Contents

443

SECTION I

GENERAL SURGERY

CHAPTER XV

WOUNDS OF THE ABDOMEN
            
The man with an abdominal wound presents one of the serious problems which the military surgeon has to face. No other group of cases furnished anything comparable to it in testing the medical resources of an army or the technical skill of its surgeons. Although gunshot wounds of the abdomen comprise a small but indeterminate percentage of the total wounded (in the American Expeditionary Forces the relative frequency was 3.3 percent of those admitted to hospital, no account being taken of the killed in action) ¹ the severity of the lesions encountered and their complex nature call for the highest surgical judgment in diagnosis and treatment. The problem is full of interest to every surgeon and challenges his best thinking.

In the war of 1914-1918, the attitude of surgeons with reference to abdominal wounds underwent a marked change. In the Spanish-American War and the Boer War the expectant treatment was followed,² and few surgeons then had the courage to argue for and practice. operative interference in these cases. The occasional patient on whom operations had been attempted almost in-variably succumbed, chiefly because intervention had been too long delayed. This inevitably resulted from fighting over an open country with a rapidly shifting line, for properly equipped hospitals could not be placed near enough to the firing line to be of any service to the seriously wounded soldier.

During the early months of the great war, conservative management of the wounded abdomen was still generally advised and practiced. This was due to the open nature of the warfare and the impracticability of establishing well equipped surgical hospitals sufficiently near the front line. On the other hand, early in 1915, after the battle lines had become fixed, there began to be evident a movement, both among the surgeons of the Allies and German surgeons, for surgical intervention, and this effort gained rapidly in favor.

The reasons for this change in surgical attitude lay largely in the growing appreciation of the truth of three factors: First, that the man with a wound of the abdomen, nonoperated, usually dies. Gibbon ³ stated that in 19 months' active service overseas he did not see a single instance of recovery following nonintervention in penetrating or perforating abdominal wounds. Second, that the time factor is vitally important because of the frequent presence of serious hemorrhage, and, with a hollow viscus injury, especially the small intestine, the rapid development of spreading peritonitis. Third, that the intra- abdominal wounds in the war just ended were definitely more grave than in preceding wars because of the nature of the projectiles used. Machine-gun bullets and high-explosive shell fragments in particular often caused extensive lacerations of the bowel wall, sometimes completely severing it, and at other times excavating large areas. These lesions were in marked contrast to the clean-cut, punctured rifle wounds of former conflicts.

444

Before proceeding with a detailed study of the wounds themselves, it is important and essential that consideration be given to certain subjects which are vital in the proper care of the abdominally wounded.
 
SPECIAL PROVISIONS FOR THE CARE OF ABDOMINALLY WOUNDED

TRANSPORTATION AND THE TIME FACTOR

It becomes evident at once that, whether with a fixed or moving line, the abdominally wounded man must be rapidly evacuated to the hospital where surgical treatment may be applied. Transportation, therefore, becomes an exceedingly vital problem. The motor ambulance made possible much of the advance in abdominal surgery by cutting down materially the time elapsing between the receipt of the injury and operation.

Various factors, however, interfere with rapid evacuation, such as the severity of the fighting, the number of wounded men, the mobility or immobility of the line, the terrain, and the condition of the roads as to number, surface, and traffic requirements. Wallace I found that in a study of 1,200 cases of gunshot wounds of the abdomen most cases arrived at a casualty clearing within the first six hours. Only those evacuations accomplished within the first six hours were considered good; after that time the chances were against the patient.

 Ambulances should be provided with heating facilities, such as the British used, that patients may be evacuated in cold weather, warmed and, therefore, less shocked and with better chances of recovery following operation. The writer took many a dead man from an ambulance in which the other occupants were stiff and cold, with the firm conviction that some of those lost might have been saved had the ambulance been provided with a heating appliance.

At times horse-drawn ambulances have demonstrated their usefulness when with roads absolutely blocked and motor evacuations at a standstill the horse-drawn vehicle has been able to carry patients over an open country and bring the seriously wounded where surgical aid was available. A certain number of such ambulances should always be included in a divisional organization, that such special evacuation emergencies may be successfully met.

A well-organized front line evacuation is essential to the proper surgical care of the abdominal case. This effort is difficult enough when trench warfare exists, but during an advance it requires the best in organization and heroism in order to insure the least possible delay. The most capable surgeon and the best equipped hospital are of no avail if the wounded do not reach the hospital within 12 hours.

The time factor is recognized as an all-important element. In general, the patient must be seen in the first 12 hours if a wounded hollow viscus is to be successfully treated surgically. Those operated upon within the first eight hours yield definitely better results. No other consideration compares in importance to the element of time; recovery percentages are inversely proportional to the time factor rather than to the organ involved or the number of lesions encountered.
 

445
 
THE NONTRANSPORTABLE HOSPITAL

Special provision had to be made far forward where abdominal cases could receive early and adequate surgical care. With a fixed line small hospitals naturally came into being, placed 6 or 8 kms. from the firing line. They were well housed, usually in wooden huts or chateaux, with completely equipped operating and radiological units and well organized wards. A surgeon of ability was placed in charge of such a unit and the staff requirements, including a small nursing quota, were limited. The French and the Belgians made this type of hospital an integral part of their organization, and as time went oneach sector of the line had its advanced hospital for the transportable wounded to care for men with wounds of the abdomen especially, of the chest, and for those suffering from shock or hemorrhage.

THE MOBILE UNIT

No consideration of abdominal military surgery can be complete without taking into account the necessity for well-organized and equipped mobile units for the care of the seriously wounded, well up toward the front. When the war of movement began in the spring of 1918 a different type from the stationary nontransportable hospital had to be evolved, so organized that it could move with an advancing or retreating line and still provide satisfactory surgical care for the severely wounded. As told in Chapter V of Volume VIII of this history, the French had devised this type of advanced hospital, known as the auto-chir, and, likewise, a lighter form, the groupe complementaire, and had made use of them on numerous occasions. The American Army adopted both these units, calling them the mobile hospital and the mobile surgical unit, respectively.⁵ They were controlled by a capable surgeon with adequate assistants; the mobile hospital had trained nurses. The bed capacity of the mobile hospital was 120; upon its ultimate standardization, 20 3-ton trucks were required for its transportation.⁶

Such a hospital should be placed as far forward as is compatible with reasonable safety to patients and personnel, but no farther. It is very difficult to maintain morale and surgical efficiency when such a unit is under shell fire. Under these conditions the patients become terrified, it is exceedingly difficult to carry on any effective work, and a hasty evacuation of all patients may be imperative. Patients with abdominal wounds who had done well after operation in such a mobile hospital have been known to die following their evacuation to the rear.

It has been claimed that proper postoperative care could not be provided by the mobile unit if an offensive with an advancing line were in progress, as military necessity would compel the moving forward of the hospital and the evacuation of all postoperative cases. The organization of such a unit may include a rear echelon consisting of a certain number of officers and enlisted men who may be left behind with a portion of the tentage and supplies to complete the 8 or 10 days of postoperative care required for abdominal cases. When such patients finally are evacuated the echelon moves forward to join the major portion of the unit. Another effective way of dealing with this problem consists

446
 
in making use of a second mobile unit, which passes the hospital already established and establishes itself in a more advanced position, while the unit behind completes the postoperative care of the nontransportable cases.
 
INCIDENCE

The true incidence of gunshot wounds of the abdomen among the members of the American Expeditionary Forces is not known nor can it ever be; many men so wounded died upon the battle field from shock and from hemorrhage. However, this cause of a discrepancy in our statistical data is counterbalanced somewhat by the fact that many men, otherwise wounded than in the abdomen, also died on the battle field. Since diagnosis tags were used by medical personnel in the trenches and at battalion aid stations, and since these tags eventually became more detailed medical records, thus including men with severe abdominal wounds but who perhaps died while being evacuated to a hospital, it is reasonable to conclude that the relative incidence of wounds of the abdomen may readily be determined from the records so made. Thus, of the 147,651 men wounded by gunshot missiles, 5,631 were wounded in the abdomen (including the pelvis),¹ a relative frequency of 3.3 percent. Considering the incidence from another viewpoint: Many of the men suffered multiple wounds, thus making the total number of wounds 170,841, and the relative frequency of wounds of the abdomen among these 3.8 percent.¹
 
MILITARY IMPORTANCE

At first thought one would say at once that from the military standpoint the soldier with an abdominal wound is much less important than the man slightly wounded. The latter may reasonably be expected to return to the firing line at an early date while the former may be months in the rear or may never return to the front. Further, the small proportion of intra abdominal lesions encountered would seem to make them of decidedly less military importance. But, as a matter of fact, the proper care of abdominally wounded men is exceedingly important from the side of the morale of the Army, for it gives its soldiers the conviction that any man badly wounded will receive the best chance for his life that surgery can give him. Humanity dictates that every combatant seriously hit by enemy fire shall have a chance to live.
 
NONPENETRATING WOUNDS

INVOLVING THE ABDOMINAL WALL

These include gutter wounds; perforating wounds running tangential to the peritoneum; wounds, with the missile lodged in some portion of the anterior or lateral abdominal wall or posteriorly in the retroperitoneal tissues or muscles of the back. Any type of projectile may cause such a wound.

The gutter abrasions make their own diagnosis. They call for complete d ebridement, with primary closure if the case can be held under the operator's observation for at least one week, during which time any symptoms of wound infection may quickly be recognized and promptly met. In times of great
 

447
 
activity, where a possibility exists of an early evacuation of the case, the wound, after débridernent, should be left widely open and treated with Carrel-Dakin dressings, a delayed primary or a secondary suture being done at a later date. It is well to remember that some apparent gutter wounds are really penetrating wounds, the gutter in these cases growing deeper as the course of the projectile is followed. When any doubt exists, X-ray diagnostic aid must be made use of.

With a perforating wound of the abdominal wall the shorter the tract the less likely is one to encounter a visceral involvement. With a through-and-through wound the missile may change its direction, after entering the soft tissues, without any impact with bone, and travel along the abdominal wall, emerging at some distance from the point of entrance. Such a case may present great difficulties in diagnosis, for the two most important local signs of a visceral lesion-muscular rigidity and tenderness-are marked, as a result of muscle traumatism, even when no abdominal perforation has occurred. The patient's general condition, with lack of pulse elevation and gastrointestinal symptoms, is the important guide to a correct diagnosis. It is well for the surgeon to bear in mind that in visceral perforations seen very early general signs may be absent. The X-ray examination is of little aid in the differentiation. With clean-cut rifle wounds no intervention is necessary, but if definite doubt of abdominal perforation exists it is necessary to operate without delay. Wounds caused by shell or grenade fragments or machine-gun bullets should be completely dissected and the rule for or against primary closure outlined for gutter wounds should be followed.

The determination of the true nature of wounds of the parietal wall, with entrance only, is at times as difficult as that when both entrance and exit exist. The same local signs frequently are present and the differentiation from cases of peritoneal penetration must be made in a similar manner. The radiological findings, however, usually make the diagnosis for or against peritoneal penetration, and too much stress can not be placed upon the necessity for this type of examination. Dissection of the tract with removal of the foreign body is the course to be followed. The rules for suture are the same as those outlined above.
 
SUBCUTANEOUS RUPTURE OF VISCERA

The literature of civil surgery and of former wars has yielded many examples of this sort of abdominal lesion; that concerning the recent conflict has been equally prolific. In the Medical and Surgical History of the War of the Rebellion numerous types and variations of subcutaneous abdominal injuries are cited in great detail.⁷ In war, as in civil life, injuries of this sort are caused by the kick of a horse or a mule, by blows upon the abdomen during a fight, or by falls from a height, the man landing upon the abdominal wall. In military surgery injuries of this type are caused also by shell explosions in the immediate vicinity of the soldier (the so-called "wind injuries" of the Civil War); they may be caused by flying pieces of wood or other solid objects in connection with shell bursts; they may result from crushing injuries from falling timbers or earth incident to the bursting of a projectile.
 

448
 
Two modes of injury which are especially important because of diagnostic difficulties are laceration of the abdominal wall by a missile without penetration of the cavity but with sufficient explosive force to rupture a viscus within, and visceral perforations from flying bone spicules without peritoneal penetration by a projectile.

The former injury if unrecognized leads frequently to a fatal result; the latter should always be thought of in connection with rib fractures in or about the hepatic or splenic areas or with comminuted pelvic fractures where the pelvic portion of the small intestine is the most frequently injured viscus, while injury of the bladder, rectum, and colon is less common.

The organs most often ruptured are the liver, spleen, kidney, and small intestine; the mesentery is not infrequently torn. The colon, rectum, and bladder are less often injured.

The lesions encountered may be grouped under the headings injuries of hollow viscera and injuries of solid viscera. Hollow visceral lesions may include contusions of the wall, minute perforations, extensive lacerations or even complete division in the case of the bowel or its mesentery. Solid visceral lesions comprise subcapsular rupture of solid organs, especially the liver, with usually a small hematoma beneath the capsule, slight tears, and extensive lacerations.

The diagnosis of subperitoneal rupture of any viscus or of the mesentery must be made by a consideration of the site and mode of injury, the presence of more or less anemia from hemorrhage, especially with solid visceral and mesenteric lesions, and, where a hollow viscus has been opened, the usual signs of peritoneal irritation with subsequent symptoms of a rapidly advancing peritonitis. Patients with serious injuries frequently come to the hospital inconsiderable shock, particularly when the degree of hemorrhage is severe or the small intestine or mesentery has been seriously injured.

Rupture of a hollow viscus calls for immediate intervention; the principles of treatment outlined under penetrating abdominal wounds should be followed. A torn mesentery with signs of hemorrhage requires immediate operation; resection is usually necessary.

With a suspected rupture of a solid viscus but without alarming symptoms or hemorrhage, an expectant attitude should be adopted. Most of these patients recover unoperated. If the hemorrhage has been severe in such a case the best working rule is to operate if the patient is seen very early, but to watch and observe for a few hours if the patient is seen six or eight hours after injury. If then the anemia is not apparently progressive one should not operate, for the hemorrhage has to all intents ceased and laparotomy will cause a renewal of the hemorrhage with a probable fatal result. If there are signs of progressive bleeding the abdomen must be opened at once.

Control of hemorrhage from a ruptured liver may be accomplished by packing or by mattress suture inserted with a large needle, blunt end first. Suture near the diaphragm or far back on the inferior surface of the liver is difficult and frequently impossible. Packing with sterile gauze should be resorted to when suture can not be effectively carried out.
 

449
 
In general a rupture of the spleen should be treated by splenectomy. The objection raised that, convalescence is hampered hy the impairment of the patient's blood forming, mechanism is practically unimportant.

In all cases in which hemorrhage has been a serious symptom transfusion should be carried out as soon as possible after the bleeding has been controlled. The ease with which this procedure may have carried out, either with the aid of paraffin-coated tubes or by the citrate method, the prompt and striking improvement in the wounded following transfusion for hemorrhage, and the abundance of robust individuals among the slightly wounded who may serve as donors, should serve to make this operation a routine procedure early in the treatment of hemorrhage.
 
PENETRATING AND PERFORATING WOUNDS

GENERAL CONSIDERATIONS

The general principles of military surgery of penetrating wounds of the, abdomen differ but little from those employed in civil life. However, many factors obtain to make the problems of abdominal surgery complex and difficult, and each case must be studied in the light of accumulated experience in the war zone.

Before the types of injuries involving special organs are discussed certain general principles influencing the diagnosis and management of abdominal wounds must be considered. As the patient is admitted the questions which arise are: (1) What is the man's general condition with respect to shock? (2) What is the diagnosis? (3) Is operation indicated?

Several factors contribute to bring the soldier with an abdominal wound to the hospital in a condition of shock. These are lack of sleep and food, exposure to cold and wet, difficult evacuation to the hospital, traumatism of abdominal and visceral walls, hemorrhage, pain, and infection, the latter being especially important in those arriving, late. Hemorrhage, save in the late cases, is the great factor in the production of shock, and the degree of shock is usually proportionate to the severity of the hemorrhage; undoubtedly many combatants die on the battle field from rapidly fatal intra abdominal hemorrhage. The abdominal cases reaching the hospital show most extensive hemorrhage when the mesenteric or pelvic vessels or when the liver, spleen, or kidney have been injured.

Hemorrhage may also occur from the abdominal wall, the omentum, and from the retroperitoneal tissues. In the latter case the loss of blood may occasionally be alarming, and the diagnosis very difficult. Since hemorrhages one of the principal causes of fatal issue in the abdominally wounded, it is therefore an important argument for the earliest possible intervention. Inactive hemorrhage from a solid viscus does not constitute an indication for operation.

Pain is usually a distinct symptom. The pain in the wounded parts, increased by difficulties in evacuation, and the pain caused by an advancing peritonitis both contribute to the shocked state of the wounded. Moreover, the degree of shock found is in approximate ratio to the time elapsing before the administration of morphine.

450

Cases with infection of the retroperitoneal tissues may show rather pronounced shock. Infection due to the presence of anaerobic microorganismsin badly lacerated muscles of the abdominal wall may be an important factor. The shock which accompanies an advancing peritonitis is readily recognized.

A hurried examination having been made to assure the surgeon that no progressive hemorrhage is present and that no splint readjustments are necessary, immediate treatment directed toward the relief of shock must be instituted. The preoperative ward of a hospital for seriously wounded should be organized to deal particularly with this condition. Since the subject of shock and its treatment is covered in detail in Chapter VII of this volume, no further reference is made to it here.

Bound up with the question of an accurate diagnosis are the questions of abdominal penetration and visceral injury. A tangential abdominal wound. especially if the tract is a short one, frequently gives no peritoneal penetration. With such wounds, where peritoneal irritation is lacking, the probability of abdominal entry is small. As demonstrated by Wallace,⁸ a wound above the pyloric level with entrance and exit to the right of the median line seldom results in visceral injury though the peritoneum may have been entered. Other through-and-through wounds mean certainly that the peritoneum has been penetrated but occasionally without visceral injury.

With only a wound of entrance, diagnosis becomes more difficult, and under these circumstances the X ray may give the greatest aid to the surgeon. Accurate information quickly obtained by means of a roentgenological examination is often the most important guide to the character of the wound with reference to visceral injury and to the exact situation of the missile.
 
SYMPTOMS OF VISCERAL INJURY

The general symptoms are those of shock, hemorrhage, pulse acceleration, and vomiting; the local ones, abdominal pain, tenderness, rigidity, and distension. The symptoms of shock are well known and its causes in abdominal injuries have been already enumerated. There may be no symptoms of shock even in the presence of serious visceral injury.

Though hemorrhage may produce little in the way of physical signs, the cardinal symptoms are a rapid, soft pulse, a low blood pressure, and obvious anemia. Patients with hemorrhage may exhibit excessive thirst and frequently show signs of air hunger. With any amount of bleeding intra abdominal fluid may be detected in the flanks, and local peritoneal signs may be moderately marked. The face is blanched and listless when the degree of hemorrhage has been considerable; it is anxious when infection in the peritoneum is progressive.

Whether there is hemorrhage or not, pulse elevation is the rule. With a peritonitis in progress the pulse rate steadily rises, and this sign is one of the valuable guides in demonstrating hollow visceral lesions. It is especially important in excluding a parietal wound with considerable muscle injury where local signs of peritonitis may be closely simulated.

Vomiting is usually present with a hollow viscus penetration and becomes more marked as peritonitis becomes more severe. Gastric wounds give earlier vomiting, with a vomitus frequently containing blood. On the other hand

451
 
wounds of the stomach sometimes occur with an entire absence of vomiting, so that this symptom can not be termed characteristic of gastric injury.

There is a great variation in the degree of abdominal pain, and the surgeon can not pass judgment upon the severity of the visceral lesion from the amount of pain the patient suffers. The location of the pain usually gives little aid in the attempt to localize the viscus injured. The most important feature of the pain from a diagnostic standpoint is its inception simultaneously with the receipt of the wound. The soldier abdominally wounded may have received a considerably dosage of morphine prior to admission to hospital, and if this has been promptly given his complaint of pain may be absent, particularly if he is seen early.

The local tenderness is the most pathognomonic sign of a ruptured hollow viscus and is the most constant physical symptom of peritonitis. It is always present and is localized over the region involved. It may be masked somewhat by morphine but it never disappears completely. Tenderness is also present with parietal wounds and contusions, but the other signs of visceral penetration are lacking.

Muscular rigidity is a very important sign of visceral injury, but it is less constantly present than tenderness. Charles ⁹ has seen cases of multiple perforation with an entire absence of muscular rigidity and again has encountered boardlike rigidity in severe wounds of the abdominal wall. Wallace ⁸ and many others have enumerated certain injuries which may be accompanied by marked rigidity without injury of an abdominal viscus. The more important are chest wounds with no abdominal lesion, wounds of the abdominal wall, and hemorrhage into the retroperitoneal tissues. The rigidity of an advancing peritonitis is generally progressive and increasing in intensity, while rigidity from introabdominal hemorrhage is less marked, and usually diminishes gradually if the hemorrhage has ceased. Muscular rigidity as well as local tenderness may be all important symptoms in diagnosing a visceral perforation when the wound of entry is in some remote region and the possibility of an abdominal lesion seems very unlikely.

Abdominal distension is not an early or important symptom of visceral penetration. It becomes more pronounced as peritonitis develops, but it is then of little value in diagnosis.
 
INDICATIONS FOR OPERATION

Certain groups of cases come to the hospital in which operation is contraindicated and these may well be considered here.

Moribund patients should he made as comfortable as possible with morphine. A so-called moribund ward is practicable and fills a good purpose. The patients in it must not be left surgically unattended but should be followed through to the end.

Cases with general peritonitis from a hollow viscus injury 24 or more hours old are generally hopeless subjects for radical treatment: Nearly 100 percent mortality occurred among such of our cases following operative intervention., The expectant plan should invariably be followed. Morphine in liberal doses, Fowler's position, heat, rest, alkaline fluids, sugar solution by rectum,

452
 
and saline solution beneath the skin are the lines of treatment to be followed. These cases must he carefully separated from the moribund class. Unless they are kept under constant supervision the rare individual whose peritoneal defense mechanism may bring him into the operable class may lose his only chance for life, because the psychological moment for operation is passed unnoticed.

Except for the individuals in whom shock is due to progressive hemorrhage the badly shocked should be kept in the preoperative ward under the eye of the ward surgeon, but the final decision as to operability or nonoperability must rest with the operating surgeon. When there is no amelioration of the symptoms of shock, operation should not be performed. As a working rule, the patient who shows no tendency to reaction within two or three hours never reacts.

Cases with through-and-through wounds of solid viscera without progressive hemorrhage do well without surgical intervention: the chances of recovery are better when no operation is performed. The presence of hemorrhage in these patients is usually differentiated from an advancing peritonitis by the pulse rate of 80 or 90. Such a patient should not be interfered with. Intervention frequently results in a renewal of the bleeding when the abdomen is opened and disaster may follow. Fluids by rectum or by hypodermoclysis with moderate morphine dosage are the indications. If the loss of blood has been considerable, transfusion may be resorted to when a fair degree of certainty exists that hemorrhage has ceased.

Short tangential wounds with unimportant abdominal symptoms, seen 8 to 10 hours after injury, are best left unoperated, but no such patient should be sent from the preoperative ward to one of the postoperative wards. During the height of an offensive military operation such an individual may easily escape the eye of the ward surgeon and a case in which operation may have become definitely indicated may be overlooked.

The basic indications for immediate operation are symptoms pointing to progressive intrabominal hemorrhage or hollow visceral penetration. Walters ¹⁰ and other surgeons speak of the obvious necessity for operation in Cases with visceral or omental protrusion; in cases with escape of gas or feces through the wound: and in eases with subcutaneous emphysema from the escape of intestinal gas, usually from the large bowel, into the tissues adjacent to the wound. Emphysema of such origin appears shortly after the receipt of the injury in contradistinction to the emphysema of gas gangrene.

In a time of great stress the problem of the seriously wounded may be a difficult one to handle. Practically it must be managed by first operating upon the best operative risks. When the stream of badly wounded becomes tremendous severe cases with wounds other than abdominal and with better hope of recovery must have the chance of life which operation gives. A frequent revision of the eases in the preoperative ward must be made that the surgeon may assure himself that no soldier with a fair chance of recovery with operation is passed by.

A good working principle is to operate if doubt of a hollow visceral lesion exists, for the mortality of such operation is exceedingly low when no visceral
 

453
 
penetration is found. In our evacuation hospitals the mortality for this type of operation was 6.5 percent. ¹¹ Exploration of the abdomen with negative visceral findings is attended, therefore, with little risk, compared to the uncertain possibilities of a serious advancing peritonitis if the man is left unoperated.

OPERATIVE TECHNIQUE

Since no type of war wound presents such a complex problem as the penetrating wound of the abdomen, the details of surgical technique must be studied with the greatest care.

There is little doubt that nitrous oxide-oxygen is the least toxic and the best borne of all the narcotizing agents, but, unfortunately, it does not give the complete muscular relaxation which is essential to efficient abdominal surgery. Its administration requires a skilled anesthetist and a bulky apparatus.

Ether is the agent universally employed and is by all odds the anesthetic of choice for abdominal cases. The ordinary open method of administration gives reasonably satisfactory results; the equipment required is the minimum. Marshall, ¹² with an extensive experience, found fewer complications when a warmed ether vapor was used, but in the American Army this method was not given a trial.

The shorter the period of anesthesia the less the degree of toxemia that will be produced and the better the prognosis for the patient.

Two general rules may be followed to aid the surgeon in his choice of an incision: (1) Plan the incision to meet the visceral injury suspected; (2) avoid, if possible, the projectile wound site, and so diminish the liability to wound infection. Always make the operative wound sufficiently ample to insure a unhampered exploration. A wound of 8 or 10 inches usually suffices. Some advise the paramedian incision as the one for general use, while others favor an opening in the median line either above or below the umbilicus, depending upon the organ probably injured. The latter is the incision of choice in most cases. For the upper abdomen some operators prefer a transverse incision or an oblique one parallel to the costal margin. This gives excellent access to wounds of the spleen, liver, kidney, or upper colon. The greatest objection to this type of incision is the difficulty in making a neat, rapid, and satisfactory closure. With a lesion below the umbilicus, and a wound in the flank, a transverse or oblique incision should be made. Such an incision affords good access to the retroperitoneal tissues and wounds of the posterior aspect of the large bowel. Transverse incisions in connection with abdominothoracie injuries are considered below under such lesions.

The general principle of wound disinfection by careful dissection must be as carefully observed in abdominal surgery as in wounds of the extremities. Careful excision of all soiled tissues must alwavs be carried out if the patient's condition permits. At times this is best done before the abdomen is opened, and the instruments used should then be discarded. Sutures may be placed if the patient may be held for 8 or 10 days, but it is safer not to close the skin. It may, however, be wiser in very serious cases to defer the abdominal wall dissection until the suture of the operative wound has been completed. Failure
 

454
 
 to carry out an efficient d ebridement will result in an infected abdominal wall, with the possibility of serious consequences.

Speed is important, but no false moves should be made. It must be remembered that an operation of more than one hour's duration usually means a shocked patient with little chance for recovery. Try to determine what organs may be excluded from the possibility of injury, but err on the side of thoroughness and keep the surgical traumatism to a minimum. Protect the skin adjacent to the wound with towels and skin clips to avoid contact of the cutaneous surface with abdominal contents. Do as little as possible, at the same time making the operation a thorough one.

If hemorrhage has been progressive or if upon opening the peritoneum there is more blood than was anticipated, seek at once for the source of hemorrhage and check it by clamp, packing, or suture. Considerable hemorrhage from the mesentery means inevitable resection.

Throughout the operation make the traction upon the abdominal wall as light as is compatible with proper manipulation of viscera. The detailed treatment to be applied in wounds of the various organs may be found below, where wounds of these viscera are discussed.

Closure should be done carefully in layers except in cases doing badly on the table, when through-and-through sutures may be employed. If the wound by the missile has crossed the line of operative incision it is better to leave the skin without sutures or but partially closed; and the same precaution against infection of the abdominal wall should be observed if a period of stress prevails and the work is necessarily somewhat more hurried.

No drainage should be employed in sutures of the small intestines, unless a very active peritonitis has developed. No drainage for stomach cases is necessary. On the other hand, drains of rubber dam (never gauze except to check hemorrhage) should always be used in wounds of the colon and rectum, especially those complicated by retroperitoneal injury. Gauze drainage for liver or spleen should be employed solely for hemostatic purposes. Drains are used to provide an outlet for leakage from the large bowel or to check hemorrhage, but should never be thought of as effectively draining the general peritoneal cavity.
 
POSTOPERATIVE TREATMENT

Practically every one of these patients has suffered a loss of body fluids, and the administration of fluid is the chief indication. For wounds of the solid viscera, stomach, and small intestines a Murphy drip of 5 per cent sodium bicarbonate solution, with or without 5 percent glucose, is the method of choice; or similar enemas at four or six hour intervals may be substituted. For wounds of the colon and rectum, hypodermoclysis with saline solution meets the indications. Saline infusion may be used in any case where considerable hemorrhage has occurred, but only as a temporary measure to tide over the man for a few hours until blood may be obtained. Transfusion in this latter group is frequently essential to recovery, and it may be necessary to repeat it. Blood grouping should always be done before transfusion, as it takes but a moment when Vincent's macroscopic test is used. Group IV
 

455
 
donors (Moss' classification) may be employed for a recipient of any of the four groups. Citrated Group IV blood may be collected at a distance and transported to the hospital in sterile bottles, and such blood may be kept for upward of 24 hours without fear of clotting. If there is no vomiting, water or very dilute alkalies may be given by mouth, but no fluid food should betaken for the first 24 hours. Restoration of the physiological activity of the bowel probably requires a still longer interval.

Pain is a constant postoperative symptom and morphine is a very important therapeutic agent for its relief. It should be given freely during the first 24 hours, moderately during the second, and sparingly or not at all during the third 24-hour period. If the patient is doing badly such a rule can not be adhered to and morphine should be given freely to the end.

Distension of the abdomen is a variable symptom, but is generally fairly marked and contributes much to the patient's discomfort. It is best treated by colonic irrigation, pituitrin, and local heat when the latter may be applied without discomfort to the patient. Fowler's position is particularly valuable in wounds of the lower abdomen, since it helps to localize the inflammatory process. It also helps to relieve the distension, especially when a rectal tube is made use of from time to time. The position may be maintained fairly continuously for the first 48 hours or 72 hours.

Vomiting is usually present and may become a distressing symptom. It is caused by the postoperative ether toxemia, the peritoneal traumatism incident to the wound and the operation, to an advancing peritonitis, or to a dilated stomach. If withholding of fluids by mouth for a few hours does not result in an early cessation of vomiting, gastric lavage with warm water, with or without sodium bicarbonate, repeated at two or four hour intervals, is the most effective means of treatment. In all the conditions named, except the advancing peritonitis, lavage usually gives effective relief.

The length of stay in the hospital in which operation has been performed and definitive treatment given should be from 7 to 10 days, or longer if military necessity will permit it. The period named brings the average patient far enough along in his wound healing and general convalescence to permit a safe evacuation to the rear. A certain small number of evacuated cases may develop postoperative complications, but military exigencies will usually allow the really bad cases to be retained forward for a longer period than the time used. Above all, careful nursing and continuous care on the part of the surgeon must be available, or many patients will suffer and a certain number succumb who otherwise might be saved. The dressing must be done by the most experienced hands available, the operator or his assistant doing this work whenever it is physically possible. If a ward surgeon dresses the wounds the surgeon himself must supervise his work and personally direct the patient's convalescence.
 
POSTOPERATIVE COMPLICATIONS

The more frequent and, therefore, important complications are infection and the development of fecal and urinary fistulae; secondary hemorrhage, nephritis, and pulmonary complications are less often seen.
 

456
 
Infection is encountered in the form of local wound infection, as a localized peritoneal abscess, or as general sepsis. The wound healing is good or bad in direct proportion to the amount of infection present in the abdominal wall. The surgeon's first effort, therefore, in combating would infection is the prevention of it by painstaking surgery at the time of operation. Careful débridement of the abdominal wall and proper placing of the abdominal incision with respect to the wound, combined with the nonsuture of skin and subcutaneous tissues in doubtful cases, are the most important details to be observed. The surgeon must also carry out only partial skin closure in the presence of an active purulent peritonitis, as the parietal wall will necessarily be more or less contaminated. Very disastrous gas and streptococcus infection of the abdominal wall may develop, usually early in the postoperative course, and such a condition calls for wide incision, combined, where possible, with Carrel-Dakin treatment. Evisceration of considerable intestinal contents may occur with an infected abdominal wound, and such cases usually do badly. Immediate replacement of viscera must be accomplished with rapid resuture of the peritoneal muscular, and fascial layers.

Localized peritoneal abscess is more often a later complication, occurring from a few days to two or three weeks after operation. It may complicate a fecal fistula. It is the most favorable outcome to be looked for in a case of diffuse peritonitis. Considerable difficulty in making a proper diagnosis of the location of the purulent collection may be encountered. It may be placed in practically any portion of the peritoneal cavity and may point in the buttocks, perineum, or flanks. Cases of infection following local leakage into the retroperitoneal tissues are the most difficult ones to diagnose and treat successfully. The indication in all these cases is drainage by the simplest possible procedure.

General sepsis is relatively infrequent. The fatal cases of peritonitis usually die within a few days before sepsis has become general. It may occur in connection with retroperitoneal infection or with badly infected operative or projectile wounds. No effective treatment has been found to combat general sepsis successfully.

Fecal fistula is a frequent complication of wounds of the small and large intestine, being encountered most often after suture or resection of the latter. It also follows operation for inaccessible rectal injuries. It may occur at any time in the postoperative course of the abdominal wound, and the fecal discharge usually appears in either the operative wound or along the original wound tract where inaccessibility has made a careful dissection impossible. Makins ¹³ cautions us to bear in mind that a bruised intestinal wall without complete entry of the lumen may at times break down with the formation of a fecal fistula. Such instances have been verified by the findings of a previous exploratory operation. Frequently a fecal fistula will close spontaneously, particularly one complicating operation. In a small proportion of cases, however, suture or resection may be required to relieve the condition. Every precaution must be taken in such an operation to isolate the general cavity from the operative field; and adequate drainage, preferably with rubber dam, should be provided. Such a type of operation is usually done in the base hospital when it has become evident that the fistula will not close spontaneously.

457
 
Secondary hemorrhage is rare as a sequel to abdominal injuries. It may occur from septic erosion of a large vessel or from a reopening of a partially healed wound of a solid viscus. The treatment is the same as for primary hemorrhage.

Any of the inflammatory processes of the lung or pleura may complicate the postoperative course of all abdominal lesion: they are, however, comparatively infrequent. The most striking pulmonary complications encountered by the writer were four cases of pulmonary embolism (diagnosis being made by symptoms, since no autopsy could be performed) which were fatal in from one-half to two hours after the intravenous injection of gum solution. The further use of this agent was discontinued by the writer and his associates.

Nephritic complications are comparatively infrequent. The indications for treatment are the same as in civil surgery.
 
TREATMENT OF VISCERAL INJURIES

In a certain proportion of cases the peritoneal cavity is opened without any injury of the viscera. Practically this group may be considered in connection with nonpenetrating wounds of the abdominal wall. The possibility of an infection of the peritoneum from the retained projectile, or from foreign material carried into the peritoneal cavity, makes them definitely more serious than the nonpenetrating wounds. The missile, however, usually becomes encysted, when it may give no symptoms whatever. In rare instances a localized peritoneal abscess may result.

Cases with short tangential wounds, and the occasional cases in which after 8 or 12 hours no alarming abdominal symptoms have developed and in which the patient's general condition is excellent, are particularly the ones in which the question for or against operation may arise.

When any doubt of the wisdom of intervening exists it is better to operate. As stated above, it is far better to open every abdomen when there is question as to visceral injury than to abstain, for the mortality after operation where no visceral lesion is found is very small and the hazard is a tremendous one if a true perforation of a hollow organ is left without operation. Further, operation always furnished an opportunity to search for and often to remove the foreign body itself.
 
WOUNDS OF THE STOMACH

Stomach wounds comprise about 7 percent of all abdominal injuries coming to the hospital for treatment. Two-thirds of all gastric lesions show no other accompanying visceral injury discoverable at operation. Wounds of other organs most often encounterel are those of the small gut. liver, colon, kidney, and spleen, in the order of their frequency. ¹

The wounds are usually two in number and are most often situated on the anterior and posterior walls. If the anterior opening is small and the organ was not distended at the time of injury there may be no protrusion of mucous membrane and no escape of gastric contents. Leakage may, however, occur into the lesser sac when none is present anteriorly. If but a single wound is present it is usually of the anterior wall. Under the conditions, great care must be exercised in excluding a posterior wall perforation, for in comparison with the
 

458
 
injury of the anterior aspect that of the posterior may be much more difficult to detect. Lesions of the borders and orifices are comparatively infrequent. Considerable damage to the gastric wall is more often seen with wounds of the lesser curvature and those parallel to the walls of the stomach. The more ragged and larger wounds are usually caused by shell and hand-grenade fragments and machine-gun bullets fired at close range.

There are local signs of peritoneal irritation, but these are definitely less marked than with injuries of the small intestine. Only moderate shock is present in most of the cases; where a severe hemorrhage has occurred the degree of shock is profound. Exceptionally, there is no gastric leakage into the peritoneal cavity. The cardinal symptom is early and persistent vomiting. Rarely, however, vomiting may be absent. Escape of gastric contents or gas may take place from the abdominal wound.

Very rarely recovery has been reported without intervention. The safe rule to follow is operation in practically every case. The best incision is the median or paramedian. The perforation of the anterior wall of the stomach is readily recognized. The opening in the posterior wall is best sought for through the gastrocolic omentum just below the stomach. Pauchet's approach, recommended by Eastman,¹⁴ is made through an opening in the mesocolon, the line of dissection passing just above the transverse colon. This frequently gives good access, but it is not recommended for general use, as repair work is more difficult when this technique is used. A ragged wound should be trimmed off rapidly before suture. The greatest difficulty in accomplishing a good closure of the stomach wall will be found with wounds involving the lesser curvature and those high up near the cardiac orifice. Gastroenterostomy should be avoided if possible, for a higher mortality results in the cases in which it is performed. Drainage in gastric cases should be employed only where a fairly well developed peritonitis is present or if suture of the stomach wound is difficult or impossible. The most important precaution to be observed in the postoperative care of these cases is careful feeding. Only water should be allowed by mouth during the first 24 hours and liquid diet for the following three days.

The seriousness of gastric lesions is in no wise comparable to the grave conditions caused by wounds of the small intestine. A favorable outcome may be possible even if the case is seen a considerable time after receipt of the injury, as peritonitis advances comparatively slowly. Numerous cases have recovered where operation has been performed from 24 to 36 hours after the receipt of the wound.

The mortality of all gastric wounds is approximately 55 percent. ¹ Uncomplicated wounds of the stomach give a mortality varying from 25 to 50 percent.
 
WOUNDS OF THE SMALL INTESTINE

The proportion of total small intestinal wounds, complicated and uncomplicated, to all abdominal lesions is approximately 22 percent.¹
           
Wounds of the colon are much more frequently encountered than those of any other complicating visceral lesion. The injuries next in order of frequency are those of the stomach and bladder, while wounds of the liver, kidney, rectum, and spleen are still less often encountered.
 

459
 
Duodenal injuries are fairly infrequent, comprising approximately 6per cent of all small gut wounds.¹ Injuries to the jejunum comprise about 23 percent; to the ileum approximately 71 percent.¹ Multiple lesions are almost universally encountered, at times reaching the number of 15 or 20, but the average number to be expected is from 4 to 6. The wound may be small or large, depending upon the character of the missile, the velocity at which it is traveling, and the angle of entry into the gut. When the projectile strikes the intestine vertically two perforations are almost invariably found. As a rule there is a certain protrusion of the mucous membrane, but if the wound is a small one there may be no pouting and such cases may show no leakage. The more nearly parallel the wound is to the long axis of the gut the more the damage to the visceral wall and the larger and more ragged is the wound itself. Extensive laceration of the intestinal wall and even complete division of the gut are not so very unusual; in such cases a considerable tearing of the mesentery is frequently found. The mesenteric lesions are especially important from the standpoint of hemorrhage and because of the necessity for resection with its added shock and operative hazard.

It is generally best to begin the exploration of the gut at the ileocecal valve, but if the wound is high up the duodenojejunal junction may be first examined. Work rapidly upward, if beginning at the ileocecal valve, or downward, if beginning at the duodenojejunal angle, being careful to replace within the abdomen every 8 or 10 inches of segment after its examination. In this way the entire ileum and jejunum with their mesentery are carefully examined for perforations. When a wound of the intestine is encountered clamp the opening tightly, protect it with a pad, and hold it outside the abdomen, and as each lesion is discovered treat it in the same way. It is a good general rule to refrain from repairing any perforation until the entire length has been examined. The writer has seen a small gut suture, requiring 15 minutes, performed upon a segment of small intestine that later had to be resected because of mesenteric injury. Exception to this rule may be taken when normal bowel and mesentery are present several inches to each side of the lesion or where a large number of perforations with their pad coverings would form a serious obstacle to efficient technique.

Careful search for complicating wounds of the stomach should then be instituted and the lesions appropriately treated. Other visceral injuries should be sought for and the colon should as a rule be the last one explored, as lesions here may necessitate performing a colostomy. If, during the course of an operation, a wound of the colon is encountered, it is wiser to treat it immediately if suture only is required.

The vast majority of small intestinal wounds are satisfactorily closed by a single purse-string suture of silk or chromic gut. Suture should always be practised if possible, as resection is attended with far greater hazard. With numerous small lesions close together, suture is preferable; if gut damage has been considerable, resection may prove to be the better procedure. The best rule is to resect only when it is impracticable to suture. A double row of sutures should always be employed when resection is done.
 

460
 
The postoperative care of cases with small gut injury has been outlined above under the general discussion of penetrating abdominal wounds.

The frequency of wounds of the small intestine an rid the high mortality attending operation for their relief make these injuries the big problem in abdominal military surgery. The mortality rate in cases of wounds of the small intestine in the American Expeditionary Forces, including the operated and unoperated, was as follows: ¹ Duodenum, 80: jejunun, 78.8; ileum, 73; small intestines (not specified), 7.9. Resection gives regularly n mortality 50 peraent higher than does suture.
 
WOUNDS OF THE COLON

Wounds of the colon represent about 22 percent of all intra-abdominal visceral injuries.¹ Perforating wounds of the colon are much less often multiple than those of the small gut because of the lack of numerous intestinal coils. The multiple lesions that occur usually involve the pelvic colon. Some of the smaller perforations may be due to minute bone spicules penetrating the bowel wall, and this type of lesion is much more difficult to recognize than an injury primarily due to a missile. Some of the wounds are large and ragged and a complete division of the bowel may be found, but less frequently than in wounds of the small intestine.

Retroperitoneal perforation with its consequent fecal leakage and cellulitis constitutes one of the difficult problems to be dealt with. The posterior perforation may be a minute one which is difficult to recognize. Injuries of the portions of the gut which are without a mesentery, the ascending and descending colons, are particularly liable to be accompanied by a serious retroperitoneal infection.

Retroperitoneal injury of the transverse colon may only be detected when the lesser sac is explored.

The symptoms and diagnosis of wounds of the colon have been considered above in the discussion of abdominal wounds under the heading of "Diagnosis." The special factors which make diagnosis difficult are the liability to retro-peritoneal infection and the inaccessibility of the splenic and hepatic flexures, particularly the former. A grave acute sepsis may rapidly develop in connection with retroperitoneal cellulitis. Further, the type of peritonitis which the surgeon encounters in connection with perforations of the colon is very likely to produce more aggravated local and general symptoms than that associated with the involvement of the small intestine.

No one incision will satisfy all requirements with wounds of the colon. A median or paramedian incision is best used when a lesion of the transverse or pelvic colon is to be dealt with. The best incision for wounds of the cecum or of the ascending and descending colon is a transverse one in the flank, for this allows an easy access to the posterior portion of the bowel and a better chance of discovering a posterior perforation. The incision of choice for either of the colonic flexures is a subcostal incision on either side, prolonged vertically downward if additional space is required. This incision is particularly valuable on the left side because of the posterior position and inaccessibility of the splenic angle. The general principles to be followed are: Suture whenever possible
 

461
 
to secure a satisfactory closure, and always employ a double row of sutures. Avoid resection; colostomy is to be preferred. If a colostomy is performed are section done at a later date sometimes gives a satisfactory result. Colostomy is to be advised with large ragged openings, particularly those occurring in the cecum, descending colon, and sigmoid. Drainage is a most important factor with wounds of the large bowel. Always drain when any doubt of the integrity of the suture line exists and in every case of proved or questionable retroperitoneal injury.

There is a slightly lower mortality record with wounds of the large intestine than with those of the small bowel, the figure for the former being 59.6 percent. ¹ The cases that do badly die from retroperitoneal sepsis, which may be most acute, or from a peritonitis secondary to fecal leakage, preceding or subsequent to operation. The wounds that are sutured do better than those in which an artificial anus is employed; the latter group gives the high mortality rate of 70 percent.¹
 
WOUNDS OF THE RECTUM

Injuries of the rectum are comparatively infrequent, constituting 2.4 percent of the lesions of abdominal viscera.¹ Complication by other injuries is infrequent. Associated lesions which may be encountered are those of the bladder and pelvic colon, or, less frequently, injuries of the small bowel.

The lesions vary in size from small perforations caused by a minute projectile or a fragment of bone to extensive lacerations. Wounds of the rectum often show a wound of entrance in the buttock or upper portion of the thigh or in the perineum. If the wound is an extra peritoneal one, fecal leakage posteriorly may occur. with the rapid development of a grave cellulitis.

Intra peritoneal injury of the rectum gives rise to a rapidly developing acute peritonitis which is still more aggravated if complicating lesions are present. A wound of entry through the buttock or perineum in a patient exhibiting symptoms of peritonitis in the lower portion of the abdomen should always make the surgeon suspicious of a rectal injury. Local tenderness in the posterior rectal wall made out by the examining finger in the rectum and associated with evidences of infection in the perineum always suggests an extra peritoneal rectal wound, especially when associated with general symptoms of a septic type.

The extra peritoneal injuries are best treated by careful d ebridement of the buttock or perineal wound, the dissection being carried upward and into the rectum. It may be necessary to open widely the lower segment of the bowl in order that complete dissection of the tract may be accomplished and that adequate drainage may be most effectively placed in the retroperitoneal tissues. Extensive lacerations of the lower segment may require a colostomy. Intra peritoneal injuries are treated by a median laparotomy with suture of the opening wherever it is possible to accomplish it. Drainage through the lower angle of the operative wound should always he practiced, rubber dam being the best material for the purpose. If a suture can not be made, owing to the depth of the rectal wound in the pelvis, a colostomy should be performed.

The mortality with wounds of the rectum is 45.19 percent.¹ Usually death is due to a rapidly advancing sepsis in the retroperitoneal tissues or to a severe spreading peritonitis.
 

462
 
WOUNDS OF THE LIVER

Wounds of the liver comprise 13.3 percent of all abdominal injuries. Approximately three-quarters of all liver lesions are uncomplicated ones. Associated wounds to be considered are, in the order of frequency, wounds of the colon, stomach, and kidney; injuries of other organs are much less often found.

Clean-cut liver wounds are very unusual. There may be any type of lesion from a small perforation to a slit or a large ragged excavation, and in some cases a loss of liver substance is encountered. Whatever the type of projectile, a large wound of exit is to be expected, and lacerations in all wounds is the rule.

Hemorrhage is always present, varying from a slight oozing to a severe and rapidly fatal hemorrhage. More often the bleeding tends to subside spontaneously. Peritoneal symptoms are to be expected from the presence of blood in the peritoneal cavity. A dullness in the flanks, particularly on the right side, may be made out if the amount of bleeding has been considerable. A case seen two or three days after the injury frequently shows a slight degree of jaundice; late jaundice usually means sepsis. Where the loss of blood has been considerable the patient exhibits a marked degree of shock.

The diagnosis of liver injury is made from the position of the wound and the symptoms of intra abdominal hemorrhage. The early appearance of jaundice should make one suspicious of liver injury.

As a general rule the expectant treatment should be followed. Operative intervention should be made when other visceral lesions are suspected, where the hemorrhage is serious and progressive, or where the foreign body retained is a very large one. The incision best suited for the management of liver wounds is a right subcostal one, though a median or paramedian approach may give adequate exposure. When the abdomen is entered, if a small wound is found without active hemorrhage, it should be left alone. A larger wound from which the bleeding has ceased should be packed or sutured, preferably the latter, for secondary hemorrhage from such a wound is not unlikely. In placing sutures use a large round needle, blunt end first, the suture being of a mattress type. This form of suture should not be drawn tightly in order to prevent its cutting through the liver substance. A properly placed suture will effectively control a very active hepatic hemorrhage. The retained foreign body, if of considerable size, should be removed in order to avoid the subsequent complication of liver abscess. The shock present should be combated with heat administration of fluids, and adequate doses of morphine. When the loss of blood has been considerable and the hemorrhage is no longer active transfusion should be performed.

The mortality rate of liver wounds is 66.27 percent.¹

A considerable number of uncomplicated wounds of the liver treated expectantly get well. The cases of this type operated upon give a mortality of about 5 percent; ¹ the mortality rate to be expected if complicating lesions are encountered is in the neighborhood of the mortality rate for liver wounds as a whole. Certain of the cases with retained foreign body in the liver develop a hepatic abscess and may succumb to sepsis.
 

463
 
Wounds of the gall bladder and bile ducts are so comparatively infrequent that any special consideration of them will be omitted. Records of but 9 cases in the American Expeditionary Forces exist, with a mortality rate of 77.78 percent.¹
 
WOUNDS OF THE PANCREAS

The cases reaching the hospital comprise about 0.2 percent of all abdominal injuries.¹ The proximity of the organ to the great vessels may give a rapidly fatal result upon the field of battle, so that a certain proportion of these cases never reach the hospital.

The accompanying lesions usually found are those of the stomach; other organs are much less frequently involved. The one important element to success is adequate drainage.

At least half of the cases prove fatal from an undiscovered or poorly drained injury. The writer had one case showing an anteroposterior wound in the epigastrium. Operation revealed a small shell fragment lodged in the head of the pancreas with an associated contusion but no penetration of the adjacent duodenal wall. The foreign body was readily removed, and wound in the pancreas drained, and the patient when last seen, seven days after operation, was convalescing satisfactorily.
 
WOUNDS OF THE SPLEEN

Wounds of the spleen are much less frequent than those of the liver, 49 only having been recorded as occurring in the American Expeditionary Forces.¹ Two-thirds of the splenic wounds show complicating lesions.

 The injuries may be of all types, from a small perforation or moderate laceration to a complete separation of a considerable portion of the organ, or an avulsion from its pedicle. The visceral injury most frequently complicating a splenic wound is a lesion of the kidney; organs less often involved are the colon and stomach.

 As with wounds of the liver, hemorrhage is the chief symptom along with the shock resulting from it. The intra abdominal signs of hemorrhage described under wounds of the liver apply equally to splenic wounds. The fluid, however, is more apt to accumulate in the left flank. The diagnosis is made from the position of the wound and the accompanying signs of hemorrhage and shock.

The best incision of approach is the left subcostal, but a left rectus incision may give adequate access. In some cases a liberal median or paramedian incision may give abundance of room. The incision for abdominothoracie injuries is dealt with in a subsequent paragraph. Where the hemorrhage is inactive and has not been of large amount abstention is the best rule to follow in an uncomplicated case. A small wound encountered at operation and showing no active hemorrhage should be left alone. Suture in the splenic tissue is less effective in controlling hemorrhage than in liver tissue, and a continuation of the bleeding is always a possibility. Packing in cases of this sort is desirable. Splenectomy, though advised by some writers, notably Depage, ¹⁵ gave practically a 100 percent mortality in the American Expeditionary Forces.¹ The treatment of hemorrhage and shock is the same as with liver injuries.

The mortality rate is 63.26 percent.¹ Hemorrhage is the cause of death in practically all the uncomplicated cases.

464
 
WOUNDS OF THE KIDNEY

Wounds of the kidney constitute 6.3 percent of all abdominal injuries; ¹ one-half of the cases are uncomplicated. Wounds of the right kidney are complicated by liver injuries in about one-third of all cases, and wounds of the left kidney are attended with splenic lesions almost as frequently. The hollow viscera most often wounded are the small gut or colon, while stomach lesions are less frequently encountered.

Wounds of the hilum include injuries to the renal vessels or to the pelvis itself. Wounds of the renal vessels are usually serious and often fatal because of the severe hemorrhage. It is well to bear in mind that an injury to any of the renal branches results in necrosis of that portion of the kidney tissue which the vessel supplies, as the anastomotic circulation of the kidney is very poorly developed. Injuries of the pelvis itself are comparatively rare and require no further discussion.

Wounds of the parenchyma may be of any type from a simple perforating or tangential wound to a very extensive laceration or destruction of the organ.

 The X ray gives important aid in arriving at a correct diagnosis in this group of eases. Hemorrhage as well is an important symptom; the bleeding may occur from the external wound, it may appear in the urine, or may form a retroperitoneal haematoma. Bleeding into the peritoneal cavity may take place if the rupture has been intraperitoneal. Under such conditions one may expect a tender and rigid abdominal wall with dullness in the flank. Under any circumstance, if loss of blood has been considerable, the patient may exhibit symptoms of shock.

Leakage of urine is present when the pelvis of the kidney has been opened or the ureter has been torn. but seldom with wounding of the parenchyma itself.

Retention of urine is seen in a certain proportion of cases. Sepsis is a later complication, which may develop where inadequate drainage has been established or where hollow visceral complications have resulted in a fecal fistula into the wound. Cases with sepsis exhibit all the symptoms common to this condition. Vomiting is frequently seen. Out of 42 cases reported by Fullerton ¹¹ arriving at a base hospital with kidney lesions 9 suffered a secondary hemorrhages 1 on the third day, 1 on the seventh day, a between the tenth and fifteenth days, and 2 after four weeks.

Conservatism should be the keynote in the treatment of lesions of the kidney. With a penetrating rifle wound and where these is no evidence of intestinal involvement rest and a liberal administration of opium are indicated. Alarming, hemorrhage, urinary leakage, advancing symptoms of sepsis, or a large retained foreign body are the principal indications for operative intervention. A complicating hollow visceral injury requires immediate operation. The incision of choice is a transverse or oblique one, which may be extended as far forward as is necessary to give adequate exposure of the kidney or to treat other complicating visceral injuries. If a small lacerated wound with a retained foreign body is encountered the removal of the missile and drainage may be all hat is necessary for a satisfactory recovery. Charles ⁹ advises under these

465
 
circumstances a débridement of the damaged kidney tissues, followed by suture and drainage. A rubber-covered clamp upon the renal vessels during this procedure gives satisfactory control of hemorrhage. If such dissection can be done with little sacrifice of kidney tissue it is a legitimate procedure. Nephrectomy should be avoided wherever possible, for in the push of advanced war surgery definite information as to the function of the other kidney must of necessity be lacking. Nephrectomy, however, must be performed where the vessels themselves are seriously damaged or where the injury to the kidney itself is extensive.

The mortality in uncomplicated cases varies from 25 to 30 percent.¹ The fatal cases far forward succumb to hemorrhage. In the rear areas sepsis and secondary hemorrhage are the chief factors leading to death.
 
WOUNDS OF THE URETER

Wounds of the ureter are infrequent. A gross injury calls for nephrectomy. Where a small wound is encountered suture may be attempted, but, in general, ureteral wounds will heal spontaneously if left alone.
 
WOUNDS OF THE BLADDER

Injuries of this organ comprise approximately 5 percent of all abdominal lesions and one-half of the cases are uncomplicated.

In Fullerton's ¹⁷ series 70 percent of the cases were complicated by intestinal or bone injuries or both. A wound of the rectum may be expected in from 10 to 15 percent of the cases. Injury of the prostate is comparatively rare.

Cathelin,¹⁸ in a series of 29 bladder wounds, found that the entrance was placed posteriorly 18 times, anteriorly 7 times, and laterally in the remaining cases. A wound of exit was present in but 5 of the patients.

Bladder perforations may be caused by either the projectile or by bone spicules from the fractured pelvis. The lesion of the bladder may be extra peritoneal or Intra peritoneal and may vary considerably in size. Legueu ¹⁹ reported 10 cases of fracture of the pelvis associated with bladder injury in which a vesical calculus was demonstrable. Other observers have not so uniformly encountered such conditions.

Hemorrhage in connection with bladder wounds is usually not serious, but its occurrence into the bladder suggests an extra peritoneal lesion. With this type of injury a considerable haematoma is not infrequently found in the vicinity of the wound in the viscus. Urinary discharge through the entrance wound is fairly uncommon. An empty bladder should make one strongly suspicious of an intraperitoneal perforation. Leakage of urine into the peritoneal cavity causes a considerable degree of peritoneal irritation with definite local signs.

Most of the extra peritoneal lesions result in pelvic cellulitis. In the long standing cases associated bone necrosis and calculus formation may be expected, and in most instances a cystitis still continues.

 Extraperitoneal lesions are best treated by wide incision down to the bladder wound, which should be sutured if possible. Ample drainage of soft
 

466
 
tissues must then be provided. Suprapubic cystotomy furnishes the best type of bladder drainage and should generally be employed.

The patients suffering from Intra peritoneal wounds, and especially those with associated visceral lesions, show such decided symptoms that little doubt exists as to the wisdom of laparotomy. Such lesions require operation with suture of the bladder wound and rubber-dam drainage down to the stitch line. A retention catheter or perineal drainage, preferably the former, meets the indications; suprapubic cystotomy should be avoided if possible.

Retained missiles are fairly frequent in the bladder. The projectile must always be sought for within the bladder, and if present removed.

In the latter stages of bladder wounds one must consider the treatment of persistent urinary fistulae, calculi, and sepsis. Cathelin's method of dealing with persistent fistulae is an efficient one. He dissects a cuff of skin and in folds it by suture down to the bladder wall, later bringing muscle and aponeurosis over it. Calculi had best be removed by the urethra, with or without crushing, but a suprapubic cystotomy may be necessary. The treatment of sepsis is supportive but the cases usually result disastrously.

The presence of complicating injuries may require considerable modification in operative technique. If a rectal or colic injury is so extensive as to make a colostomy imperative a suprapubic cystotomy should be dispensed with.

Very variable statistics will be found in the literature concerned with the mortality rate following uncomplicated bladder wounds, but the average is about 50 percent. Where the bladder injury is associated with a lesion of the small intestine a much higher figure is reached, running to 75 or 80 percent. The causes of death are sepsis, general peritonitis, or, much more rarely, a secondary hemorrhage from the pelvic vessels.

An important prophylactic precaution in the avoidance of bladder wounds consists in an invariable order that soldiers should empty their bladder before going into action.
 
ABDOMINOTHORACIC INJURIES

These wounds comprised 4.6 percent of all the thoracic injuries coming to the evacuation hospitals, American Expeditionary Forces, for treatment. ¹¹

The lesions encountered are thoracic, diaphragmatic, and abdominal. Duval, ¹⁹ gives the general rule that with a wound of entry in the chest the thoracic lesions are more apt to be the serious ones, while with entry through the abdomen the abdominal injuries are more often the graver ones. Frequently there are several wounds of the lungs and more than one lobe is occasionally involved. The types of lesion are the same as one encounters in simple thoracic wounds. A certain amount of hemothorax is always present, but the amount is variable. In approximately one-third of the cases a hollow abdominal organ is penetrated. An uncomplicated liver injury is more common than one of the spleen.

The wound in the diaphragm may be made either by the projectile or by a fractured rib. The diaphragmatic wound may vary from a small puncture to an irregular opening of large size; the shape is often slit-like, in which instance it
 

467
 
is usually not more than an inch long. In about 10 percent of the cases herniated abdominal organs will be found in the chest cavity. The omentum is the structure most often seen, and if viscera are extruded into the thorax the omentum usually accompanies them. Next in order of frequency are the spleen, stomach, and transverse colon. Practically all herniations occur through a wound in the left side of the diaphragm.

Through-and-through wounds involving both sides of the diaphragm are seldom encountered in the hospital, for most of them die far forward before evacuation can be carried out.

The pathognomonic symptoms of abdominothoracic wounds are dyspnea, the breathing being rapid and labored; sudden pain in the abdomen at the time of the receipt of the wound; hemothorax; abdominal rigidity over the corresponding half of the abdomen, especially in its upper part; and shock, which is partially dependent upon the degree of hemorrhage and partially upon the respiratory distress. Other abdominal symptoms such as definite local tenderness and vomiting may also be evident.

COURSE OF TREATMENT DEPENDENT UPON TYPE OF INJURY

If there are separate wounds of entrance for the abdomen and chest and the latter one is not a blowing wound, the abdomen should be opened immediately if a hollow visceral lesion is suspected, and the chest should be left undisturbed. With a wound of entrance on the right side of the chest authors vary as to the procedure to be adopted. If the missile is a small one and X-ray examination localizes it in the liver the wisest course is nonintervention. If the chest wound is a blowing one it should be closed by suture in which the muscular layer is included. No blowing chest wound should ever be allowed to get past the regimental dressing station without a closure of the muscular layers of the wound by suture. This rule should be adhered to even if asepsis can not be maintained. Many advise operation in this group of patients, the steps being de bridement of the chest wall, and exploration of the diaphragm and liver through an intercostal or vertical wound after the chest lesions have been cared for. The foreign body is removed from the liver when possible, the diaphragm sutured, and finally the chest itself closed without drainage.

With a wound on the left side above the level of the eighth rib and an associated abdominal injury which is high up a different problem is presented. Excellent constructive surgery has been developed by Duval ²⁰ in cases of this type. He uses a vertical incision upon the chest wall, beginning near the thoracic wound. The ribs are sectioned in order to allow access to the thorax, and the prolongation of the wound downward opens the abdomen; at times this lower extension may be continued obliquely forward. Other surgeons employ a transpleural approach to the upper abdomen with an incision which is roughly transverse. A rib may be resected, or access may be had through an intercostal space with the aid of Lilienthal's rib spreader. The chest wound is carefully dissected out and any soiled rib or loose bone fragments are removed. With fresh instruments the pleural cavity is explored and any lesions encountered are taken care of. The wound in the diaphragm is then sought for and enlarged up to 5 or 6 inches; this exposure gives a satisfactory

468
 
approach to the abdomen. Any abdominal injury found is cared for and the diaphragm closed. It is generally wise to obliterate the pleural space lowdown by suturing the diaphragm to either the lung or the lateral pleural wall. The chest should be closed without drainage.

With a wound on the left side and the point of entrance below the eighth rib the lesion within the abdomen is apt to be more serious. In this group of cases the abdomen should first be dealt with through a separate incision. At times it is possible to deal properly with the opening in the diaphragm from below. If the pulmonary injury warrants intervention it may be explored through an intercostal space. The closing off of the pleural cavity is accomplished as described in the preceding paragraph.

Duval to the contrary notwithstanding, the mortality rate is principally due to the lesion in the abdomen, and is distinctly higher when a hollow viscus has been penetrated.
 
LESSONS IN CIVIL ABDOMINAL SURGERY GAINED FROM THE WAR

The contributions of the war to civil abdominal surgery may be summarized as follows: (1) It has shown the types of tangential wounds which are not infrequently without visceral lesions. (2) It has taught the unwisdom in most cases of relying upon the so-called expectant treatment of abdominal wounds and the soundness of early radical operation. (3) It has demonstrated the wisdom of waiting an hour or so before operation is attempted in cases with severe shock. This rule applies, of course, only to those instances where shock is not due to active hemorrhage. (4) It has taught us the best methods of handling abdominothoracic injuries. (5) It has given an unusual opportunity to review the whole subject of abdominal drainage, strengthening our convictions that the general peritoneal cavity can not be drained, but that it is possible to drain a single focus within the peritoneal cavity. (6) It has emphasized again and again that speed, dexterity, simplicity of technique, and the minimum of traumatism are essential to success.
 
REFERENCES

(1) Based on Sick and Wounded Reports made to the Surgeon General. On file, Historical Division, S. G. O.
(2) La Garde, Louis A.: Gunshot Injuries. William Wood and Company, New York, 1916, 2d ed., 262.
(3) Gibbon, John H.: Treatment of Gunshot Wounds of the Abdomen. Journal of the American Medical Association, Chicago, July 19, 1919, lxxiii, 187.
(4) Wallace, Cuthbert: A Study of 1200 Cases of Gunshot Wounds of the Abdomen. British Journal of Surgery, Bristol, 1916-17, iv, No. 16, 679.
(5) General Orders No. 70, G. H. Q. A. E. F., May 6, 1918.
(6) Organization and Operation of Mobile Hospital Units, by Col. E. C. Jones, M. C., undated. On file, Historical Division, S. G. O.
(7) Medical and Surgical History of the War of the Rebellion. Washington, Government Printing Office, 1876, Surgical Volume, pt. 2, 3-208.
(8) Wallace, Cuthbert: A Preliminary Note on the Treatment of Abdominal Wounds in War. Journal of the Royal Army Medical Corps, London, December, 1915, xxv, 591.
(9) Charles, R.: Gunshot Wounds of the Abdomen at a Casualty Clearing Station. British Medical Journal, London, March 23, 1918, I, 337.
(10) Walters, C. Ferrier, Rollinson, H. D., Jordan, A. R., and Banks, A. Gray: A series of 500 Emergency Operations for Abdominal Wounds. Lancet, London, February 10, 1917, i, 207.
(11) Based on reports of surgical operations at evacuation hospitals, A. E. F., undated. On file, A. G. O., World War Division, Medical Records Section.
(12) Marshall, Geoffrey: Anesthetics for Men with Wounds of the Abdomen. British Journal of Surgery, Bristol, 1916-17, iv, No. 16, 733.
(13) Makins, Sir George: A Study of One Hundred and Eleven Cases of Perforating Wounds of the Gastrointestinal Canal which Occurred amongst a Consecutive Series of Two Hundred and Two Perforating Wounds of the Abdomen in which the Presence of Visceral Injury was Certain. Journal of the Royal Army Medical Corps, London, 1916, xxv, No. 1, 1.
(14) Eastman, James Rilus: The Question of Operation in Gunshot Abdominal Wounds. Journal of the American Medical Association, Chicago, September 28, 1918, lxxi, 1036.
(15) Depage, A.: Note sur les plaies pénétrantes de l'abdomen traitées a l'ambulance de la Ocean a la Panne. Bulletins et mémoires de la société de chirurgie de Paris, March 14, 1917, xliii, 691.
(16) Fullerton, Andrew: Gunshot Wounds of Kidney and Ureter as Seen at the Base. British Journal of Surgery, London, 1917, v. No. 18, 247.
(17) Fullerton, Andrew: Observations on Bladder Injury in Warfare. British Journal of Surgery, Bristol, 1918, vi, No. 21, 24.
(18) Cathelin, F. Blessures de guerre de la vessie. Lyon chirurgical, 1918, xv, No. 1, 109.
19) Legueu, F.: Des calculs vesicaux chez les blesses de la vessie. Bulletin de l’académie de médecine, Paris, December 5, 1916, lxxvi, 445.
(20) Duval, Pierre: Plaies thoraco-abdominales. Comptes rendus de la conférence chirurgicale interalliée pour l'étude des plaies de guerre, 3d session, November 5-8, 1917. Archives de médecine et de pharmacie militaires, Paris. 1918, lxix, 355.