled in 1921 to visit a number of patients in Government, hospitals who had had chest wounds and to see certain others living near by who came in voluntarily for examination. This officer secured sufficient accurate information to determine the extent, nature, and causes of disabilities in those whom he examined and to make it possible to estimate dependably the conditions of others not examined but whose records could be found. Here again, although our records are incomplete, it is probable that enough has been learned to give a reliable average of results, since the more seriously incapacitated were hospitalized, and they form a larger part of the observations.

The series here recorded consists of 104 patients whose records are available. They have been divided into four groups, according to the nature of the first operation performed: (1) Excision of parietal wound; (2) limited thoracotoiny; (3) thoracotomy of necessity; and (4) thoracotomy of election.

The history of each patient is outlined, the cause of death presented, or the nature and causes of disabilities stated. The treatment of each patient is criticized. Similarly, the treatment of each group is criticized. Finally, reasons are given to show how the entire series could have received better care from the time injuries were inflicted until the final results were established.



1. W. M. July 22, 1918: Bullet, perforating wound, left chest. Entrance below and mesial to angle of right scapula; thence through tissues of back, through left chest, to lodge in left side of neck. Duration, unknown. Hemopneumothorax moderate. Condition good (?). Operation: Excision of parietal wound. Removal of foreign body. Pleural wound not exposed. Closure without drainage. July 24, 1918: Fluid not increased; partial pneumothorax persists. Evacuated in good condition. Returned to duty in three months.1921: Records available. Disability less than 10 percent and due to chronic pleuritis.

Recovery would have been more certain, rapid, and complete had hemothorax been removed and lung sutured at time of operation.
W. B. July 27, 1918: Bullet, through-and-through, sucking wound, left chest. Entrance, posterior triangle above left clavicle; exit, in fourth interspace, lateral to midclavicular line. Hemopneumothorax; heart slightly displaced to right. Condition good (?). Duration, unknown. Operation: No rib injuries recognized. Wounds excised and closed without drainage. Hemothorax not aspirated. Pressure anesthesia disclosed no air leakage. August 1, 1918: Uneventful recovery. Healing smooth. Left lung slightly cloudy. Left diaphragm excursions reduced one-half. Bight chest clear. Estimated disability less than 10 percent and due to pleuritis.

Hemothorax should have been aspirated at time of operation and again later if persistent. Absence of rib injuries and of air leakage contraindicated thoracotomy.
 3. D. C. July 27, 1918: Bullet, penetrating wound, left chest. Entrance, upper anterior aspect of chest; lodgment, just to left of aortic arch; moves with respiration, probably intrapulmonary. Moderate hemopneumothorax; heart displaced to right. Duration, six hours. Condition good (?). Operation: Entrance wound excised. Pleural wound not found. No rib injuries. No air leakage recognized. Closure with one gutta-percha drain. Hemothorax not aspirated. July 31, 1918: Smooth convalescence. Both sides of chest cloudy; left more so. Left diaphragm moves, though but slightly. Foreign body at root of left lung. Tissue emphysema. Early removal of drain showed that drainage of superficial tissues can be helpful and harmless. August 2, 1918: Evacuated in good condition. Estimated disability less than 10 percent and due to pleuritis.
Perforation in pleura should have been located and accurately closed with muscle to prevent emphysema. Hemothorax should have been aspirated. Removal of foreign body not indicated.

4. E. L. D. July 28, 1918: Shell fragment, penetrating wound, left chest. Entrance through left scapula; lodgment in left lung, where foreign body 1 cm. in diameter can be seen to move with respiration. Hemopneumothorax moderate. Duration, 24 hours. Condition, poor. Operation: Gutter wound, left arm, excised. Wound through left scapula excise and drained, as patient's condition did not permit of thoracotomy. Gum acacia given before and during operation. August 3, 1918: Smooth recovery. Pleuritic exudate still present and heart displaced to right. Evacuated in good condition. Disability, estimated at less than 10 percent, as this man returned to active duty, can be attributed to pleuritis.
No radical operation was possible under the conditions. Aspiration of hemothorax should have been done.      
5. R. E. H. July 30, 1918: Shell fragment, perforating wound, left chest. Entrance second left interspace, midelavicular line; lodgment beneath skin of back at eighth interspace. Massive hemothorax; heart much displaced to right. Duration, 30 hours. Patient treated  6¾ hours because of shock. Gum salt intravenously raised systolic blood pressure 30 mm. and made operation feasible. Operation: Excision of both entrance and exit wounds with removal of foreign body. No rib damage. Positive-pressure analgesia forced out part of hemothorax, which was not aspirated. No air leakage from lung recognized.


Pleural openings closed with muscle. Wounds closed without drainage. August 3, 1918: Convalescence and healing smooth. Pleuritic exudate subsiding, but heart still displaced to right. Evacuated in good condition. July 20, 1921: Passed through several hospitals in France. No further treatment. Discharged from service March 7, 1919. Disability, 10 percent. Dyspnea on extraordinlary exertion. Utnable to lift heavy weight. Limited expansion at left base. X-ray shows chronic pleuritis; adhesions between left diaphragm and ninth rib and little mobility of diaphragm. Heart competent. Vital capacity, 71 percent.
Disability due to pleuritis. Rating at 10 percent is seemingly too low in view of vital capacity. Conditions at time of operation prevented more radical treatment. Aspirations should have been performed; possibly one-way catheter drainage. Systematic breathing exercises would have reduced disability to less than 10 percent. Limited thoracotomy might have been employed.

6. J. F. M. July 31, 1918: Shell fragment, penetrating wound, right chest. Entrance above and mesial to spine of scapula. Large hemothorax; heart displaced to left; foreign body one cm. in diameter in lung hilum close to aorta. Duration and condition not noted. Operation: Entrance wound excised. No rib injuries. Hole in pleura plugged with muscle. Superficial wound drained with gutta percha. August 1, 1918: Toxic and uncomfortable. Pleural effusion increased; heart displaced to left anterior axillary line; 500 c. c. bloody fluid aspirated. Immediate relief. August 3, 1918: Condition good. Fluid not reaccumulating. Evacuated. July 30, 1921: A second aspiration was subsequently performed at a base hospital in France. Returned to United States September 15, 1918. No further treatment. Disability rating 20 percent. Dyspnea on slight exertion. Chronic eough. Fibrous pleuritis right side, marked above. Diaphragm deformed and movements restricted. Peribronchial thickening right side. Foreign body in lung in front and lateral to aorta. Myocardium slightly defective. Vital capacity 63 percent.

Disability due to pleuritis, pulmonary sclerosis, and myocardial deficiency. Unable to follow active occupation. Disability rating of 20 percent is apparently low. More radical operation and removal of foreign body had man's condition permitted would have been wiser. Man states he had suffered from dysentery for two days precedIing injury so his condition may have been unfavorable. Immediate aspiration, perhaps catheter drainage, should have been used. Subsequent breathing exercises could have reduced disability.

 7. F. J. S. August 10, 1918: Shell fragment, through-and-through wound, right chest. Entrance to left of spine at level of angle of scapula, thence through vertebre and ribs up- ward and to right to exit through posterior triangle of neck, fracturing first rib. Duration 21 hours. Condition critical. Operation: Entrance and exit wounds excised. Pleural defects plugged with muscle and superficial wounds sutured; 500 c. c. of gum salt given intravenously during operation raised pressures from 98/60 to 108/64. Soon after operation pressures had fallen to 60/30. Transfused 500 c. c. citrated blood. Pressures raised to 100/60, temporarily. Death within a few hours. Necropsy: Kissing wound at apex of right lung. Splenization right upper lobe. Moderate hemothorax. Dilatation of right heart.

This man could have recovered had it been possible to operate when hewas in good condition. Removal of right upper lobe would have been necessary. An illustration of the need to protect the pulmonary circulation.

8. N. E. L. September 12, 1918: Shell fragment, penetrating, right chest. Entrance wound at tip of right clavicle. Moderate hemothorax. Foreign body 1.8 cm. in diameter within lung. Cough and hemoptysis marked. Duration five hours. Condition good. Operation: Entrance wound and tract excised. No rib damage. Entrance intopleura not found. Superficial drainage. No aspiration. September 14, 1918: Healing smooth. Pleuritic effusion reducing. Heart no longer displaced to left. Evacuated in good condition. 1921. Records available. Disability less than 10 percent.


Immediate aspiration should have been performed.

 9. W. J. McC. September 12, 1918: Shell fragment, penetrating wound, right chest wall. Entrance in third interspace at right sternal margin to lodge beneath skin over fifth rib in anterior axillary line. Small hemothorax; heart not displaced. Some hemoptysis. Duration unknown. Condition good. Operation: Excision of wound of entrance; resection of upper right margin of sternum and of fractured third rib. Pectoralis major muscle split for exposure. Foreign body 2 cm. in diameter removed. Pleura not opened. No suture closure. Gutta-percha drain. September 17, 1918: One aspiration removed straw-colored serofibrinous fluid. X ray shows some cloudiness at right base. Diaphragm and its excursions normal. December 24, 1918: Returned to active duty. August 3, 1921: Little disability. No cough. Rating 10 percent. Heart normal. In lungs there is slight increase in peribronchial strive; otherwise normal, as is the diaphragm. Basal chronic pleuritis, right. Vital capacity, 90 percent.

Disability is due to pleuritis which could have been eliminated with aspiration at time of operation and subsequent breathing exercises. This is an example of pulmonary injury caused probably by indirect violence. In this instance thoracotomy was not indicated as it would have been had there been splenization of the lung.

10. F. K. September 12, 1918: Bullet, through-and-through wound, left chest wall. Entrance at anterior axillary line, level of fourth rib; exit at posterior axillary fold at level of sixth rib. Small hemothorax; dyspnea. Duration seven hours. Condition good. Operation: Wounds of entrance and exit excised; connecting tract laid open and excised. No rib injury nor pleural laceration. Superficial wound closed with drainage. September 18, 1918: Pleuritic effusion persists in small amount, but is decreasing. Evacuated in good condition. No further records. Estimated disability less than 10 percent.

Aspiration, perhaps repeated, should have been employed.

11. P. O. September 26, 1918: Bullet, through-and-through wound, right chest wall. Entrance just below right sternoclavicular joint; exit below and external to right nipple. Small hemothorax. No dense shadows. Duration five hours. Condition fair. Operation: Ten and a half hours later. Wounds of entrance and exit resected; connecting tract opened. No rib or pleural damage. Lacerated outer one-half of pectoral muscle removed. Partial closure. Superficial drain. September 30, 1918: Condition excellent. No increase in pleuritic effusion. Evacuated. No further records. Estimated thoracic disability zero.

Aspiration, especially if performed under fluoroscopic control, would have been wiser.

12. E. L. October 1, 1918: Bullet, through-and-through wound, neck and right chest. Entrance above inner third clavicle; exit, right axilla, level of fifth rib. Cold, exhausted; pulse feeble and rapid. Extreme interstitial emphysema of face, neck, arms, and thorax. Grunting dyspnea. Exhaustion due to long, hard ride. Duration 12 hours plus (?). Operation: Excision of entrance wound. Air escaped from tissues in bubbles. Wound packed with gauze and a tube drain inserted. Condition prevented further intervention. Death 14 hours later. Necropsy: Left pleural cavity, negative pressures persist. Right pleural cavity, positive pressure; lung compressed and diaphragm depressed by large hemopneumothorax. Through-and-through wounds of trachea in episternal notch and of left upper lobe. Peritracheal emphysema; trachea not obstructed. Death due to obstruction of venous return from head.

This man arrived too exhausted to stand operation. If he had been received earlier tracheal repair or tracheotomy would have been effective.

13. R. M. October 13, 1918: Shell fragment, through-and-through wound, left chest wall. Entrance beneath anterior axillary fold; exit below angle of scapula. Much bone


damage. Small hemothorax. Condition good. Duration 22 hours. Operation: Entrance and exit wounds excised. No pleural injury. Wide resection of scapula. October 19, 1918: Evacuated in excellent condition. No further records. Estimated thoracic disability zero.

Even this limited hemothorax should have been aspirated.

14. W. J. W. October 14, 1918: Shell fragment, penetrating, right chest wall. Entrance wound right axilla, level of fifth rib. Foreign body, 0.3 by 0.7 cm., beneath right nipple. Operation: Entire wound excised. Foreign body removed. No rib or pleural injury. Partial closure. October 16, 1918: Wound clean. No abnormal chest findings. Excellent condition. Evacuated. No further records. Estimated thoracic disability zero.
15. G. S. October 14, 1918: Shell fragment, penetrating wound, left chest. Entrance just below spine, left scapula. Moderate hemothorax. Foreign body 0.6 by 0.8 cm. in lung. Intrapulmonary hemorrhage. Duration, nine hours. Condition, fair. Operation: Two and a half hours later. Wound of entrance excised. Fractured scapula resected. Pleural defect closed with muscle. Partial closure. October 15, 1918: Comfortable. Stations changed. No further records.

If this man's condition had permitted, a thoracotomy of election should have been performed, as it would have caused less risk than a splenized lung.

16. B. W. November 1, 1918: Shell fragments, 0.5 by 0.8 cm., and 0.5 by 0.12 cm., penetrating wounds of back and right chest. Sucking wound at entrance at ninth rib below angle of right scapula. Small hemopneumothorax. Heart displaced to left. Complete section of cord. Duration, 24 hours. Condition, poor. Operation: Wound of entrance excised. Contused muscle, broken ribs, and vertebrae resected. Closure with muscle flaps. Superficial drainage. Died within six hours. No necropsy.

Lethal injury. Treatment with morphine would have been wiser.

17. H. G. H. November 2, 1918: Shell fragments, penetrating; seven wounds of back and right chest. Moderate hemothorax; hemoptysis. Duration, six hours. Condition, poor. Operation: One and a half hours after admission. Multiple wounds excised and superficial foreign bodies removed. Fractured spinous processes resected. Wound in pleura closed. November 18, 1918: Condition, fair. Fluid in chest despite previous aspiration. No pneumothorax. Febrile. Died some days after evacuation. No further notes.

This man's life might have been saved by proper drainage, as he died presumably from emphysema.

18. H. C. H. November 12, 1918: Bullet through-and-through wound, right chest. Entrance over third rib, right parasternal line; exit, right posterior axillary line at level of fifth rib. No hemoptysis. Moderate hemopneumothorax and cardiac displacement. Some tissue emphysema. Duration, 34 hours. Condition, fair. Operation: Exit wound excised. Slight sucking. No rib injury. Closure with superficial gutta percha drain. Entrance wound excised. No rib injury. Closed. Chest aspirated. November 16, 1918: Temperature normal. No cardiac displacement. Small pleuritic effusion. Condition, excel- lent. Evacuated. February 5, 1919: Discharged from service. No further treatment. August 12, 1921: Dyspnca on slight exertion. No cough. No limitation in respiratory excursions. Slight fibrosis at right apex. Diaphragm free. Heart: Rapid action; otherwise normal. Vital capacity, 110 per cent. Disability rating, 25 percent.

Disability granted is due to cardiac excitability and not to previous injury. This man's treatment, in view of the duration of injury before operation, has been justified by the results. An intercostal catheter drain might have done harm and accomplished no more good.


The number (18) treated by extrapleural excision is 20 percent of the entire series. It would have been much larger had not so many of the less seriously wounded been sent to hospitals farther away.
Fatalities.-There were four deaths (22 percent). Two (7, 12), lethally affected at time of operation, could have recovered if operated upon earlier. One (16) was fatally injured because of a cord lesion; one (17), despite multiple injuries, might have recovered had the principles of primary drainage been under-stood and suitable apparatus been available. One (17) was operated six hours after injury; the others after 24 hours. No death can be attributed to the operation itself. One is chargeable to surgical error, the failure to provide drainage. This gives a surgical mortality of 5 percent.
Disabilities.-The ultimate disabilities of 13 of the 14 who recovered are quoted from allowances made by the War Risk Insurance board or by estimates made in comparison therewith. Three (11, 13, 14), recovered without disability from thoracic lesions. Five (1, 2, 3, 4, 10) had disabilities of less than 10 percent. Two (5, 9) had ratings of 10 percent. One (6) was rated at 20 percent and one (18) at 25 percent. The ratings of two (5, 6) are low and the rating of one (18) is high. The result in one (15) can not be surmised. A generous estimate of the average disabilities for this group is 10 percent. The average interval before return to duty was about 90 days. Pleuritis, arising from hemothorax, caused the disabilities in all but one (18, disordered action of the heart). In four, hemothorax was caused by transmitted violence as the parietal pleura had not been punctured. Three of these (11, 13, 14) recovered without disability, and in the fourth (10) it was less than 10 percent. On the average 24 hours had elapsed between receipt of injury and operation.

Had it been possible to have hospitals close to the zones of conflict or to have had effective sifting of the wounded and expeditious transportation, the mortality rate would have been less. Had the proper use of intercostal catheter drains been understood or had the members of this unit been permitted to give patients individual postoperative attention, particularly in earlier and perhaps repeated aspirations, the disability rate (10 percent) would have been reduced. Also had continued after-care, especially breathing exercises, been a routine, the average duration of disabilities (90 days) would have been less.


Virtually all wounds of the chest should be treated by prompt excisions of injured extrapleural tissues, which without adding to immediate dangers give protection against inflammation in parietal tissues that can cause pyothorax and frequently reveal unsuspected deeper lesions demanding more radical operation.

Parietal excisions should be performed under positive pressure gas analgesia, because these minor operations can thus be more rapidly and safely conducted. Discomforts are less than when local anesthesia is employed; there is less likelihood of spreading infections and the exposure of a sucking wound is not accompanied by pulmonary collapse. Administration of ether by open methods is unwarranted.


Aspirations or suitable drainage of hemothorax should be almost routinely employed with parietal excisions to reduce pleuritis and be followed by systematic exercises to minimize the effects of pleuritis. This combination assures the most complete, undelayed recoveries.


The chief objections to routine parietal excision have been: (a) It increases immediate dangers; (b) it is too time-consuming, especially during periods of active fighting when simple aspiration with or without air replacement suffices: (c) it overburdens the forward hospitals and nurses.

(a) Immediate dangers are not increased and ultimate dangers are reduced. (b) Less time and attention are required to obtain healing if wounds receive prompt attention. Reasons have been given to show why simple aspiration is inadequate and ofttimes dangerous in spite of the many excellent recoveries that may be secured. Air replacement is unsound therapy. Its one excuse is to stop hemorrhage from lacerated lung and can only be effective by producing pulmonary compression. Not only is pulmonary compression undesirable from every standpoint, but also it must be controlled by manometric determinations which are time-consuming. (c) Virtually all wounds must be given some attention and all but those receiving trivial injuries must be hospitalized. Hospital facilities of an army have physical limitations which can be increased only by reducing the durations of disabilities. When the numbers of severely wounded overtax forward hospitals, those suffering from less severe chest wounds can be sent as far back as they can be delivered within 24 hours while still in good condition. This plan in the long run would be more effective than the giving of makeshift early treatment that assures more prolonged and less complete recoveries. Parietal excisions, with and without immediate aspiration or primary drainage, are applicable to the least severe chest wounds or to those in such critical condition that no further intervention may be attempted.


1. E. B. July 20, 191S: Bullet, through-and-through wound, both sides of chest. Entrance at left nipple; exit above right nipple. No physical signs of intrapleural involvement. No X-ray examination. Duration nine hours. Condition (?). Operation: Entrance and exit wounds excised and connected. Fractured ribs and sternum resected. Wounds in lung sutured and in pleura closed with muscle. No aspiration. July 23, 1918: Hemothorax with increased pleuritic exudate, right side. Dyspnea. July 24, 1918: Aspiration unsuccessful. Condition poor. Hospital moved forward. Patient left in poor condition August 2, 1918. Died. No further notes.

Notwithstanding severe injuries affecting both pleural cavities and anterior mediastinum, this man had the power to recover had proper primary drainage been employed.

2. A. R. July 20, 1918: Shell fragment, penetrating wound, sucking, right chest. Entrance over scapula. Large hemothorax. Foreign body present. Duration (unknown).Condition (?). Operation: Lacerated tissue excised; fractured scapula and rib resected. Muscles closed over sucking wound in pleura. Foreign body not removed. Severed supra-scapular nerve sutured. July 24, 1918: Uncomfortable convalescence. Pleuritic exudate disappearing. Evacuated in good condition. 1921. Records available. Disability below10 per cent.


This man's recovery was due to good luck rather than to good management. Hole in lung should have been repaired; chest aspirated, possibly drained if the wound was of long duration because of the sucking type.

3. A. F. G. July 21, 1918: Bullet, through-and-through, sucking wound, left chest. Entrance over sixth rib, anterior axillary line; exit, paraspinal line, level of twelfth rib. Hemopneumothorax moderate. Duration (unknown). Condition (?). Operation: Excision of wounds of entrance and exit. Resection of fractured ninth, tenth, and eleventh ribs. No lung injury recognized at operation. Closure of parietal defect. July 24, 1918: Rapid convalescence. Still some heniopneumothorax though disappearing. Evacuated in excellent condition. No further records. Disability (?).

Immediate aspiration was probably employed but not recorded. Sub- sequent aspiration was indicated, possibly primary drainage if the sucking wound was of more than a few hours' duration.

4. C. J. D. July 22, 1918: Shell fragments, multiple wounds, right jaw, neck, shoulder, and right chest, sucking, perforating. Wound of entrance in second interspace anteriorly. Moderate hemopneumnothorax. Heart displaced to left. Duration (unknown). Condition good. Operation: Extrathoracic wounds excised and packed. Wound of entrance excised. Foreign body, thought to be in lung, not sought. Hemothorax aspirated. Parietal wound closed. July 24, 1918: Condition good although pleuritic exudate is not receding. Evacuated in good condition. August, 1918: Phlebitis right leg. August 30, 1918: Returned to United States. February 6, 1919: Discharged from service. August 5, 1921: Disability attributable to leg alone. Vital capacity 87 per cent. Slight pleural thickening at right base. Heart competent. Foreign body in posterior chest wall.

Another recovery attributable to good fortune. An undrained sucking wound causing no pyothorax. Quite probably this man's condition after excision of multiple wounds prohibited more radical intervention.

5. W. D. S. July 22, 1918: Bullet, perforating wound, left chest. Entrance fifth interspace, midelavicular line; lodgment in ninth interspace, close to vertebra. Moderate hemopneumothorax. Duration (unknown). Condition (?). Operation: Excision of wound of entrance. No rib injury. Opening into pleura closed. Incision over foreign body posteriorly. Foreign body removed. No rib injury. Hemothorax aspirated through defect in parietal pleura which was then closed. July 24, 1918: Easy convalescence. Evacuated in good condition. November 20, 1918: Returned to organization (Infantry). December 20, 1918: Readmitted to hospital because of trouble in knee, supposed to be synovitis. May 10, 1919: Discharged from service. Given a disability of but 10 per cent because of chronic synovitis; none because of chest injury. August 5, 1921: Synovitis proved to be due to sclerosis of pyramidal tract due to injury to spine. Heart and lungs quite normal. Vital capacity, 98 percent.

Without knowledge of man's condition at time of operation, or duration of wound, it is unwise to say that more radical treatment would have assured better repair. It could have secured none as a zero disability attests.

6. J. H. T. July 23, 1918: Shell fragment, penetrating, sucking wound, right chest. Entrance over scapula; compound fracture of scapula and three subjacent ribs. Small hemothorax. Duration (unknown). Condition fair. Operation: Excision of entrance wound; resection major portion of fractured scapula and ribs. Very large foreign body removed from pleural cavity. Hemothorax aspirated. No lung injury required repair. Closure of pleural defect with muscle flaps and subcutaneous fat. July 24, 1918: Superficial healing good. Some pneumothorax persists. Uncomfortable but otherwise in excellent condition. Evacuated later much improved. 1921: Official record show disability less than 10 percent.


This recovery is another to be attributed to good fortune as the pleural cavity was much soiled because of a large sucking wound. Suitable drainage would have hastened recovery and made it the more certain.

7. M. D. July 27, 1918: Shell fragments; wounds of left neck and arm with penetrating wound of left chest. Entrance over eighth rib near angle of scapula. Moderate-sized heinothorax. Duration (unknown). Condition (?). Operation: Excision of entrance wound. Fractured eighth rib resected. Inner table of rib found extending into pleural cavity. Hemothorax evacuated. No note on foreign body. Lung distended normally. Muscle closure of pleural defect. Layer closure of superficial structure; drained with gutta-percha. July 31, 1918: Excellent recovery. Pleuritic exudate slight and diminishing. Hemothorax here due to bleeding from rib. Left chest slightly hazy. Foreign body 0.5 by 0.5 cm. in lung. Diaphragm motion present; restricted on left. Evacuated in good condition. Later returned to duty and then lost. Disability (estimated) 10 percent.

Removal of so small a foreign body from the lung is contraindicated. The case illustrated the wisdom of exploring all chest wounds. The rib injury would otherwise have escaped attention until late complications arose. This man would have benefited by a day or two of primary drainage or one postoperative aspiration.

8. L. E. July 28, 1918: Bullet, through-and-through, sucking wound, right chest. Entrance over right third rib, parasternal line. Exit over ninth rib below angle of scapula. Small hemothorax. Condition poor. Duration (unknown). Operation: Excision of en- trance and exit wounds. Fractures of fourth, fifth, sixth, seventh, eighth, and ninth ribs resected. Hemothorax evacuated. Splenized lung not resected because of patient's poor condition. Pleural defect closed with muscle. Death within 12 hours. No notes on necropsy.

This man was lethally injured by the time he came to operation. Nevertheless the splenized lung should have been resected, as allowing it to remain could at the best only postpone death. Earlier operation and multiple transfusions could have been effective.

9. M. S. September 26, 1918: Shell fragment, perforating right chest. Entrance right upper chest, anterior. Exit right lower chest, at level of tenth rib, posterior. Foreign body under skin near vertebra. Right hemothorax, moderate. Abdomen rigid, indicating possible injury to diaphragm. Duration 9 ½ hours. Condition good. Operation: Eight and a half hours later. Wound of entrance excised. No rib injury. Pleural defect closed by stitching inflated lung to margins and by muscle flaps superimposed externally. Incision over foreign body which was removed and dark blood evacuated. Again no rib injury. Hemothorax aspirated through pleural defect. No lung injury found. Pleura closed with muscle. September 30, 1918: Condition good. Little pleuritic fluid evacuated. January 15, 1919: Returned to duty. Disability (estimated) 10 percent.

Growing experience led to more effective operations and attempts were made to expose lung injuries. Same fault of not using primary drainage is noteworthy.

10. S. F. S. September 26, 1918: Shell fragment, through-and-through, sucking wound; right chest. Entrance at eighth rib, posterior axillary line. Exit at third rib above scapula. Small hemothorax. Duration 14½ hours. Condition poor. Operation: Begun after six hours' treatment for shock. Entrance wound excised. No rib injury. Hemothorax aspirated. Pleural opening closed. Exit wound excised. Fractured rib resected. Chronic adhesive pleuritis prevented any exploration. Large intrapleural cavity drained. Incomplete layer closure. October 4, 1918: Has done fairly well except for persistent effusion in lower chest from which 1,300 c. c. of blood-stained fluid was aspirated. Culture


negative. Hereafter improvement was more rapid. Evacuated in good condition. Returned to duty in six months. Disability (estimated) 10 percent.

Recovery would have been hastened by earlier aspiration of pleuritic effusion, though primary drainage would have been still better.

11. M. S. September 26, 1918: Bullet, through-and-through wound, right chest. Entrance wound, anterior, at margin of rectus muscle; exit wound posteriorly over seventh rib. Small hemothorax. Splenization in right lower lobe suspected. Duration 10 hours. Condition fair. Operation: After six and a half hours' preparation. Entrance wound excised. No rib injury. Pleural opening and superficial wound closed. Exit wound excised. Fractured seventh rib resected. Hemothorax aspirated through this pleural opening. Lung injury not repaired. No splenization noted. Parietal pleura defect plugged with muscle. Wound closed in layers. No drainage. October 5, 1918: Wound healing good. Pneumothorax absorbed. Pleuritic effusion persists, but not aspirated. Pneumonia left upper lobe. October 6, 1918: 500 c. c. bloody fluid aspirated from right chest. October 7, 1918: Partial pneumothorax evacuated; condition fair. August 4, 1921: No treatment subsequent to evacuation. Suffers from dyspnea and cyanosis on exertion. Chronic bronchitis, peribronchitis, fibrous pleuritis; dome of diaphragm adherent to ninth rib; myocardial deficiency. Vital capacity, 57 percent. Disability allowance of 10 per cent much too low; should be 40 percent.

This man's treatment was improper partly because of battle pressure. A more finished operation with suture of lung wounds should have been performed and with drainage and postoperative care would have hastened recovery and reduced disability, notably in preventing recurrence of pneumothorax.  An excellent example of the wisdom of exploring all through-and-through wounds to exclude or to remedy rib injuries. The occurrence of a contralateral pneumonia is noteworthy because so infrequent with positive pressure gas analgesia.

12. J. G. September 27, 1918: Bullet, through-and-through, sucking wound, left chest. Entrance at second rib, parasternal line; exit below and posterior to angle of scapula. Fluoroscopic diagnosis: Moderate hemothorax. Both lobes perforated; splenization of lower part of upper lobe; heart displaced to right; no rib injuries recognized. Duration 12 hours. Condition poor. Operation: Nine hours later during which he was treated for shock. Wound of entrance excised. No rib damage. Closed without drainage. Exit wound excised; revealed sucking wound, fractures of sixth and seventh ribs, which were resected. Fragments of sixth rib had penetrated visceral pleura and were removed; 200 c. c. of fluid blood aspirated. Clots not removed and lung not repaired because of weak condition of patient. October 5, 1918: Stormy convalescence. Considerable effusion with pneumothorax. Evacuated in fair condition. No subsequent treatment. October ?, 1918: Returned to United States. January 23, 1919: Discharged from service. July 29, 1921: Underweight. Dyspnea on extra exertion. Pleuritic thickening base of left lung. Dome of left diaphragm adherent to sixth rib. Shallow pneumothorax cavity beneath seventh, eighth, and ninth ribs covered by thick scar. Lung parenchyma corresponding to injury and to pneumothorax cavity does not function. Myocardium fair. Vital capacity 90 percent effected by compensatory emphysema. Disability allowance of 18 percent is low because heart muscle is only competent at rest and has narrowed reserve power.

This man's experience emphasizes important points. The folly of conservatism in treating through-and-through wounds. These rib fractures untreated would have led to persistent pneumothorax. Under favorable circumstances open thoracotomy and radical treatment of lung defects would have been indicated. Splenization may not always cause death but by the resultant scar restricts pulmonary elasticity and causes disability. The


persistent partial pneumothorax is an example of the permanent total pneumothorax that occurs when differential pressures are not employed. Suitable drainage would have corrected this fault. Neglect of breathing exercises through absence of all continued treatment increased the total disability materially and perhaps prevented this man from returning to his pre-war occupation.

13. H. B. September 27, 1918: Shell fragment, penetrating left chest. Entrance wound second left interspace, parasternal line; small foreign body moves with respiration. Hemoptysis marked; hernothorax large; considerable pneumothorax. Duration 24 hours. Condition good. Operation: Two hours later. Wound of entrance excised. Fractured fourth rib at costochondral junction resected. 1,000 c. c. of blood and some clots removed. Wound of entrance into lung not found. Foreign body not sought. Closure without drainage. October 2, 1918: Rapid recovery. Pleuritic exudate slight. No pneumothorax. Evacuated. November 5, 1918: Returned to duty 39 days after injury. April 21, 1919: Discharged from service. No further treatment. August 20, 1921: Complains of pain in left chest and dyspnea with slight exertion. Chronic pneumonia left upper lobe. Chronic pleuritis left base. Adhesions between diaphragm and ninth rib restricts its motion Myocardium, subcompetent. Vital capacity 112 percent, due to contralateral emphysema Foreign body present near base of lung. Disability 25 percent, due to heart.

This man's disability is due to scar tissue inside the lung and out. His prompt recovery indicates that the immediate treatment was adequate. The ultimate results show that proper postoperative care, including well systematized exercises, would have brought such a recovery as would have permitted him to return to his original occupation of farming. Again and again the heavy toll placed upon the wounded through failure to provide suitable after-care is exemplified.

14. E. McF. September 27. 1918: Bullet, through-and-through, sucking wound, right chest. Entrance, third rib anteriorly; exit, ninth rib posteriorly. Small hemothorax; large pneumothorax. Duration nine and a half hours. Condition grave. Resuscitation for 25 hours. Operation: Thirty-six hours after injury. Excision of wound of entrance and of exit in posterior axillary line. Compound comminuted fractures of fourth, fifth, sixth, seventh, eighth, and ninth ribs exposed and resected. Lung badly lacerated and splenized. Lacerations trimmed and sutured. Patient's condition was thought to contraindicate resection of splenized lung. Hemothorax removed. Pleural defect closed with muscle flaps over incompletely expanded lung. September 29, 1918: Transfusion 400 c.c. citrated blood. October 5, 1918: Partial pneumothorax above pleuritic exudate. Four hundred c.c. serosanguinous fluid exudate aspirated. Cocci in clumps. October 8, 1918: Somewhat improved. October 21, 1918: Died. Double empyema and peritonitis.

Shock prevented earlier operation and was thought to contraindicate resection of splenized lung. Obviously, resection should have been performed and with suitable drainage demanded by a sucking wound of this duration (36 hours) would have saved this life.

15. I. D. September 28, 1918: Shell fragment, 0.5 by 0.3 cm., penetrating, sucking wound, right chest. Entrance over spine of right scapula, thence through lung, diaphragm and liver to lodge in upper pole of right kidney. Large hemothorax. Duration eight hours. Condition poor. Resuscitation for seven hours. Operation: Fifteen hours after injury. Entrance wound excised. Comminuted fracture of tenth rib resected. Wound in lung insignificant and not repaired. Hemothorax aspirated and clots removed. Parietal defects closed with muscle flaps. No drainage. October 3, 1918: Had done well for a few days. Signs of bronchopneumnonia developed yesterday in left lung and fluid in right chest increased. Foul-smelling fluid containing many bacteria aspirated from pocket near angle of right


scapula. Open drainage. October 4, 1918: Died (sixth day after operation). Necropsy: Open drainage had caused some collapse. A second encapsulated empyema had not been reached. Left lower lobe and lower part of left upper lobe almost solid with confluent patches of bronchopneumonmia. Abdomen contained small amount of blood. No peritonitis.

Another of the exceptional instances of bronchopneumonia after positive pressure analgesia. Otherwise this man might not have succumbed not, withstanding his poor condition. Should have been drained.

16. O. O. M. September 2S, 1918: Shell fragments. Wounds of legs and head, and penetrating, sucking wound, left chest. Entrance over ninth rib below angle of scapula. Moderate hemothorax. Heart displaced to right. Foreign body in left lung. Duration 39 hours. Condition poor. Long journey, exposure and anemia. Resuscitation five hours. Operation: Forty-four hours after injury. Wounds in scalp and thigh excised Wound of entrance to chest excised. Fractured ninth rib resected. Hemothorax aspirated through defect in parietal pleura. Inflation of lung brought lacerations in left upper lobe into view which were bleeding profusely and so were sutured. Foreign body could be felt in lung but removal was not attempted as patient's condition was critical. October 18, 1918: Uneventful recovery. Diffuse pleuritis, left lung. No consolidation. Foreign body, 1.1 by 0.8 cm., present near hilum. Evacuated in good condition. December 28, 1918: Returned to duty (three months). No subsequent record. Disability (estimated) 10 percent.

This recovery when operation on a sucking wound had been delayed 44 hours illustrates the wisdom of denying none of the wounded the chance to live no matter what the operative mortality rate might be. It would have been wiser under the conditions to have employed primary drainage as the recovery without empyema was most fortunate under the conditions.

17. S. G. October 12, 1918: Bullet, through-and-through wound, left chest. Entrance at tip of left clavicle; exit at vertebral border of left scapula at level of its spine. "Small hemnothorax. Hemoptysis. Mustard gas burns, right side of face. Duration seven and a half hours. Condition good. Operation: Six hours later. Wounds of entrance and exit excised. Fracture of (?) rib resected. Muscle closed into pleural defect to protect exposed lacerated but adherent lung. Gutta-percha drain. October 14, 1918: Mustard gas burns on face much worse. Few rHles noted in right chest. Slight left pleuritic effusion. Evacuated in good condition.

Official records show death from bronchopneumonia (mustard gas) 25 (lays later. Chest disability would have been less than 10 percent.

18. H. B. October 15, 1918: Bullet, through-and-through wound, left chest. Entrance over second rib anteriorly; exit over fourth rib posteriorly. Moderate hemothorax. Duration 16 hours. Condition poor. Operation: Entrance and exit wounds excised. Fractured second rib anteriorly and third and fourth ribs posteriorly resected. Hemothorax aspirated. No lung repair. Pleura closed with muscle flaps.
October 16, 1918: Patient continues cyanotic. Tachycardia. No increase in pleural effusion. Died at noon. No necropsy. Death attributed to myocardial fatigue.

Less extensive operation incompatible with recovery; more extensive not indicated. Digitalis before and after operation with hypertonic glucose intravenously might have been effective. May be called a lethal injury.

19. J. C. October 15, 1918: Bullet, through-and-through wound, right chest. Entrance at inner end of clavicle; exit beneath spine of right scapula. Small hemothorax. Much bone damage. Condition grave. Duration 29 hours. Resuscitation 10 hours. Operation: Thirty-nine hours after injury. Entrance wound ignored. Exit wound excised. Scapula turned forward. Comminuted fracture of scapula and one rib resected. Hemothorax


aspirated. No lung injury repaired. Closure with superficial drainage. Stations changed. Learned of death but not its cause. No necropsy. Death attributed to shock.

Another example of a simple injury made lethal by delay and exposure.

20. L. L. October 19, 1918: Shell fragments, multiple wounds; compound fracture right femur; penetrating right chest. Entrance eighth interspace posterior axillary line. Foreign body, 2 cm. by 2 cm., immobile in right upper chest. Small hemopneumothorax; heart displaced to left; middle and upper lobes involved. No hemoptysis. Duration 20 hours. Condition serious. Operation: Entrance wound excised. Fractured ninth and tenth ribs resected. Small hemothorax. Clots and fibrin removed. Hole in lung sutured. Foreign body not removed. Area of splenization not resected because of lack of space. Closure .with superficial drain. Wound of thigh excised. Compound fracture of femur splinted. October 22, 1918: Pleuritic effusion increased and was aspirated. Bronchopneumonia and pleurisy, both lobes of left lung. Death. Necropsy: Bronchopneumonia left upper and lower with fibrinous pleurisy. Right lower lobe contained a tunnel wound in which were bone fragments; widely splenized. Fibrinous pleurisy. Upper and middle lobes normal.

This man's chance for recovery depended upon his chest repair. It would have been wiser here to have opened the chest widely, excised the lower lobe and neglected the thigh wound even if this meant ultimate sacrifice of leg. Another example of the serious import of splenization and of contralateral bronchopneumonia.

21. R. H. November 2, 1918: Shell fragments, left wrist; through-and-through wound, right chest. Entrance wound upper anterior chest; exit, posterior. Small hemothorax. Superficial wound left wrist. Condition good. Duration 24 hours. Operation: Entrance wound excised. Lung adherent. Closure with muscle flap. Exit wound excised. Slight rib injury resected. Hemothorax aspirated. Closed without drainage. Wrist wound dressed. November 11, 1918. Easy convalescence in spite of right-sided pleurisy and return of partial pneumothorax. Evacuated. Could not be traced.

A more radical operation at exit would have repaired lung injury, prevented recurrence of pneumothorax, and reduced the pleuritic effusion which should have been aspirated. Drainage would have been better.


The number treated by limited thoracotomv (21) was about 20 percent of the series. The number would have been larger had the unit had more of the less severely injured to treat and had the possibilities of such operations been appreciated.
Fatalities.-There were eight deaths (40 percent). One (17), slightly wounded, died on the 25th day from bronchopneumonia due to mustard gas. His chest disability would have been less than 10 percent. Three (8, 18, 19) were lethally affected at time of operation because of delay and exposure. One (18) died from myocardial exhaustion which had been incurred before injury. Three developed pyothorax (1, 14, 15). Two (1, 14) might well have recovered had more radical surgical treatment and primary drainage been employed. The third (15) had advanced contralateral pneumonia which possibly was traceable to failure to drain. One (20) suffered from other injuries, including a compound fracture of a thigh. Had the thigh wound been given less attention and the splenized portion of a lung resected, a recovery, perhaps with amputation, was conceivable. Duration from injury to operation was


from 9 to 39 hours, average 23 hours. Mortality chargeable to surgical errors, too conservative treatment of injured lung, and failure to drain is 15 percent.
Disabilities.-Late disability ratings are dependably established for 11 of the 13 who recovered. Two (4, 5) made complete recoveries. Two (2, 6) are rated at less than 10 percent; five (7, 9, 10, 11, 16) at 10 percent; one (2) at18 percent, and one (13) at 25 percent. Two (11, 12) have unjustly low ratings. The disability of one (13) is due to cardiac incompetence. Two (3, 21) were not estimated. The average disability is probably not far from 10 percent, but is set at 13 percent to err on the safer side. The few notes available showed from 39 to 180 days interval before return to duty, average about100 days.

Pleuritis is the cause of disability in all but the one (13) due to myocardial deficiency. An average disability of even 13 percent is gratifying when it be considered that some of these wounds were "suckers" and their duration about 24 hours, and also that none received proper postoperative care. The two who received unfairly low ratings had complications, one (11) had pneumonia and the other (12) a persistent pneumothorax and a lung scar resulting from splenization.

It is again evident that a reduction of the interval between injury and operation would have reduced mortality and disability rates; the latter (13 percent) would also have been favorably influenced by less conservatism in operating, a more general use of primary drainage and constant after-care. The duration of disabilities estimated at one hundred days upon the few notes at hand is higher than the truth and avoidably high.


Operations that can be classed as limited thoracotomies are applicable to those wounds that can not be safely treated by parietal excisions and to those severely wounded who can tolerate but little more than parietal excisions. The same methods, especially positive pressure gas analgesia, are indicated. Limited thoracotomy gives opportunity to repair less significant lung injuries, to aspirate pneumothorax with a cannula instead of a needle and at times to remove clots. Its disadvantages are incomplete exposures and the consequently great dangers of overlooking lesions that should be repaired. Its advantages are shorter and less trying operations and less interference with parietal integrity.


The more severe the injury, the greater the necessity for prompt intervention. Operations of the limited thoracotomy type were employed too infrequently and possibly can be used more generallv and more effectively hereafter if the routine after care is improved so that postoperative complications would receive prompt recognition and correction.

During periods of active fighting many of the wounded who might be well served by limited thoracotomy could be transported to evacuation hospitals for first treatment. A great danger lies in too conservative operating upon lung wounds. The objection to limited thoracotomy is its ease and rapidity of performance


formance and apparent safety when real conservatism would be a more radical operation.

The need for preoperative resuscitation of those in shock and the prevention of postoperative shock becomes more and more evident. Gum acacia proved of value when properly used and was a source of danger when improperly used, which was the rule. Hypertonic glucose and other solutions may be of more benefit than gum and should be provided. The most help comes from blood transfusions. Larger amounts of blood should be made available.


1. A. B. July 21, 1918. Shell fragment, large, freely bleeding, sucking, tangential wound, left chest. Entrance at fifth interspace, midelavicular line. Tissue emphysema. Large pneumothorax. Small hernothorax. Condition poor. Duration unknown. Operation: Lacerated soft parts excised; broken ribs resected. Lung wounds repaired. Flap closure. Death in 12 hours. Necropsy: Slight hemothorax. Lung repair and parietal closures adequate and apparently secure.

Multiple transfusions might have prevented this death.

2. H. K. July 27, 1918: Shell fragments; penetrating and through-and-through sucking wounds both chests. Main wound entrance seventh rib, right scapular line; exit seventh rib, left scapular line. Bilateral hemopncumothorax, right larger. Condition (?). Duration unknown. Operation: Entrance wound excised, exposing sucking wound. Fractured seventh rib resected. Crater defect in lung contained bone fragments. Fragments removed, bleeding controlled, lung approximated over defect but incompletely as man's condition contraindicated wider exposure. Pleural defect closed with aid of wire rib stay. Superficial wound drained. Wound of exit excised. No rib damage. July 31, 1918: Good recovery. Postoperative tissue emphysema showed that repair of lung wound had not been airtight. Left side, effusion absorbed. Right side, effusion is increasing. August 1, 1918: Spontaneous discharge of serum from right side. Partial plletlmothorax. Evacuated in good condition. February 1, 1919: Returned to duty in 180 days. Disability 10 percent.

Slightly wider exposure, more accurate closure of lung wound, right-sided drainage and aspiration of left chest would have been wiser.

3. V. I. P. July 27, 1918: Shell fragment, through-and-through, sucking wound, left chest. Entrance over 6th rib lateral to mnidelavicular line; exit ninth rib, posterior axillary line. Moderate hemothorax; heart displaced to right. Splenized lung seen fluoroscopically. Condition (?). Duration 30 hours. Operation: Entrance and exit wounds excised; fractured eighth and ninth ribs resected. Splenized lung excised. Pleural closure effected with wire rib stay. Layer closure. Superficial drain. July 31, 1918: Fluid in left chest receding. Patient sitting upright. Evacuated. Condition good. August 15, 1918:Pneumonia, left side, followed by empyema and treated with rib resection. Drainage tract closed spontaneously. February 8, 1919: Discharged from Army. July 21, 1921: Under-weight, languid, dyspneic. Scoliosis, reduced expansion, left chest; chronic pleuritis, limited motion and adhesions of diaphragm. Vital capacity 78 per cent. Disability 40 percent (estimated).

Pneumonia caused by compressed lung because of effusion.

Drainage would have lessened, had it not prevented this complication.

4. T. B. B. July 28, 1918: Shell fragment, through-and-through wound, left chest. Entrance seventh interspace anterior axillary line; exit tenth rib posterior axillaryline. Condition (?). Duration unknown. Operation: Wounds excised; fractured ninth and tenth ribs resected; lacerated and splenized portions of lower lobe resected. Two tears in diaphragm sutured. Wound in liver not treated. Pleural closure incomplete even with


rib stay so reenforced with muscle. July 31, 1918: Slight effusion; no peueinothorax. Motion of diaphragm reduced. Evacuated. July 30, 1921: Pain and dyspnea on extraexertion. Chronic pleuritis at left lease with adhesions between dome of diaphragm and eighth rib. Excursions of diaphragm restricted. Wire rib stay has parted. Vital capacity 81 percent. Heart competence slightly reduced. Disability rating of 35 percent is too high.

Wire stay should have been removed. Primary drainage and postoperative exercises would have reduced disability.

5. M. D. August 1, 1918: Shell fragment, penetrating right chest. Entrance wound ninth rib, paravertehral line. Foreign body, two cm. by three cm., in lung, which is cloudy. Moderate hemothorax; heart slightly displaced; diaphragm immobile. Condition good. Duration (?). Operation: Entrance wound excised; ninth rib, fractured just anterior to its angle, reseated; heniothorax evacuated; foreign body removed from posterior aspect lower lobe. Lung wounds sutured. Pleura closed after ribs were approximated with wire stay. Layer closure of soft parts. No drain. July 21, 1918: Effusion to level of angle of scapula. Good condition. Healing satisfactorily. Subsequently evacuated ingood condition. No further treatment. January 31, 1921: Discharged from service. August 19, 1921: Slight pain and dyspnea on exertion. Fibrous pleuritis at right base. Diaphragm adherent to eighth rib and excursions reduced. Heart competent. Wire rib stay broken. Vital capacity 81 percent. Disability of zero is too low.

Wire rib stay should have been removed. Aspiration, or better still drainage, would have reduced pleuritis. Lack of proper exercises prevented a perfect recovery.

6. M. P. August 1, 1918: Shell fragments, multiple, penetrating, sucking wounds, right chest, liver and colon. Entrance chest, ninth rib, anterior. Condition poor. Duration 10 hours. Operation: Resection compound fracture ninth rib. Multiple lung wounds, liver and colon. Wounds sutured as rapidly as possible because of man's condition. August 2, 1918: Died. Necropsy: Hemothorax right, splenization of middle lobe, collapse of lower lobe. Wound repair adequate.

Lethal injury at time of operation.

7. A. H. B. August 1, 1918: Shell fragment, left scapula, sucking, penetrating wound, chest. Bullet wound, right shoulder, penetrating. Foreign body, 1.5 cm. by 3 cm. in left utpper thorax. Chest hazy; diaphragm fixed. Condition desperate. Duration 12 hours. Operation: Excision entrance wound; resection left scapula and fractured fourth rib. Excision and suture of bleeding wound in upper lobe. Foreign body removed. Rib stay and muscle flap closure. Bullet wound ignored as patient had been carried this far with two gum-salt infusions. August 2, 1918: Slightly better. Blood transfusion. Fluid up to angle of scapula. Slight pneumiothorax. August 3, 1918: Condition about the same. Unit moved to another station. August 5, 1918: Died. No notes. No necropsy.

Severely wounded man in shock. Death probably due to pyothorax. Should have had primary drainage and blood transfusions before operation.

8. E. K. August 8, 1918: Shell fragments, right wrist, superficial, and through- and-through, suicking woIund, right chest. Entrance just to right of spine above angle of scapula. Tissue emphysema here. Exit at second rib midclavicular line. Right chest hazy. Condition poor. Weather hot, water scarce, troops fatigued. Duration 12 hours. Operation: Entrance wound excised. Compound fracture eighth rib resected. Lung adherent, lacerated and contains many bone fragments, resected and repaired as well as exposure and adhesions permitted. This area was drained as there was no connection with pleural cavity. Exit wound excised; fractured second rib resected. This tract also drained for same reason. August 15, 1918: Pyothorax and septicemia caused death. Aspirations, open drainage, and transfusions were futile. No necropsy.


This man was operated upon during a pestilence of flies that could not be kept off dressings or out of wounds; doubtful if it would have been possible to avoid this death.

9. J. F. C. August 8, 1918: Shell fragments, through-and-through wound, right arm and right chest. Notes few and vague. Entrance at angle of tenth rib; exit, anterior aspect of chest. Condition (?). Duration 55 hours. Operation: Entrance wound excised; fractured tenth rib resected. Large hemothorax and many clots removed. Lung sutured. Pleural defect repaired by stitching diaphragm to margins. Anterior entrance wound excised; rib not injured. Tear in lung not found. Wire rib stay used to obtain closure of pleural defect. Wound in arm excised and packed with gauze. August 17, 1918: Slight effusion persists. Evacuated in good condition. August 17, 1921: Notes by Doctor Byrne. "No parenchymatous lesions; dome of diaphragm adherent to seventh rib is flattened." Disability, vital capacity and myocardial competence unknown. Disability (?).

This patient illustrated the need of breathing exercises to reactivate a diaphragm, particularly after it has been sutured to the parietes. An excellent example of the wisdom, accepting all risks with faintest chance for recovery. Without operation this man's life expectancy would have been zero.

10. W. B. August 8, 1918: Shell fragment, large, penetrating, sucking wound, left chest. Entrance ninth rib, costochondral juncture. Condition good. Duration unknown. Operation: Wound excised; fractured ninth rib resected. Foreign body, found on diaphragm which was lacerated, was removed. Diaphragm sutured. Hemothorax and clots removed. Lung and diaphragm sutured. Closure with wire rib stay. August 17, 1918: Uncomfortable convalescence. Slight pneumothorax. Pleuritic effusion moderate and not aspirated. Suppuration occurred but did not cause pyothorax because the repair of parietal pleura had been adequate. Evacuated in good condition. February 13, 1919: Discharged from service having had otitis media which caused deafness. August (?), 1920: Wire rib stay removed. August 31, 1921: Some pain. No dyspnea. Underweight. Chronic pleuritis left base; diaphragm adherent to parietes. Reduced myocardial reserve power. Disability, 38 percent, not all due to chest. Vital capacity, 78 percent. Disability, 15 percent, due to chest, is liberal.

Wire rib stay should have been removed earlier. Excellent example of obtaining firm parietal pleural healing to prevent pyothorax. Aspiration and breathing exercises would have limited disability.

11. G. F. August 8, 1918: Shell fragment, perforating left chest. Entrance below middle of left clavicle; exit into soft parts beneath scapula. Condition poor. Duration six hours. Operation: Wounds excised; fractured ribs (second, third, and fourth) resected. Laceration and splenization left upper lobe. Laceration repaired; splenized lung not resected. Hemothorax aspirated. Pleural defect closed with muscle flaps. Posterior wound drained after removal of foreign body. Returned to shock ward with pulmonary edema. No response to treatment. Death in 16 hours. Necropsy: Pulmonary edema, hypostatic congestion both lungs. Dilatation, right heart. Splenization left upper lobe. Suture line intact.

This patient might have been saved, probably would, had he been wounded later when experience was larger. He would have been better prepared for operation. Splenized lung should have been resected. Too high positive pressures were used in a futile attempt to reinflate lung. Abruptly increased peripheral resistance added to that already present sufficed to cause dilatation of right heart and thus to provoke pulmonary edema.

12. German soldier. August 8, 1918: Shell fragment, seton wound over anterolateral aspect sixth rib, left. Condition poor. Duration 56 hours. Operation: Wound excised.


Fractured sixth rib resected. Lower one-half of upper lobe and most of lower lobe splenize. Fibrinous exudate on visceral pleura. Pleural cavity contained thin, red fluid (streptococcus?). Pericardium used to assist closure of pleural defect. Death soon after operation. No necropsy.

Illustrates severe types of splenization caused by tangential injury without laceration of lung. Also shows the greater reactability of visceral pleura upon which the fibrinous exudate had been formed. It is doubtful if prompt operation could have prevented a fatal issue because removal of almost all of the left lung would have been necessary.

13. M. S. August 9, 1918: Shell fragment, penetrating wound, left chest. Wound of entrance over sixth costal cartilage at sternal articulation. Moderate hemopneumothorax. Condition (?). Duration four hours. Operation: Wound of entrance excised. Margin of sternum and sixth costal cartilage resected. Hemothorax evacuated. Foreign body removed. No notes on lung injury and repair. Lung completely inflated. Closure difficult because of a defect close to sternum. August 16, 1918: Interstitial emphysema about wound and reappearance of pneumothorax showed that pleural closure had been inadequate. Moderate pleuritic exudate. By this time, emphysema, pneumothorax and effusion all were less. Excellent general condition. January 4, 1919: No further operative treatment; indeed, none of any kind. Discharged from service. Disability 80 percent. August 17, 1921: Tachycardia and dyspnea follow slight exertion. Frequent pain referred to lower left chest. Dome of left diaphragm adherent to parietes. Chronic pleuritis left base. Cardiac response to exercise poor. Left heart enlarged. Vital capacity, 55 percent.

Disability due in part to myocardial incompetence which in part is attributable to intrathoracic lesions. Rating of 80 percent is too high. Aspiration after operation and suitable care would have hastened convalescence and reduced disability.

14. E. K. K. August 11, 1918: Shell fragment, perforating wound, left chest. Entrance second interspace just lateral to sternum. Exit through scapula to lodge in infraspinatus muscle. Condition poor. Duration five hours. Operation: Excision entrance wound; resection second costal cartilage and part of second rib. Hemothorax aspirated; clots removed. Lung drawn out and perforation sutured. Lung sutured to close defect in parietal pleura which could not be approximated even with two wire rib stays. Foreign body removed from infraspinatus muscle. No attempt made to open chest posteriorly, as no rib injury was present and patient's condition was poor. Despite blood transfusions and infusions, patient died within 12 hours. Necropsy: Moderate splenization left upper lobe. Posterior perforation not closed. Hypostatic congestion right lung.

Extent and duration of anatomic injuries and severity of operation do not explain death, which must be attributed to general and myocardial exhaustion previous to injury. Splenization should have been noted at operation.

15. W. S. August 12, 1918: Shell fragment, penetrating wound, right chest. Entrance sixth interspace anterolateral aspect. Condition good. Duration four hours. Operation: Wound excised; ninth and tenth ribs, fractured in posterior axillary line, resected. Tangential wound of lung. Moderate splenization. Hemothorax evacuated. Resection and suture of lung. Pleural closure obtained with two wire rib stays. Layer closure of soft parts. No drainage. August 17, 1918: Limited pleuritic effusion. Condition excellent. November -, 1918: Returned to duty. Disability less than 10 percent.

Aspiration should have been performed. Wire rib stays should have been removed. Despite errors and omissions in treatment, the man is reported to have participated in active fighting again.

16. P. F. M. August 17, 1918: Shell fragments (three), penetrating wounds, right chest. No notes on condition or duration. Condition probably unsatisfactory, as was


the rule with the wounded from this division. Moreover, the hospital facilities were wretched. For example, when this man was treated the electrical plant was not working. There was no X ray; no current for headlight. Operation: Wound of entraiiec over tenth rib in midaxillary line excised; fractured rib resected; pleural cavity opened widely and heinothorax evacuated. Injuries to lung not severe. Two small foreign bodies not removed. One perforating wound of diaphragm and penetrating would of liver. Exploration failed to locate foreign body. Tract in liver packed with gauze which was brought out through laceration in diaphragm. The diaphragm was then sutured to parietal pleura for partial closure and to protect pleural cavity from bile. Operation unsatisfactory cause of no headlight. Balance of closure secured with aid of a wire rib stay. August 20, 1918: Condition excellent. Slight pleuritic exudate. Evacuated shortly. February 28, 1919: Passed through several hospitals in France. No further surgical treatment until today when wire rib stay was removed. April 29, 1919: Because of persistent pain two shell fragments were removed from lung. September 18, 1919: Pain still persists. Foreign body removed from liver. August 4, 1921: Pain in lower right chest; dyspnea on exertion; hemoptysis at intervals. Parietes affected by sear and removal of two ribs, one additional rib and part of sear from futile removal of foreign bodies. Restricted expansion of right lower chest; restricted motion of right diaphragm; chronic adhesive pleuritis. Heart competent. Vital capacity, 66 percent. Disability of 35 percent is fair.

Foreign bodies might better have been removed at first operation, although they did no harm. Wire rib stay should have been removed in a few weeks. Subsequent care would have reduced disabilities. It is noteworthy that removal of foreign bodies did not relieve pain, introduced hemoptysis, an increased disability.

17. J. T. A. September 12, 1918: Shell fragment, sucking, tangential wound, left chest; entrance fourth interslace, left midaxillary line; exit ninth interspace, mid-scapular line. Moderate hemopneumothorax; tissue emphysema; heart displace to right; left diaphragm motionless. Condition poor. Duration five hours. Operation three hours later: Wounds of entrance and exit excised and united. Fractured eighth and ninth ribs resected. Multiple tears in parietal pleura. Wound of entrance in upper lobe not exposed. Wound of exit, lacerated and contained many bone fragments, was resected and repaired. Hemothorax removed. Closure obtained with one wire stay; suturing lung to parietes, vas unsatisfactory and assured subsequent emphysema. Superficial closure also incomplete. Shock prevented exact methods. September 21, 1918: Recovery in spite of complications. Interstitial emphysema developed, followed, as it commonly is, by suppuration, and finally an open pyothorax. Evacuated in fair condition. February 18, 1919: Suppuration continued. Ribs resected and wire stay removed. July 7, 1920: Hospitalized for three months because of suspected, but not proved, pulnmonary tuberculosis. August 11, 1921: Underweight; frail; pain and disability upon exertion. Parietal scars and defects limit expansion of lower left chest. Pericardium adherent to diaphragm and diaphragm to parietes; pleural thickening. Heart competent. Vital capacity, 106 per cent. Disability rating of 75 per cent is higher than findings warrant.

A more finished operation should have been performed to secure better healing. Drainage could have been employed to reduce dangers of pyothorax. This patient illustrates the harm coming from provisional suturing of sucking wounds. His after-care was not satisfactory.

18. F. F. September 12, 1918: Shell fragment, perforating wound, left chest. Entrance posterior axillary fold at level of angle of scapula. Foreign body, .4 cm. by 1 cm., in erector spinae muscles. Condition (?). Duration six hours. Operation: Entrance wound excised. Fractured ninth rib resected. Hemothorax evacuated. No lung injury seen. Second incision posteriorly over shell fragment failed to discover tile foreign body but revealed more serious rib injuries. A third incision made lateral to the second. Rib fractures resected; pleura opened, revealing tangential injury in lung containing bone fragments. Fragments removed; lung resected and sutured; more blood and many large clots


removed. Closure obtained with One rib stay. September 21, 1918: Some pneumothorax persisted. Slight pleuritic effusion absorbed after aspiration of 75 c. c. sterile bloody fluid. Temperature varied from 99 0 to 102 0 in spite of smooth healing; 97 0 on evacuation. 1921: Records show disability less than 10 per cent.

Another illustration of the more serious injury occurring at exit wound and of the wisdom of thorough exploration of all possible parietal injuries. Had the posterior wound been neglected, pyothorax and possibly death would have occurred. The wire rib stay should have been removed.

19. J. R. September 12, 1918: Shell fragment, penetrating wound, left chest; entrance lateral to spine at level of tenth rib. Foreign body in lung (3.5 cm. by 3.5 cm.). Considerable hemothorax. Shock treatment. Duration unknown. Operation: Entrance wound excised. Fractured ribs resected. Entrance wound into lung not found. Foreign body not removed because patient's condition was critical. Large hemothorax evacuated. Closure with wire stay unsatisfactory. Prognosis for healing, poor. Shock treatment. September 21, 1918: Stormy convalescence. Superficial suppuration but no open pyothora. Pneumothorax with pleuritic effusion; 600 e. c. serosanguinous fluid aspirated and superficial drainage instituted. Convalescent four months. Final records not available. Estimated disability 15 percent.

Attempt should have been made to repair entrance wound into lung. Primary drainage should have hastened recovery.

20. H. W. September 12, 1918: Shell fragments, multiple wounds, viz, compound fracture upper third, right tibia, foreign bodies in knee, calf and thigh; sucking wound, left chest. Entrance at eighth rib, anterior axillary line; foreign bodies in upper left chest and below dome of diaphragm, thought to be in lung tissue. Hemoptvsis and hematemesis. Condition so critical that thorough examination was impossible. Shock treatment for eight hours before operation. Duration (?). Operation: Wounds in extremities excised and foreign bodies removed. Entrance wound into chest excised. Fractured eighth rib resected. Foreign body not found. Search for perforation in diaphragm unsuccessful. Large hemothorax evacuated. Closure with one wire stay. Transfused. September 13, 1918: Condition continued fair until sudden weakening at noon. Transfusions unavailing. Died at 2.30 p. in. Necropsy: Peritonitis, lesser cavity, hemoperitoneum, through and through wounds of spleen and stomach. Chronic nephritis.
This patient's condition was so critical that preoperative examinations were restricted. Intra-abdominal injuries were suspected but a laparotomy was impossible. His one chance was that he could take care of his peritoneal wounds spontaneously. His injuries were lethal when he reached the hospital.

21. H. B. (German soldier). September 13, 1918: Shell fragment, perforating wound, right chest. Entrance, ninth rib posteriorly; foreign body beneath right clavicle. Condition fair. Duration 15½ hours. Operation: Entrance wound excised. Fractured ninth rib resected. Upper lobe adherent, previous pleurisy. Lower lobe, craterlike defect, excised and sutured. Hemothorax evacuated. Wire rib stay. Inflated lung sutured into parietal pleural defect. Foreign body removed from beneath clavicle. No rib injury. Closure without drainage. September 21, 1918: Evacuated; condition excellent. Disability estimated at 10 percent.

22. M. R. S. September 17, 1918: Bullet, through-and-through wound, right chest. Entrance just below angle of scapula; exit just above nipple, both sucking. Slight hemothorax. Condition good. Duration unknown. Operation: Entrance wound excised. Ninth rib, incompletely fractured, resected. Large laceration, lower lobe, resected and repaired. Hemothorax evacuated. Pleural repair incomplete. Lung sutured into defect. Wound of exit excised. No rib injury. Lung wound closed with purse-string suture and sewed into parietal defect. September 20, 1918: Condition remarkably good. Healing excellent. 1921: Records show disability of less than 10 percent.


 Treatment given to this man was effective except for postoperative exercises. Another example of the value of exploring all through-and-through bullet wounds even though the wounded are in splendid condition.

23. O. B. (German soldier). September 13, 1918: Shell fragment, through-and-through wound, left chest. Entrance below outer third of clavicle; exit below angle of scapula. Condition (?). Duration unknown. Operation: Excision of entrance and exit wounds revealed fractures of second, third, fourth, fifth, sixth, seventh, and eighth ribs in axillary line. Resected ribs exposed deep gutter wound in left upper lobe which was adherent and made lung resection and repair difficult. Large hematoma evacuated from beneath pectoralis major muscle. Plastic closure. September 20, 1918: This man's recovery was remarkable. None others survived such severe multiple rib injuries. Another example of the wisdom of operating despite unfavorable prognosis.

24. H. D. B. September 26, 1918: Shell fragments, through-and-through wound, left thigh, and penetrating wound, right chest. Foreign body, 8 cm. by 10 cm., beneath lower sternum. Entrance posteriorly over twelfth rib. Condition (?). Duration five hours. Operation two hours after admission: Through-and-through wound left thigh excised and drained. Large entrance wound of chest excised. Fractured eleventh and twelfth ribs resected. Foreign body had lacerated diaphragm and liver and lodged in diaphragm beneath sternum. No lung injury. Bile in pleural cavity. Large hemothorax removed, likewise foreign body. Wounds in diaphragm sutured. Drain inserted between liver and diaphragm and diaphragm sutured to parietal pleura to aid in closure and to exclude drain from pleural cavity. October 8, 1918: Satisfactory convalescence. Slight pleuritic effusion. Suppuration in superficial wound but no open pyothorax. Disabled for six months. Ultimate disability less than 10 per cent (estimated).
Delayed recovery due to lack of proper exercise. Primary drainage indicated because of bile in pleural cavity and tendency to cause empyema.

25. C. P. F. September 26, 1918: Shell fragments, through-and-through wounds, left arm near axilla, and left chest. Entrance, seventh interspace, midaxillary line; exit, tenth interspace midscapular line. Paresis of flexors of first finger. Moderate hemopneumothorax. Heart displaced to right. Condition poor. Resuscitation 6½ hours. Duration 11 hours. Operation: Excision wound in upper arm; suture of injured nerve trunk. Wounds of entrance and exit excised and joined. Fractured tenth rib resected. Lacerations in lower lobe and in diaphragm repaired. Parietal closure without drainage. September 29, 1918: Condition and healing satisfactory. July 28, 1921: Duly severe symptoms referable to arm. Slight fibrous pleuritis, left base; diaphragm free. Vital capacity 97 percent. Disability (chest) zero.

Operation not only protected this man from death, but initiated a perfect recovery. Illustrates advantage of exploring all wounds and giving deep injuries proper treatment. Thoracotomy, even when patients are in poor condi- tion, can be less dangerous than hemothorax.

26. W. S. September 26, 1918: Bullet, penetrating left chest; entrance, third interspace, parasternal line. Bullet beneath sternum moves with respiration, but not with heart beat. No hemoptysis. No hemothorax. Friction rub audible over precordium. Mediastinal emphysema (?); hemopericardium (?); pulse slow but irregular. Condition fair. Duration five hours. Operation six and one half hours later: Entrance wound excised. Bullet found with its nose penetrating the wall of a serous cavity, probably pericardial. Bullet removed and hole closed with suture. Path of bullet extrapleural. Wound closed tight. September 29, 1918: Signs and symptoms cleared since operation. Healing excellent. 1921: Disability less than 10 percent.

Removal of this foreign body was required. Result shows methods were good.

 27. M. J. T. September 29, 1918: Bullet, through-and-through, sucking wound, right chest. Entrance, ninth rib, posterior axillary line; exit, slightly lower in anterior axillary


line. Both plugged with gauze. Hemoptysis. Moderate hemopneumothorax. Resuscitation for four hours. Duration nine hours. Operation: Entrance and exit wounds excised and joined. Portion of ninth rib, found dangling into pleural cavity, removed, and ends of ninth rib resected. Gutter wound in lung excised and sutured. Lung could not be entirely reinflated, so stitched to parietes. Large hemothorax evacuated. Closure unsatisfactory. Patient's condition prevented a finished operation. Given 700 c. c. of gum salt solution on operating table. October 18, 1918: Complicated convalescence ending in open pyothorax. December 1, 1918: Drainage ceased. No further operations. July 29, 1921: Suffers from pain and dyspnea on moderate exertion. Chronic pleuritis. Diaphragm attached high to parietes. Myocardial competence slightly impaired. Vital capacity, 77 percent. Disability allowance of 20 percent is low.

Partial inflation of lung at operation indicated need of one-way drainage, as it suggested probable empyema. Even this imperfect operation, made possible by exploration, contributed to recovery as it eliminated bone fragments and osteomyelitis of ribs as a complication of empyema.

28. C. H. September 26, 1918: Bullet, perforating wound, right chest. Entrance close to spine at level of angle of scapula; exit from chest at second rib to lodge beneath the skin. Small bemothorax. Condition poor. Duration 14½ hours. Operation: Sixteen hours later; delay for resuscitation. Wound of entrance excised. Incision made to remove foreign body. Wounds united. Fractured second, third, fourth, fifth, and sixth ribs resected. Parietal pleura lacerated; visceral pleura intact. Hemothorax (1,500 c. c.) removed. Airtight pleural repair. Wound closed without drainage. October 2, 1918: Rapid recovery. Healing smooth. August 19, 1921: No treatment subsequent to operation. Pain with sudden exertion. Overweight. Myocardium competent. Restricted parietal mobility due to malunion of fractured ribs. Vital capacity 87 per cent. Disability of 20 per cent is high.

No better results obtainable without after-care. Removal of rib fragments and immediate reinflation protected this man against empyema and pulmonary compression.

29. T. E. L. September 26, 1918: Bullet, through-and-through wound, right chest. Entrance at level of twelfth rib posteriorly; exit at fifth interspace, anterior axillary line. Moderate hemothorax. Paralysis of right diaphragm. Hematuria. Condition fair. Dura- tion nine hours. Operation four hours later: Entrance wound excised; incision carried along twelfth rib, which was fractured and was resected. Kidney delivered. Subeapsular clots removed and large transverse tear repaired. Laceration in liver drained and in diaphragm sutured. Hemothorax evacuated. Puncture wound of lung not repaired. Parietal pleura closed. Exit wound excised; no rib injury found. Closed tight. October 18, 1918: Good recovery. Slight amount of pleuritic effusion. Free drainage of urine and bile, but temporary. Evacuated in good condition. Later suffered from influenza. July 30, 1921: Complains of pain and weakness in right chest. Some fibrosis of lower lobe of lung. Diaphragm but little affected. Pronounced scoliosis. Vital capacity 92 per cent. Disability rating 100 per cent is ridiculously high. Needs only vocational training to be self-supporting.

Suitable postoperative care would have reduced this man's disability and enabled him to enter a profitable occupation.

30. J. A. September 27, 1918: Bullet, perforating, sucking wound, right chest. Entrance over eleventh rib, midscapular line; exit through diaphragm into liver. Large hemothorax. Condition fair. Duration 14 hours. Operation: Twelve hours later: Entrance wound excised; fractured twelfth rib resected. Perforations in lower lobe, diaphragm and liver repaired. Foreign body not found. Hemothorax, bile-stained, evacuated. Closure unsatisfactory. October 3, 1918: Residual pneumothorax absorbed. Slight pleuritic exudate. Condition good. July 29, 1921: Suffered from influenza before leaving France. Otherwise recovery without complications or after-care. Complains of pain in right chest


and is easily fatigued. Cardiac competence is fair. Fibrous pleuritis at right base. Dia- phragmatic excursions limited by adhesions. Disability of 20 percent is adequate.

Recovery by good fortune. Bile in pleural cavity usually produces an intense reaction and needls primary drainage.

31. C. C. September 26, 1918: Bullet, penetrating, sucking wound, right chest. Entrance, sixth interspace, midaxillary line, lodgment back of heart. Condition wretched. Blood pressure 68/50. Duration 9½ hours. Resuscitation treatment, pressures raised in 3 hours to 90/65. Operation: Eleven hours later: Entrance wound excised; fractured sixth rib resected. Three holes in lung repaired; fourth could not be reached. Bullet removed from behind heart. Difficulty in checking hemorrhage from azygos vein. Large, clotted hemothorax removed. Tight closure obtained. Gum salt given at close of operation. October 10, 1918: Developed contralateral pneumonia after operation. Empyema developed; wound opened revealing a bronchial fistula, the probable cause of empyema and possibly attributable to unclosed perforation. Death fourteenth day. No necropsy.

Man's condition thought to be too precarious to justify opening chest widely enough to repair fourth perforation. This might have saved life. So far as known this is the only bronchial fistula that occurred. This is an excel- lent example of resuscitation.

32. G. S. (German soldier). September 27, 1918: Shell fragment, sucking, penetrating wound, right chest. Entrance over tenth rib anterior axillary line. Foreign body three cm. by three cm. lodged in liver. Large pneumolielnotlhorax. Condition (?). Duration 25 hours. Operation 5 hours later: Entrance wound excised; fractured rib resected. Hemothorax removed. Two holes in lower right lobe sutured. Foreign body removed from liver; hole in diaphragm sutured. Wound closed without wire rib stay. September 30, 1918: Aspirated; no fluid obtained. October 6, 1918: Wound ruptured; seropurulent fluid escaped. Tube inserted; end covered with gutta-percha valve. October 18, 1918: Evacuated in good condition. Lung expansion excellent.

Sucking wound, 30 hours old; bile in pleural cavity; all indicated primary drainage. Rupture of incision from within of wound closed without rib stay indicates the value of that suture.

33. M. L. September 28, 1918: Bullet, through-and-through wound, right chest. Entrance, third rib, parasternal line; exit, costal margin, anterior axillary line from which bile was escaping. Condition poor. Duration 38 hours. Operation: Entrance wound ignored. Fractured sixth, seventh, and eighth ribs resected. Hemothorax evacuated. Holes in lung and diaphragm sutured. Drainage to liver wound. No notes on convalescence or at discharge. January 7, 1919: Small amount of fluid aspirated. August 28, 1921: Pain and dyspnea only after sharp exertion. Fibrous pleurisy. Diaphragm fixed. Costophrenic sulcus obliterated. Disability 20 percent. Vital capacity 71 percent.

This man's condition prevented complete operation. Had primary drainage been used with proper after care disability would have been less. Aspirated second day, vet fluid was withdrawn four months later. An example of slow absorption of effusions as well as the wisdom of operating even if there has been delay.

34. S. D. September 28, 1918: Bullet, through-and-through wound, left shoulder and chest. Entrance just to left of vertebra; exit, high in axilla. Condition very poor. Duration 23 hours. Operation: Seven hours later. Wounds excised; fractured second and third ribs resected. Laceration in lung repaired. Large hemothorax evacuated. Closure. Blood transfusion. Died in one hour.


Earlier operation would have been effective. Other risks quite as forbidding had recovered.

35. H. H. September 28, 1918: Bullet, through-and-through, both wounds sucking, left chest. Condition poor. Duration 36 hours. Operation: Resection sixth, seventh, and eighth ribs. Excision of splenized ling; suture of hole in pericardium; repair of laceration in diaphragm; evacuation of hemothorax. Closure. Five hundred c. c. gum salt for shock. Death in two hours. Necropsy: Small amount of blood in pericardium. Left lung partially collapsed. Splenization incompletely removed. (Exposure inadequate at operation.)
This man's death due to delay and exposure, as he could have been saved with early operation. Another example of the necessity to secure adequate exposure.

36. F. McC. September 29, 1918: Shell fragment, through-and-through, sucking wound, right chest. Entrance below sixth rib, anterior axillary line; exit over eleventh rib below scapula. Condition bad-cold and shocked. Duration 10 hours. Operation: Three hours later. Entrance wound not treated. Exit wound excised; fractured rib resected. Lower lobe lacerated; contained indriven rib fragments and was bleeding profusely. Fragments removed; sutured; chest closed tight. October 2, 1918: Wound opened spontaneously; discharged 300 c. c. turbid fluid containing streptococci. October 5, 1918: Dyspneic, cyanotic, delirious. Contralateral pneumonia. October 6, 1918: Died. Necropsy: Purulent bronchitis, right; fibrinouss pleuritis, right; collapse and splenization of lung, right; massive bronchopneumonia, left; pericarditis; vegetative endocarditis; infarction of kidney.

Man's condition prevented extensive operation. The small chance there was was forfeited by failure to drain.

37. J. A. L. September 30, 1918: Shell fragment, penetrating wound, right chest. Entrance, seventh rib, posterior axillary line. Large hemothorax. Condition grave. Duration 55 hours. Operation: Five hours later. Wound excised. Perforated rib resected. Hemothorax evacuated. Much bile present. Foreign body in liver. Thick fibrinous pleuritic exudate. Liver tear repaired. Diaphragm sutured. Man too low to stand further operation. Peritonitis present. October 5, 1918: Homolateral bronchopneumonia. October 6, 1918: Died. Necropsy: One thousand three hundred c. c. fluid in chest; lung collapsed. Tract of projectile in liver led to large thrombosed vein. Foreign body found, covered with fibrin, Iying between columiiiae cariiete of right ventricle.

Lethal injury at time of operation after 60 hours. Early and more complete operation would have saved him.

38. O. W. September 30, 1918: Bullet, penetrating wound, sucking, right chest. En- trance, seventh interspace, posterior axillary line. Condition bad. Duration 66 hours. Operation four hours later: Entrance wound excised; fractured eighth, ninth, and tenth ribs resected. Pleural cavity cleaned and closed. October 4, 1918: Incision opened spontaneously. October 5, 1918: Died. Necropsy: Lower lobe collapsed except where splenized. Empyemia. Wounded lung contained bone fragments.

Lethal injury at time of operation after 70 hours. Illustrated a common error. Operation should be sufficiently radical to give chance for recovery even at risk of death on table Early, complete operation would have saved.

39. C. K. October 14, 1918: Shell fragment, perforating vound, left chest. Entrance, eighth interspace, midaxillary line; exit, eighth interspace, midscapular line. Foreign body under skin. Moderate hernopneunmothorax. Condition good. Duration 8 1/2 hours. Operation: Excision of entrance wound. Fractured ninth rib resected. Bone fragments driven into diaphragm, which was repaired and sutured into defect to close pleura after evacuation of large hemothorax. Foreign body removed. 1921. Records available. Disability less than 10 percent.


Early operation led to prompt recovery.

40. (German soldier.) October 12, 1918: Bullet wounds, right arm, and penetrating, right chest. Entrance over tenth rib, posteriorly. Condition poor. Duration 54 hours. Operation: Amputation arm; gas gangrene. Excision of entrance wound; resection fractured tenth rib. Repair of lacerated diaphragm. Evacuation of hemothorax. Closure. Death within a few hours.
Injury lethal at time of operation.

41. H. C. October 15, 1918: Shell fragment, sucking, penetrating wound; right chest. Entrance, third rib, high in axilla. Condition poor. Duration 28½ hours. Resuscitation 72 hours. Operation: Entrance wound excised; fractured third rib resected. Oldpleuritic adhesions and patient's condition made radical operation impossible. Foreign body in upper lobe not sought. Hole in lung sutured. Gauze drain. October 29, 1918: Died. Necropsy: Abscess in lung and liver.

Rare instance of abscess forming about foreign body. Hole in lung should have been left open. Fear of bronchial fistula is not well founded.

42. W. L. October 15, 1918: Shell fragment (1 by 1.2 cm.), penetrating wound, right chest. Entrance through middle of clavicle; lodgment in right, upper lobe. Condition poor. Duration 22 hours. Operation three hours later: Entrance wound excised; com- minuted fractures of clavicle, first and second ribs resected; fragments removed from lung. Lung repaired. Hemothorax evacuated. Pleural defect closed with muscle. Died during night; cause unknown. No necropsy.

Severe injury made lethal by exposure and delay.

43. L. B. October 15, 1918: Bullet, through-and-through, sucking wound, left chest. Entrance just above left nipple; exit, ninth rib, paravertebral line. Moderate hemothorax. Condition poor. Duration 24 hours. Operation: Entrance wound not treated. Exit wound excised; fractured ninth rib resected; wound in lower lobe sutured; pleura closed; no drainage. October 16, 1918: Died. Cause of death probably shock.

Another moderately severe injury made lethal by exposure and delay.

44. J. K. November 1, 1918: Bullet, through-and-through, sucking (exit) wound, right chest. Entrance in anterior axillary fold; exit above eighth rib, paravertebral space. Small hemopneumothorax. Condition (?). Duration 11 hours. Operation one and one-half hours later: Wound of entrance untreated as it was found to be smooth on inside. Exit wound excised. Fractured eighth rib resected. Liquid and clotted blood removed. Wounds of entrance and exit in lower lobe sutured. Wounds in upper lobe not found. Parietal pleura closed fairly accurately with aid of one wire rib stay. November 16, 1918: Convalescence stormy. Despite aspiration, wound broke down from within with spontaneous discharge of pyothorax. Had asthmatic attacks. Returned to duty in 90 days. Disability (estimated) 10 percent.

Imperfect closure of parietal pleura is always a source of danger. Fortunately, this rupture occurred after adhesions had formed so that collapse was obviated. Primary drainage was indicated.

45. A. R November 2, 1918: Bullet, through-and-through wound, left chest. Entrance, fifth interspace, posterior axillary line; exit, ninth rib, midscapular line. Slight interstitial emphysema about exit wound. Fluoroscope revealed moderate hemopneumothorax; fracture of ninth rib and involvement of left lower lobe. Condition fair. Duration 11 hours. Operation: Exit wound excised; shattered rib resected; pleura opened widely; many rib fragments removed. Splenized and lacerated lower lobe resected and sutured after evacuation of hemothorax. Wound of entrance found on internal examination to be smooth so not disturbed. One wire rib stay. Pleural closure incompleted so reinforced with muscle. November 8, 1918: Uneventful recovery. Wire rib stay removed. November 11, 1918: Evacuated in excellent condition. No further records obtainable. Disability estimated at 10 percent.


Treatment here was good. Illustrates the wisdom of attacking worst wound first and thoroughly, and letting the internal examination determine whether any further operation is needed. Early removal of wire rib stay was beneficial. Resection of injured lung assured recovery. Primary drainage had been safer because pleural closure was inadequate and pyothorax would likely have led to spontaneous opening.

46. F. K. November 2, 1918: Multiple wounds; shell fragments, one, through-and-through, three penetrating right chest; one bullet, penetrating abdomen. Chest woundsucking and emphysematous. Condition poor. Duration unknown. Prolonged resuscita-tion. November 3, 1918: Operation: Exit wound over ninth rib excised and fractured ribresected. Wounds in lower lobe sutured. Two wounds in diaphragm sutured to excluded herniated and wounded liver, in which foreign body was not sought because of patient's condition. Bile-stained hemothorax evacuated. Drained with tube armed with flap valve obtained from gas mask. Fair approximation of pleura. Skin closure not attempted because of empyema. November 4, 1918: Died in spite of attempts at resuscitation. Necropsy: Valve drain had functioned perfectly. Wounded lung inflated notwithstanding pulmonary edema. Bullet found in retroperitoneal tissues, only injury to kidney. Death due to myocardial incompetence.
Severe multiple injuries with cold and exposure made condition lethal by time of admission to hospital. Recovery with early operation possible. Operative treatment good.

47. J. R. A. November 2, 1918: Shell fragment, through-and-through, left chest. Entrance, third interspace, nipple line; exit, seventh rib, posterior axillary line. Wounds dirty. Condition serious. Duration 27 hours. November 3, 1918: Operation: Both wounds excised. Fractured rib at exit resected. Bone fragments in lung. Four injuries repaired. Liquid and clotted hemothorax (800 c.c.) removed. Both pleural reflections hemorrhagic. Entrance wound on inner aspect not examined. Pleural closure satisfactory. Operation hastened and terminated by patient's condition. November 5, 1918: Never regained strength. Seven hundred and fifty c.c. thin, bloody fluid aspirated. Streptcoccus (?). Cyanosis and dyspnea. Died. Necropsy: Left chest contained 500 c.c. thin, bloody fluid. Fibrinous pleurisy. Pericarditis with effusion. Acute dilatation of right heart; pulmonary edema. Fracture of fourth rib at entrance wound.

Wounds, not of themselves lethal, had become so through cold and delay. Recovery was easily attainable with early operation. Drainage should have been employed, but could not have altered, merely postponed, the outcome.

48. J. C. November 2, 1918; Bullet wound, through-and-through, right chest. Entrance, sixth interspace, midaxillary line; exit, twelfth rib, paravertebral line. Moderate hemothorax. Mitral insufficiency. Dirty wounds. Condition poor. Duration 46 ½ hours. Operation: Entrance wound excised; no rib injury. Exit wound excised; fractured eleventh rib resected. Large hemothorax removed. Laceration in diaphragm and upper pole of kidney repaired. Lung inflated and laceration repaired. Pleura closed tight. Gutta-percha drain to kidney and liver. November 7, 1918: Developed jaundice, edema of extremities and pleuritic effusion. Died. Necropsy: Large pleuritic effusion. One laceration in diaphragm had been overlooked. Acute diffuse hepatitis. Liver wounds necrotic. Acute fibrinous pericarditis.

Another reparable injury made fatal by delay. Line far in advance and transportation of wounded almost impossible. Drainage should have been employed.

49. A. D. November 4, 1918: Bullet, through-and-through wound, left chest. Entrance above left clavicle; exit, ninth rib, paravertebral line. Large hematoma at entrance wound. Left radial pulse absent. Condition poor. Duration (?). Operation:


Wound of exit excised and fractured ninth rib resected when patient stopped breathing. Oxygen had given out. Injection of adrenalin into heart and direct massage started cardiac contractions. Incision closed. Death in two and one-half hours. Necropsy: Hematoma at entrance wound and absent radial pulse due to section of subclavian artery.

Positive pressure analgesia too rich in nitrous oxide when oxygen supply failed. Cardiac resuscitation not prompt enough to save central nervous system from fatal degeneration. Man was probably lethally injured.


Operations, called thoracotomies of necessity, were performed upon 49, or approximately 55 percent of the series. Included are many of the most serious wounds, thus treated because more ideal methods were impossible not- withstanding the exposures obtained were inadequate for deep repair.
Fatalities.-There were 22 deaths, mortality rate of 45 percent. It is noteworthy that the mortality rate in the first half of the series is 32 percent and in the second 58 percent despite the fact that after greater experience the later treatments were better. The difference is due to the colder weather, rain, and greater difficulties in transportation.

One-half of the deaths occurred within 24 hours after operation. One (1) was due to acute anemia and could have been avoided with multiple transfusions; one (11) was due to too high positive pressures; one (49) to too high concentration of nitrous oxide (failure of oxygen supply) in the administration of analgesia. Four (6, 12, 14, 46) had received injuries sufficiently serious to cause death even if treated promptly. Operation was performed 10, 56, 5, and 24 hours after injury; average 26 hours. Five (34, 35, 40, 42, 43) had received injuries not severe enough to jeopardize life if promptly relieved, but were rendered lethal by exposure and delay. Operation was performed 30, 36, 54. 25, and 24 hours after injury; average 33 hours.

Two (20, 47) of the half of deaths that occurred more than 24 hours after operation took place within 3 days. One (20), within two days, was lethally injured, the other (47), within three days, had become fatally affected by delay. The balance (7, 8, 31, 36, 38, 39, 41, 48) survived operation from 5 to 14 days. Five developed pyothorax (7, 8, 31, 36, 38), which was due to incomplete opera- tion, e. g., failure to excise splenized lung (38), to close a bronchial fistula (31). the only one in the series, and to institute primary drainage. The average duration before operation was 34 hours. One (37) died because a shell fragment that had not been removed from a liver was transported to the heart and contributed to a fatal septicemia. Another shell fragment not removed from a liver (42) caused a fatal acute hepatitis. A shell fragment in a lung (41) caused a lung abscess because the track was sutured; this was the only lung abscess noted.

Four deaths were caused by obvious surgical errors-a failure to transfuse(1), too high pressures with administration of anesthetic (11), too high concentration of anesthetic (49), and suturing instead of draining the tract of a foreign body in an adherent lung (41).

Five deaths which occurred in the second half (31,36,37,38,48) were operated upon on the average of 43 hours after injury. None were given the benefits of a


complete operation, yet they survived on the average one week. Operation was hurried and unfinished in each instance because condition of the individuals was so poor. Hindsight seems to teach that one or two might have survived, if at the cost of greater immediate risk an opportunity for ultimate recovery had been provided. Mortality chargeable to surgical errors and accidents (limited supply of blood for transfusions, exhaustion of supply of oxygen, use of too great positive pressure in analgesia, closure of tract in lung, failure to complete operations in spite of impending death and to use primary drainage) is 13 percent.

Disabilities.-Late disability ratings are available for 23 of the 25 survivors. Two were zero (5, 25); six less than 10 percent (15, 18, 22, 24, 26, 39); four at 10 percent (2, 21, 44, 45); two at 15 percent (10, 19); two at 20 percent (27, 28); two at 35 percent (4, 16); one at 40 percent (3); 75 percent (17); 80 percent (13); and 100 percent (29). Two ratings are low (25, 27) and five are high (4, 13, 17, 28, 29) as shown by physical examination, fluoroscopy, resistance exercises and estimations of vital capacity.

According to the figures the average disability was 21 per cent. This is higher than the facts would justify, but is accepted to be safe. Only four were returned to duty, 2 in 90 days (15, 39) and 2 in 180 days (2, 24), giving an aver- age of 135 days, which also is too high, but may be accepted as a safe estimate.

Pleuritis remains a constant factor in producing disability, but there is added, because of the increased severity of injuries, greater interference with parietal integrity, notable in multiple rib injuries, more frequent diaphragmatic lacerations, and greater destruction of lung tissue requiring resections. Like- wise more complicating lesions appear--liver, kidney, pericardium, and peritoneum. A disability rating even of 21 percent is not entirely discreditable when it be considered that the average duration before operation was 19 hours and that none of these men received proper after-care. The evil effects of pre-operative delay in the more serious injuries is apparent. Those who recovered with disabilities of 20 percent or less were operated upon in 15 hours on the average; those above 20 percent in 27 hours.

It was noted above in discussing fatalities that more finished operations and more frequent use of drainage would have reduced the mortality rate. The same applies even more directly to reductions in duration and extent of disabilities because sucking wounds are common, soiling of pleura with bile is frequent, and with urine is occasional. The need for better immediate and continued after-care is self-evident.


Thoracotomies of necessity will be performed upon the less severely injured when parietal excisions and limited thoracotomies reveal unexpected lesions that require more radical immediate intervention and upon those so severely injured that parietal and deep repair must be made through one opening. Advantages are the greater rapidity, avoiding making a separate incision and thus not impairing parietal integrity by surgical wounds added to the projectile destruction. Disadvantages are the frequent failures to obtain satisfactory exposure to make proper intrathoracic repair and the temptation


to avoid risks of operative deaths by performing incomplete operations when finished operations are needed to obtain ultimate recoveries.


Operations of this type will inevitably be more frequent than other serious procedures and demand greater consideration. Improved facilities in advance of mobile hospitals and the establishment of a thoracic surgical division would make better methods possible. If the thoracic wounded were provided with proper treatment from front to base, not only could the less severely injured be shunted to hospitals farther toward the rear when the fighting is active, but the more severely injured could be evacuated earlier. This might well provide for two-stage operations that would secure recoveries in types that now seem to be almost hopeless.

The need for more effective prevention and treatment of shock which includes promptness as a first requisite as well as continued and consecutive care under unified control is indisputable. Simplifying and perfecting the technical details appear now to be easy. The combination is desirable and attainable. Returns to active duty were few and delayed.


1. F. P. July 30, 1918: Shell fragment, penetrating wound, left chest. Entrance, left sternoclavicular articulation; foreign body lodged in lower lobe. Large hemothorax. Heart displaced to right. Condition (?). Duration (unknown). Operation: Entrance wound excised; sternum resected; pleural defect closed with muscle. Fourth rib resected. Hemothorax evacuated. Lacerated lung repaired. Hematoma in lung and foreign body not removed. Wire rib stay. Layer closure. No drainage. Postoperative interstitial emphysema from wound of entrance. Moderate pleuritic effusion. Later was operated upon at a base hospital, anterior and posterior drainage for empyema. May 19, 1919: Drained again at Walter Reed Hospital and wire stay removed. November 14, 1919: Discharged. July 28, 1921: Dyspnea and pain on exertion. Deficient expansion, multiple parietal scars, chronic pleuritis, immobile diaphragm, heart displaced to left, cardiac competence fair. Disability 50 per cent. Vital capacity 58 per cent.

This an early experience; battle rush; no assistant. Operation incomplete. Foreign body should have been removed and intrapulmonary hematoma evacuated. This with drainage would have prevented empyema. Proper after-care would have reduced disability at least by half. Rib stay should have been removed early.

2. D. July 31, 1918: Shell fragment, penetrating wound, left chest. Entrance wound over (?) rib, foreign body in upper lobe. Moderate hemothorax. Condition (?).Duration (unknown). Operation: Entrance wound excised; fractured rib resected; pleural defect closed. Thoracotomy at site of election. Hemothorax evacuated. Foreign body removed. Lung repaired. Wire rib stay. Layer closure. No drain. August 3, 1918:Limited pleuritic effusion. Continued improvement. Lost.

Notes too meager for any judgment.

3. A. L. B. August 8, 1918: Shell fragments, left arm, shoulder, leg, penetratingleft chest, which contained two foreign bodies. Entrance over fourth rib, axilla. Largehemothorax. Condition (?). Duration 18 hours. 'Operation: Foreign bodies removedfrom leg, elbow and shoulder. Entrance wound excised; fractured fourth rib resected;thoracotomy at site of election. Large hemothorax evacuated. Lacerations in upper lobe


repaired. No note of foreign bodies. Two wire rib stays failed to provide satisfactory closure at anterior angle. No drainage. August 10, 1918: Much interstitial emphysema from incomplete closure. August 13, 1918: Chill. Increased pleuritic effusion; 650 c.c. blood-stained fluid aspirated. August 16, 1918: Fluid reaccumulating. Aspirated and found to contain much fibrin. Rib resection and open drainage. August 17, 1918: Subsequent progress excellent. Disability for seven months. Ultimate disability 15 percent (estimated).

Primary drainage would have hastened recovery and reduced disability.

4. W. S. August 9, 1918: Shell fragment, perforating, right chest; entrance wound, posterior axillary line at level of angle of scapula; foreign body lodged in mediastinum posterior to aorta. Enormous hemothorax. Condition bad. Duration eight hours. Operation: Thoracotomy at site of election to give immediate opportunity to check bleeding because of severe and increasing acute anemia. Injured azygos vein found and ligated with great difficulty. Death occurred as bleeding laceration in upper lobe was being sutured. Hemostasis imperative but attempted too late even at eight hours.

Transfusions might have been effective. A serious and usually fatal injury.

5. M. B. August 11, 1918: Shell fragment, through-and-through wound, lower left chest. Large hemothorax. Burns to head and face. Both ear drums ruptured from shell explosion. Condition bad. Blood pressures 90/50. Duration six hours. Resuscitation attempted; pulse became barely perceptible. Chance of any recovery lay in immediate hemostasis. Operation: Thoracotomy at site of election to expose bleeding points surely and with least delay. Lacerations in lower lobe sutured. Death from acute dilatation of right heart.

Injuries lethal at time of operation. Questionable if earlier intervention could have succeeded.  

6. L. G. September 12, 1918: Bullet, perforating wound, left chest. Entrance wound, second interspace below middle of clavicle. Foreign body under skin below angle of scapula. Large hemotborax. Free hemoptysis. Heart widely displaced. Interstitial emphysema more prominent posteriorly. Condition (?). Duration unknown. Operation: Entrance wound excised. No rib damage. Closed. Bullet removed. No rib damage. Closure. Thoracotomy at site of election. Hemothorax evacuated. Wounds in upper lobe sutured. Entrance wound surrounded by zone of splenization which was not excised. One wire rib stay. Layer closure. Superficial drain. September 13, 1918: Greater emphysema about entrance and exit wounds because closure had not been effective. Good condition. September 16, 1918: Aspiration of sterile, pleuritic effusion. Continues to improve. Five months later he developed empyema and was treated by open drainage. Notes on present condition not obtainable.

Imperfect closures of entrance and exit wounds was poor surgery. Splenized lung should have been excised. Note the late empyema. Phrenic nerve should have been blocked.

7. J. J. M. September 12, 1918: Shell fragment, penetrating, left chest. Entrance, sixth interspace midaxillary line. Foreign body lodged near heart, moving with each contraction. Moderate hemopneumothorax. Heart slightly displaced. Condition good. Duration five hours. Operation: Six hours later: Entrance wound excised. No rib damage. Pleural defect plugged with muscle. Fourth rib resected, hemothorax evacuated Wounds in upper and lower lobes sutured. Foreign body at hilum, close to pulmonary artery and not removed for fear of injuring vessel. Phrenic nerve blocked with 1 per cent cocaine. September 14, 1918: Interstitial emphysema from wound of entrance. Considerable pleuritic effusion. Diaphragm on left side seen with fluoroscope to be in extraordinarily high position. September 17, 1918: Exceptionally comfortable convalescence but usual


amount of pleuritic effusion. Absorption of fluid not noticeably favored by diaphragmatic paralysis. Diaphragm now crowded into low position by fluid. September 18, 1918: Six hundred and fifty cubic centimeters fluid aspirated, contained cocci in pairs and chains. December 18, 1918: Discharged from service. Disability 15 percent.

Foreign body should have been removed as immediate risks are less than ultimate dangers. Plugging of pleural defects with muscle again unsatisfactory. In this instance the visceral pleura might have been sutured to the parietal at close of operation. Comforts of phrenic nerve block demonstrated. Effusion due rather to bacterial irritation than to faulty absorption. Primary catheter drainage would have been safer and hastened recovery.

8. H. G. (German soldier). September 13, 1918: Shell fragments, multiple injuries to right chest and arm. Entrance to chest over fourth rib laterally. Large foreign body 0.3 by 3.5 cm. deep in chest. Right diaphragm unaffected. Moderate hemothorax. Subcutaneous emphysema. Condition poor. Duration unknown. Operation: Arm wound excised and foreign bodies removed. Wound of entrance excised. Many foreign bodies, including bits of clothing, removed. Fractured fourth rib resected. Hemothorax evacuated. Upper lobe not exposed because of old adhesions. Bleeding had ceased. No note of removal of foreign body. Collasped lower lobe was inflated nearly to normal. Good closure obtained with one wire rib stay. Postoperative shock. Systolic pressures were raised from 60 to90 with 600 c. c. gum salt intravenously. September 20, 1918: Evacuated in good condition. Two aspirations and opening of focus of suppuration in superficial wound.

Man's poor condition precluded completed operation. Superficial wound,because of emphysema, should have been drained. Proof that inflatable lung is trustworthy.

9. N. P. (German soldier). September 13, 1918: Shell fragment, through-and-through, left chest. Entrance at angle of scapula; exit, above scapula. Moderate hemothorax. Left diaphragm moves one-half as widely as right. Condition (?). Duration unknown. Operation: Entrance wound excised. Fractured scapula and fourth rib resected. Tunnel wound in upper lobe resected and sutured. Exit wound above scapula resected. Closure with one wire rib stay. September 17, 1918: Uncomfortable. Some-residual pneumothorax and pleuritic effusion. September 21, 1918: Conditions little changed. Evacuated.

Lung repair imperfect else no pneumothorax. Serious injuries warranted primary drainage.

10. G. A. September 14, 1918: Shell fragment, penetrating, left chest. Entrance oversecond rib below middle of clavicle. Foreign body, 3 by 0.7 cm. substernal and deflected with each heart beat. Large hemothorax clouding left chest. Heart displaced to right. Very dyspneic. Condition fair. Duration seven hours. Operation: After five hours' resuscitation: Entrance wound excised. Thorax opened by resecting fourth rib. Hemothorax, 600 c. c., evacuated. Perforations in lower margin of upper lobe repaired. The exit wound in lung revealed by escaping air with positive pressure. Foreign body removed from contact with aorta. Closure with one wire rib stay. Condition good. September 17, 1918:Very uncomfortable. Two hundred centimeters dark bloody fluid aspirated. No bacteria found. September 21, 1918: Small pleuritic effusion persists. Disabled for 180 days. No other information. Estimated disability 15 percent.  

Primary drainage. Phrenic block and exercises would have been wiser.

11. F. J. C. September 18, 1918: Shell fragments (grenade) right hip, through back and chest. Entrance wound not found. Foreign body in right lung, seven cm. in diameter. Pleural cavity obscured. Diaphragm immobile. Heart displaced to left. Condition (?). Duration 12 hours. Operation: Superficial wounds excised and foreign bodies


removed. Entrance wound to chest not found. Thoracotomy at site of election; 2,000 c. c.of bloody fluid evacuated, little clotting. Long, edematous laceration, posterior aspect right lower lobe resected and sutured. Multiple areas of contused lung. Foreign body not found. Pleural closure perfect. One wire rib stay. Preoperative blood pressures, 115/94. Postoperative blood pressures, 100/94. Six hours later systolic pressures suddenly fell to 60. Seven hundred cubic centimeters gum salt elevated pressures to 78/50. In 45 minutes they fell to56/40. Transfused 600 c. c. citrated blood; rose to 88/60, where they remained for two hours, then sudden collapse and death. Necropsy: Repair of both pleura is good. Small area of splenization in lower right. Wound of entrance to chest posteriorly from two injured vertebrae. Small hemothorax. Serofibrinous pleurisy. Death explained neither by anatomic lesion nor operative procedure. Attributed to medullary anemia.

Operation alone could have led to recovery. Multiple and large transfusions required to rehabilitate circulation.

12. J. L. September 27, 1918: Bullet, perforating, right chest. Entrance through back of neck; bullet lying beneath skin over eighth rib lateral to scapula. Condition poor. Duration 53 hours. Operation after five hours' resuscitation: Entrance wound in neck notdisturbed. Incision over eighth rib; evacuation of hematoma; removal of foreign body; resection of fractured rib (eighth). Pleural closure. Thoracotomy at site of election. Large hemothorax evacuated. Splenized gutter wound with bronchial fistula in lower lobe excised. One rib stay. Layer closure. October 4, 1918: Aspirated fluid containing many streptococci. Chest drained. Local anesthesia. October 8, 1918: Evacuated. Condition bad. October 13, 1918: Died. Septicemia. No necropsy.

Severe wound, long duration, bronchial fistula, feeble patient. Should have been drained at operation. No less complete operation would have provided opportunity for recovery.

13. A. B. R. September 30, 1918: Bullet, penetrating, left chest. Entrance at left costovertebral angle. Foreign body lodged beneath left clavicle and moves with each heart beat. Dyspneic though hemothorax is small. Condition fairly good. Duration 24 hours. Operation: Entrance wound excised. Fractured twelfth rib resected. Pleural closure. Thoracotomy at site of election; 500 c. c. bloody fluid aspirated. Two holes in lower lobe and one hole in upper lobe sutured. Bullet removed from left apex. One wire rib stay. Layer closure. No drainage. October 6, 1918: Smooth convalescence. Small effusion. July 10, 1919: Wire rib stay removed. Graduated exercises begun, which were beneficial even at this late date. September 14, 1919: Discharged from service. July 26, 1921: Below normal weight; lacks endurance. Fibrous pleuritis. Obliteration of costo- phrenic sinus. Defective expansion. Disability of 30 percent is low. Vital capacity 71 percent.

Early and continued breathing exercises would have reduced disability. Wire rib stay should have been removed promptly.

14. P. McG. October 10, 1918: Bullet, penetrating, right chest. Entrance outer one-third of right clavicle. Foreign body in liver. Interstitial emphysema. Moderate hemothorax. Heart displaced to left. Three lobes of lung perforated. Condition poor. Duration eight hours. Operation: After three and one-half hours' resuscitation: Entrance wound not touched. Thoracotomy at site of election. Inner aspect of entrance wound smooth. Wound at apex of lung could not be exposed. Three tears in lung repaired. Hemothorax evacuated. Bullet could not be felt in liver. Hole in diaphragm sutured. Layer closure. One wire rib stay. Superficial wound of entrance excised and drained because of emphysema. During convalescence developed empyema which was drained and wire stay removed. February 1, 1919: Drainage ceased. August 11, 1921: Dyspnea after sharp exertion. General condition good. Chronic adhesive pleuritis at right base limits excursions of diaphragm. Disability, 22½ percent. Vital capacity, 96 percent.


Primary drainage indicated, particularly by liver injury. Proper exercises would have reduced duration and extent of disability which is due to pyothorax. Bullet in liver is not harmful.

15. WV. B. November 2, 1918: Bullet wounds, penetrating right arm and through-and-through, right chest. Entrance, sternoclavicular joint; exit, sucking at right anterior axillary fold. Hemothorax slight. Pneumothorax large. Provisional sutures in exit wound showed this to have been a "sucker." Foreign body in right forearm. Operation: Wound in forearm excised and bullet removed. Wound of exit excised; compound fracture of fourth rib resected and chest opened; 60 c. c. fluid blood and no clots evacuated. Focus of splenization at extreme tip of middle lobe and adjacent triangle of upper lobe. No lung operation. (Patient probably in critical condition.) Wire rib stay. Layer closure. Defect in parietal pleura reinforced with muscle. November 17, 1918: Stormy convalescence after good immediate recovery. Finally a pyothorax developed containing anaerobes and streptococci. Rib resection and drainage. Continued improvement. Ultimate result unknown.

Incomplete operation, probably because of patient's condition, not justified. If it were, leaving splenized lung assured pyothorax and demanded primary drainage.

16. R. B. November 5, 1918: Shell fragment, through-and-through wound, right arm, penetrating wound, right chest. Entrance, anterior axillary line over fourth rib. Foreign body (0.8 by 1 cm.) resting on diaphragm. Moderate hemothorax. Cold and severely shocked. Duration unknown. November 6, 1918: Continued hemoptysis. Fair recovery from shock. Blood pressures 110/82. Operation: Entrance wound and fractured fifth rib resected; 120 c. c. liquid and 1,200 c. c. clotted blood removed. Foreign body removed. Two holes in middle lobe sututred. Wound in lower lobe not found because old adhesions prevented exposure. Wound in forearm excised. November 8, 1918: Exploratory aspiration, putrid fluid. Local anesthesia. Rib resection. Insertion of air-tight glass tube armed with check valve. November 16, 1918: Evacuated in good condition. Little pneumothorax. Still draining. No further record.

Wound of long duration, weak patient, chronic pleurisy-all demanded primary drainage, which could have been effective in preventing pyothorax.


Thoracotomies for the repair of intrathoracic lesions were performed at the sites of election, i. e., through the middle portions of the fourth or fifth rib upon 16 patients, 15 percent of the entire series. The injuries were serious enough to require prompt attention and permitted or demanded wide exposure so that repair of all lesions might be possible immediately.
Fatalities.-There were four deaths, a mortality of 25 percent. Three (4.5, 11) of the four deaths occurred within six hours; two (4, 5) during operation. Each was in critical condition, each suffering from acute anemia or exemia, and in each it was essential to secure prompt hemostasis for which thoracotomy of election offers largest opportunities. The 4th (12) lived 16 days, dving of septicemia from pyothorax which primary drainage might have prevented. Mortality chargeable to surgical shortcomings is 6 percent. Operations were performed 8, 6, 12, and 58 hours after injury, average 21 hours. It is noteworthy that the one who lived the longest and who might have recovered is the one (12) whose wound was of the longest duration.
Disabilities.-Late ratings are available for only 6 of the 12 survivors. Three (3, 7, 10) at 15 percent; one (14) at 222 percent; one (13) at 30 percent; and


one (1) at 50 percent, giving an average of 20 percent. Six (2, 6, 8, 9, 15, 16) are not estimated. Two were returned to duty (10, 3) in 180 and 210 days, average 195 days' duration of convalescence. The average late disability (20 percent) and duration of immediate disability (195 days) are obtained from insufficient data to be reliable but as both are high they are accepted as safe.

Pleuritis is the chief cause of disabilities as the parietal destructions and lung lesions are less severe than those in Group III and there were fewer complicating injuries. Pleuritis was the cause of immediate distress to all but two (2, 13); six (1, 3, 6, 14, 15, 16) developed pyothorax which had to be drained; four (7, 8, 9, 10) were aspirated one or more times. The average duration of wounds prior to operation was 16 hours for those who recovered and 21 hours for those who died. The wounds in the men who received less than the average disability rating of 20 percent had been of 13 hours' duration; in those above 20 percent, 18 hours; in those who developed empyema, 20 hours. The evil effects of delay are again illustrated; likewise repetitions of indications for primary drainage, notably in duration of injury.

An average disability rate of 20 percent will not reflect upon the nature of thoracotomy of election when it is recognized that 50 percent of the survivors developed pyothorax because of failure to employ primary drainage and that only one (13) received any nonsurgical after-care. Disabilities from pyothorax alone usually exceed 40 percent. Breathing exercises, though delayed for months, benefited one patient materially. Again it is evident in this group that more finished operations and the use of drainage, in brief, more radical treatment, would have reduced both mortality and morbidity rates. Likewise earlier treatment promises better results than later. Blocking the phrenic nerve should have been a routine. It was not done often enough to prove or to disprove its apparent aid to defense. Reduction in discomfort alone warrants its use.


Thoracotomies of election will be performed more frequently as their value is demonstrated by the better results obtained when more radical measures are employed. They can be used most effectively in the following varieties of wounds: When immediate hemostasis is required, the lesion deep-seated and a generous exposure of all possible sources of hemorrhage must be obtained by the shortest and surest approach; when the nature and location of injury make a thoracotomy of necessity coincide with the site of a thoractomy of election: when the exposures obtained by parietal excisions or attempted limited thoracotomies, and the condition of the patient warrant employment of the most adequate means to effect deep repair. The advantages of being enabled to examine the entire side of a chest and to effect all repair possible in intrathoracic procedures are sufficient to compensate for certain disadvantages. These are the occasional occurrence of chronic pleuritic adhesions sufficient to make pulmonary mobilization impracticable, the additional impairment of parietal integrity and more prolonged operations. Pleuritic adhesions occur so infrequently in soldiers as to be an almost negligible handicap. A few observations indicate that the parietal injuries occasioned by thoracotomy of election performed even with rib resection, if properly treated subsequently, will not


add appreciably to disability. Moreover, it is probable that the provision of suitable primary drainage with consequently lessened dangers of pyothorax will justify intercostal thoracotomy unless rib fractures are present. Important as the time factor may be in operating upon seriously distressed patients, the other factors of better work done with less added handicaps of limited exposure often obtained with harmful traction make slightly more prolonged operations the safer.


Thoracotomies of election will prove more serviceable now that principles are better understood, means for primary drainage are at hand, and technical abilities in intrathoracic operations are developing widely and rapidly. Indeed, it is reasonable to predict that improvement in operative procedures coupled with more competent care of the wounded will show that reductions in both mortality and morbidity rates will be such that even the less seriously wounded, having more than limited hemothorax, will be better served by thoracotomy. Thoracotomy of election would be the method

Returns to active duty were few and long delayed. There was no suitable after-care and the need for replacements never became acute in our Army. It is perhaps idle to guess what might have been accomplished. Few, indeed, of the wounded, who will be fit again for duty, will require more than 90 days for recovery, which must be within 10 percent or 15 percent of normal. Smooth healing and suitable after-care would make such recoveries from thoracotomy of election the rule rather than the exception.


It is desirable to learn from all failures, ignorance of proper methods, lack of facilities for ideal care, faulty organization and administration and the like, the remedies that will assure better protection of the wounded hereafter.


The following table correlates the essential facts.


 A mortality rate of 37 percent will appear to be woefully high until it benoted that 53 percent of the deaths resulted from injuries initially lethal or made so by delay, hemorrhage, infection and exposure. The 18 remaining deaths are charged to therapeutic failures which include not only operative mistakes but also lack of adequate supplies of blood for transfusion, of hypertonic


glucose solutions, and insulin, failure to employ the intravenous administration of digitalis, and proper immediate and remote after-care. In other words 47 percent of the deaths might have been reduced had materials and method snow available been used. Operations have been held entirely responsible for12 deaths (10 percent for the series and 30 percent of the fatalities) inasmuch as this number survived long enough to die from complications which more exact methods could have obviated. The incompleted operations were not finished because the patients were in such critical condition that continuation appeared to be unwise, yet survival for several days is assumed to prove such conclusions to have been erroneous.

Criticism will be directed at the large proportion (53 percent) classed as lethally affected who were subjected to operation. In some of these patients the fatal involvement was not recognized and could not have been recognized until exposures obtained by operation revealed irrecoverable lesions. Other men apparently equally or even more severely handicapped made excellent recoveries. Errors in judgment were numerous and because decisions must be made promptly without full information they will always be frequent.

The orders of the chief consultant, surgical services, A. E. F., to operate on every man who had the least chance for recovery, were proved to have been based upon sound surgical judgment as well as upon a high conception of obligation. A few of the men who died during or soon after operation conceivably might have survived without operation or had the operation been less exact. There was no proof of this. In several instances, postoperative blood trans- fusions would have averted death, but this was too often impossible because donors were not available. As experience grew and methods became more radical, the results improved. In retrospect, our regrets are that supposed conservatism allowed us to evade the responsibility of performing forbidding but necessary operations and thus to deny some men their only chance to survive.


Interpretation of the causes of disability is most important. It will determine the chief factors in producing death, the greatest disability, as well as show how more complete recoveries could have been hastened.


Disabilities due in part to irreparable defects include destruction of parietes, costal and diaphragmatic, or such costal lesions as required the use of the diaphragm in effecting closure, cardiac lesions and injuries ordinarily remediable but not repaired because conditions prevented finished operations. These and the following estimates are based upon the results actually obtained and not upon what might have been done in the light of present knowledge. Therapeutic failures consist mainly of omissions of postoperative aspirations and transfusions which would have favored defense and repair. They are by no means to be charged to ignorance and neglect since it was frequently impossible to give desirable individual attention and donors were often not to be had when most needed. No account has been taken of the failure to provide suitable exercises to aid in reestablishing the conditions for normal respiration. There is record of but one instance (Group IV , 3) in which breathing exercises were used and then 10 months after operation. Some of the men possibly continued the training advised, and, when possible, instituted before evacuation after operation.

Operative failures, with a few exceptions, were omissions. Incomplete procedures and neglect of drainage were the worst faults to which probably can be added the failure to introduce two-stage operations.

The worst showing is in the extended period of immediate disability (130 days). It shows that recoveries were more protracted and less complete than they should have been. On the other hand the permanent disability ratings, an average of 16 per cent for the series which errs on the higher side, is proof of what was accomplished.

In order that the full significance of the ultimate disability ratings and the reasons therefor are appreciated three facts are emphasized: (1) An average disability of 16 percent was obtained in 66 men; 63 percent of them had been subjected to major thoracotomies, some with multiple rib resections, some with resections of part or most of a lobe and in some the diaphragm was used to secure parietal closure. (2) An understanding of what this means will be easy when it is known that the average disability rating of soldiers treated in the camps in the United States for post-pneumonic empyema was reported as being approximately 50 percent. (3) The results obtained in treating the wounded can not be attributed to unusual operative skill or to especially effective after-care, but only to the employment of methods which had been determined by intrathoracic physiologic actions and reactions.


The principal secondary influences that determine recoverability from chest wounds are (a) the condition of men when wounded, (b) the promptness with which relief is provided, and (c) its effectiveness.

Fighting men can not always be well fed and watered or kept free fromfatigue and the mental depression inevitable with reverses, all of which have untoward effects upon the wounded. Our Army experienced no continued and but few localized defeats, so this factor was eliminated. Cheerfulness was an almost unbroken rule.


Promptness depends upon getting the wounded back to hospitals and in keeping hospitals as close to zones of conflict as relative safety permits. Transportation is essential to both. Transportation handicaps, however avoidable or unavoidable, caused great hardships to the wounded and materially increased mortality and morbidity rates.

Early wound dressing was very good. Sucking wounds should seldom be sutured provisionally. Firm pressures swathes to provide support and fixation would have been helpful. Morphine should be given in full doses and repeated. Prophylaxis and treatment of shock can be improved. In addition to means to get men warm and dry, provisions are needed for more general employment of intravenous remedies, particularly in advance of mobile hospitals. A satisfactory method of preserving blood, based upon the work of Weil,49 will provide for a limited number of transfusions. In addition some hyper-tonic carbohydrate solution, insulin and digitalis now seem to be essential and should be used in spite of the less dangers of inducing secondary hemorrhage to combat certain death from prolonged hypotension.

The need for further preoperative care, including physical and fluoroscopic examinations, and of consecutive after-care, providing similar examinations, early and progressive activation has been established.

The efficacy of surgical procedures based upon physiological principles in providing opportunities for recovery with the least danger, delay and disability has been proved and warrants positive statements.

All chest wounds, severe enough to demand more than temporary dressings, should be treated by operations conducted under positive pressure gas analgesia.

Parietal excisions, usually with paracentesis, suffice to protect the least seriously injured and to hasten recoveries without added danger. They may likely prove a first-stage operation in the care of those so seriously wounded that any additional primary intervention except perhaps aspiration or the insertion of an intercostal drain is prohibited.

Limited thoracotomy is applicable in the treatment of the less severely wounded in whom deep repair can be effected through slight enlargements of entrance and exit wounds and from whom the bulk of hemothorax can be aspirated by large cannulte introduced through a defect in parietal pleura.

Limited thoracotomy can be employed as a first-stage operation when deep injuries require an amount of repair that could not then be tolerated. Under such conditions, hemostasis and primary drainage would be required.

Thoracotomy of necessity, a more extensive application of limited thoracotomy, must be used when immediate deep repair is required and when there are reasons for using the wound defect in the parietes to secure exposure. If it be found that the repair required can not be effected with obtainable exposure or if the patient's condition prevents satisfactory operation, hemostasis, primary drainage and parietal closure can be secured and the complete operation postponed, to be performed through a separate incision.

Thoracotomies of election will find wider application as primary procedures, and if two-stage procedures are practicable, they will be the preferable secondary


operations. They afford sufficient exposure for a satisfactory examination of a pleural cavity, permit repair of most visceral and diaphragmatic lesions, heal well and impose little disability. Indeed, with the development of simple and effective methods of primary drainage thoracotomny of election may prove to be the safest treatment of massive pneumothorax unless the patient's condition be poor.

The value of two-stage operations, should they prove feasible, will be greater than in affording better care. A good proportion could be sent, after the primary operation, to evacuation hospitals where the final operation would be performed. Thus the mobile hospitals could be relieved for more strictly emergency service.

We have attempted to indicate our therapeutic failures so positively that similar mistakes need not be repeated. Contributory causes for failure will be mentioned for the same reasons, namely, so that repetitions may be avoided.

Success can be realized if personnel competent to give treatment and the materials essential to that treatment are so organized and disposed that the wounded may be well and promptly served. From a medical viewpoint, service to the wounded men is all important; from a military standpoint, service to the fighting men is most important. The exaggerated individualism of civil surgeons lead them to misunderstand or to fail to appreciate the responsibilities of their colleagues in the Regular Service to Army organization and administration. Similarly, the medical officers of the regular service apparently underrated the personal aspirations of doctors and nurses to provide their patients with the best care. Both sides were unprepared to fulfill their obligations and neither hesitated to hold the other responsible, justly and unjustly, for their own deficiencies. Jealousy, suspicion, distrust, and resentment prevented cooperation. The wounded man paid the bill with avoidably high mortality and disability rates.

There was a rapidly progressing improvement in all departments which showed that had the war continued solutions might have been found for even the most involved problems. Thereby was indicated a means to preparedness.

An army must be an autocratic organization, but many evils peculiar to autocracies can be minimized. This can be accomplished effectively so far as the Medical Corps of the United States Army is concerned by developing cooperation in advance. Civil surgeons can and should prepare themselves not merely to give professional services but to give them under the restrictions of military methods, the worst attribute of which is inflexibility. Similarly, the officers in the Regular Medical Corps, kept directly in contact with progress and changing requirements of surgical practice, will find more liberal interpretations of regulations, which are fixed products of past experiences and often are literally opposed to immediate necessities. National security should be enough of an incentive to produce the necessary personal adaptation and coordination of effort.



(1) Burlingame, C. C., Lieut. Col., M. C.: Military History of the American Red Cross in France. On file, Historical Division, S. G. O.
(2) Cannon, W. B., Licut. Col., M. C., and Yates, J. L., Lieut. Col., M. C.: Report on the "Services of the Laboratory of Surgical Research, American Army at Dijon," December 7, 1918. On file, Historical Division, S. G. O.
(3) Yates, J. L.: The Significance of Vital Capacity in Intrathoracic Therapy. Archives of Surgery, Chicago, 1925, x, No. 1, pt. 2, 477.
(4) Miller, William Snow: Key Points in Lung Structure. Radiology, 1925, iv, No. 3, 173.
(5) Graham, E. A., and Bell, R. D.: Open Pneumothorax: Its Relation to the Treatment of Empyema. American Journal of the Medical Sciences, Philadelphia, 1918, n. s. civil (War Medicine), 839.
(6) Barcroft. Personal communication.
(7) Bowditch, Henry L.: On Pleuritic Effusions, and the Necessity of Paracentesis for their Removal. American Journal of the Medical Sciences, Philadelphia, 1852, xxii, n. s., April, 320.
(8) Pryor, John H.: Immobility of the Diaphragm Following Pleural Exudates. New York Medical Journal, April 22, 1916, ciii, 781.
(9) Middleton, W. S.: Costodiaphragmatic Adhesions and their Influence on the Respiratory Function. American Journal of the Medical Sciences, Philadelphia, 1923, clxvi, 222.
(10) Keller, William L.: The Treatment of Chronic Empyenia where the Recognized Surgical Procedures have Failed to Produce Obliteration. Annals of Surgery, Philadelphia, 1922, lxxvi, No. 5, 549. Ibid., No. 6, 700.
(11) Gwathmey, James T.: Anesthesia. The Macmillan Company, New York, 1924, 2d ed., 136.
(12) Steiner. Personal communication.
(13) Cloetta: Über die Zirkulation in der Lunge und deren Beinflussung durch Über und Unterdruck. Archiv f ür experimentelle Pathologieund Pharmakologie, Leipzig, 1911, lxvi, Nos. 5-6, December 20, 409.
(14) Karsner, H. T., and Ghoreyeb, A. A.: Studies in Infarction. III. The Circulation in Experimental Pulmonary Embolism. Journal of Experimental Medicine, New York, November 1, 1913, xviii, 507.
(15) Denny, George P., and Minot, George R.: The Coagulation of Blood in the Pleura Cavity. American Journal of Physiology, Baltimore, 1916, xxxix, No. 4, 455.
(16) Elliott, T. R., and Henry, H. G. M.: Infection of Hemothorax by Anaerobic Gas-Producing Bacilli. British Medical Journal, London, March 31, 1917, 413; April 1, 448.
(17) Delrez, L.: War Wounds of the Joints. United States Naval Medical Bulletin, Washington, 1920, xiv, No. 4, 537.
(18) Yates, J. L., Middleton, W. S., Drane, Robert, and Gwathmey, James T.: Laboratory of Surgical Research, Central Medical Department Laboratory, American Expeditionary Forces, France, A. P. O., No. 721. Boston Medical and Surgical Journal, Boston, 1919, clxxx, 405.
(19) Bradford, Sir John Rose: Gunshot Wounds of the Chest as Seen in Hospitals on the Lines of Communication. War Medicine, Paris, 1918, ii, No. 1, 10.
(20) Soltau, A. B. and Alexander, J. B. On Gunshot Wounds of the Chest as seen at a Base Hospital at Nancy. Quarterly Journal of Medicine, Oxford, 1916-17, x, July, 1917, 259.
(21) Karsner, H. T., and Ash, J. E.: Studies in Infarction. It. Experimental Bland Infarction of the Lung. Journal of Medical Research, Boston, 1912, xxii, n. s., No. 2, 205.
(22) Willems, C.: La mobilisation active immédiate. Méthode générale de traitement des lésions articulaires. Archives médicales Belges, Paris, 1918, lxxi, No. 3, March, 225. Ibid., Traitement de L'arthrite purulente par l'arthrotomie simple suivi de mobilisation active immédiate. Bulletins et memoires de la société de chirurgie de Paris, June 19, 1918, xliv, 1098.
(23) Lockwood, A. L., and Nixon, J. A.: War Surgery of the Chest. I. British Medical Journal, London, January 26, 1918, i, 105; II. Idem., February 2, 1918, i, 145.
(24) Duval, Pierre: Les plaies de guerre du poumon. English translation by Col. Thompson, Masson et Cie., Paris, 1917.
(25) Gask, G. E., and Wilkinson, K. D.: Penetrating Gunshot Wounds of the Chest and their Treatment. British Medical Journal, London, December 15, 1917, ii, 781.
(26) Gask, G. E.: Gunshot Wounds of the Chest. Transactions of the College of Physicians, Philadelphia, November 5, 1918, xl, 3 s., 199.
(27) Bastianelli, R.: Treatment of Chest Wounds. Surgery, Gynecology and Obstetrics, Chicago, 1919, xxviii, No. 1, 5.
(28) Davies, H. M.: The Surgical Treatment of Gunshot Wounds of the Chest. Lancet, London, January 29, 1916, I, 232.
(29) Bradford, Sir John Rose: On Gunshot Wounds of the Chest. British Medical Journal, London, August 4, 1917, ii, 141.
(30) Tuffier, T.: The Secondary Surgical Treatment of Chest Wounds. War Medicine, Paris, 1918, ii, No. 1, 16.
(31) Leslie, R. M.: The Medical Aspects of Chest Injuries. Practitioner, London, 1916, xcvi, March, 301.
(32) Soltau, P. B.: Wounds of the Chest. War Medicine, Paris, 1918, ii, No. 1, 1.
(33) Cannon, W. B.: Traumatic Shock. War Medicine, Paris, 1918, ii, No. 6, 1367.
(34) Depage, A.: General Considerations as to the treatment of War Wounds. Transactions of the American Surgical Association, Philadelphia, 1919, xxxvii, 15.
(35) Cowell, E. M.: Plastic Transcostal Thoracotomy. British Medical Journal, London, November 3, 1917, ii, 581.
(36) Cloetta: Beitr age zur Physiologie und Pathologie der Lungencirculation und deren Bedeutung für die intrathoracale Chirurgie. Archiv für klinische Chirurgie, Berlin, 1912, xcviii, No. 3, 835.
(37) Capps, J. A.: Clinical Study of Pain Arising from Diaphragmatic Pleurisy and Subphrenic Inflammation. American Journal of the Medical Sciences, Philadelphia, 1916, cli, No. 3,333.
(38) Gray, H. M. W.: Notes on Surgery of the Chest. British Medical Journal, London, November 3, 1917, ii, 580.
(39) Dobson, J. F.: A Preliminary Note on the Treatment of Infected Hemothorax. British Medical Journal, London, February 2, 1918, 148.
(40) Mozingo, A. E.: The Surgical Treatment of Empyema by a Closed Method. Journal of the American Medical Association, Chicago, December 21, 1918, lxxi, 2062.
(41) Blankenhorn, M. A.: A Closed System of Drainage for Penetrating Wounds of the Chest. Journal of the American Medical Association, Chicago, December 14, 1918, lxxi, 1994.
(42) Whittemore, W.: A Series of 100 Consecutive Empyemata. Boston Medical and Surgical Journal, Boston, November 13, 1919, clxxxi, 575.
(43) McKenna, H.: Operation for Empyema. Journal of the American Medical Association, Chicago, August 31, 1918, lxxi, 743.
(44) Roberts, J. E. H., and Craig, J. G.: The Surgical Treatment of Severe War Wounds of the Chest. British Medical Journal, London, November 3, 1917, ii, 577.
(45) Petit de la Villeon, E.: Extraction of Projectiles from the Pleura and Diaphragm. War Medicine, Paris, 1918, ii, No. 1, 24.
(46) Roux-Berger, J. L.: Plaies de la pleure et du poumon par projectiles de guerre. Lyon chirurgical, 1918, xv, No. 1, 1.
(47) Rist, Flandrin, Bernard, Somerville, et al. Discussion on "Do we Expect to find Pulmonary Tuberculosis Following Gas or Thoracic Wounds?" War Medicine, Paris, 1919, ii, No. 6, 982.
(48) Péhu and Daguet: Recherches Cliniques et radioscopiques sur certaines séquelles lointaines des plaies pleuro-pulmonaires de guerre. Lyon chirurgical, 1918, xv, No. 3, 291.
(49) Weil, Richard: Sodium Citrate in the Transfusion of Blood. Journal of the American Medical Association, Chicago, January 30, 1915, lxiv, 425.