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 XVIII



Battle-Incurred Compound Fractures About the Hip Joint

Marshall R. Urist, M. D.

Battle-incurred compound fractures about the hip joint have been a major problem and have been attended with almost equally unsatisfactory results in all the recorded wars of history. This is true of the Napoleonic Wars; the Crimean War; the War of the Rebellion; the Wild West era of American development, which had all the effect of warfare; the Anglo-Boer War; the Spanish-American War; the Russo-Japanese War; World War I; and the Spanish Civil War. The records show that in all of these conflicts, compound fractures produced by war wounds in the region of the hip joint were more often septic, were more frequently associated with systemic infection, and healed more slowly and more unsatisfactorily than other joint injuries. They also had the poorest functional results and had the highest case fatality rate of all joint injuries.

The background of this chapter is a detailed analysis of 29 intensively treated, personally observed battle-incurred fractures in the region of the hip joint (fig. 59). These 29 fractures were selected from 154 similar injuries which occurred in a total of approximately 25,000 (chiefly battle incurred) orthopedic injuries treated in the 18 general hospitals which made up the 802d Hospital Center in the United Kingdom Base between May 1944 and July 1945. The reason for the selection of these 29 cases was threefold: (1) The patients were all treated intensively; (2) they were all personally observed and, because they were all classified as nontransportable, the period of observation was unusually long, ranging from 12 to 20 weeks; and (3) they were all followed up for periods varying from 6 months to 2 years. From this selected sample, it is possible to determine the special problems inherent in the reaction of the hip joint to injury and to evaluate the methods of treatment employed against methods previously employed and those recently proposed. The analysis of these cases is shown in the following tabulation:















Superficial debridement



Debridement and arthrotomy


1The intracapsular classification includes involvement of the head of the femur in 20 cases; of the neck of the femur in 5 cases; and destruction of articular surfaces in 18 cases.


FIGURE 59.-Semidiagrammatic showing of distribution of 29 battle-incurred compound fractures of hip joint. Circles represent locations of main fractures. Numbered circles represent fractures described in detail in case histories in text. Since each fracture was the result of a missile injury, and since missile fractures are practically always comminuted, each circle may be considered the point from which a stellate lesion radiated to involve inferior and superior segments of the bone. Circles also, however, indicate the points of maximum displacement of the fractures.





Suppurative arthritis



Incision and drainage



Excision head of femur









Pulmonary embolism


Deaths (both from homologous serum hepatitis)


Known end results



Less than 25-percent disability



Disability moderate or not yet determined



Disability more than 50 percent



Death from sepsis and undetermined other causes after 2 years


Other bone and joint injuries



Septic fractures


Sciatic nerve injuries


Laceration femoral vein


Chest injuries


Intra-abdominal injuries



Gastrointestinal tract





Genitourinary tract injuries


Nonorthopedic operations












Incision-drainage septic wounds


2Known end results include 15 cases in which avascular necrosis of the femoral head occurred.
3Including 6 major operations on the gastrointestinal tract.

Although the immediate results of wounds of the hip joint were no more encouraging in the early days of World War II than they had been in any other recent war, the situation had changed in one important respect: These wounds are not lethal in themselves, but they are frequently associated with other serious wounds, and in former wars the men who sustained them succumbed in large numbers on the battlefield or in forward areas. In World War II, many of them survived serious abdominal, thoracic, and other wounds which, of course, took precedence of bone injuries, and the fractures about the hip could therefore be treated. This was because of the speed of first aid; the competent abdominal, thoracic, and other urgent surgery performed in forward hospitals soon after wounding; the liberal use of whole blood and blood substitutes; the wise use of chemotherapeutic and antibiotic agents not available in earlier wars; and the excellent supervisory care along the whole chain of evacuation. Orthopedic surgeons in World War II were thus confronted with a challenge of almost unique magnitude, as compared with the responsibilities which had confronted specialists in this field in earlier wars.

Emergency Measures

Transportation of the wounded soldier from the battlefield to the hospital was more rapid, more efficient, and more humane in World War II than in


any other recorded conflict. The splints provided for transportation of casualties with injuries about the hip joint were modified from those employed in World War I. The most popular was the Keller-Blake half-ring modification of the Thomas splint, in which the leg is maintained in full extension. The traction obtained by this method was helpful in the control of pain, but adequate immobilization of the hip joint was not secured unless the torso was also bandaged to the litter. Immobilization of the torso is inherent in the use of the Liston and Hayes splints, which do not seem to have been employed on the battlefield in World War II, but the application of bandages, as just mentioned, was equally effective when the Keller-Blake modification of the Thomas splint was used, as it was in 19 of these 29 patients. The other 10 were transported in improvised splints, including splints made from rifles. One rifle was applied from the axilla to the lower extremity and the other from the groin to the foot. Both were held in place by bandages.

Patients strapped to litters for transportation in ambulances or airplanes automatically obtained good immobilization and efficient splinting for short periods of time between medical installations. Medical-aid men had been thoroughly schooled in the dictum that a soldier with a fresh fracture should be moved about or otherwise disturbed only when it was absolutely necessary. This practice alone probably saved as many lives as splints saved in World War II. Its importance was apparently not realized in World War I, when the Thomas splint was considered to be lifesaving.

Morphine, plasma, whole blood, and careful avoidance of any measures that would produce disturbance and pain at the fracture site reduced the incidence of shock originating on the battlefield in injuries about the hip joint and reduced the incidence of death from that cause before casualties reached a forward (field) hospital to a level not much higher than obtains in ordinary civilian automobile accidents. Unless a very large artery had been lacerated, pressure bandages were sufficient to control hemorrhage for several hours. During the time these particular men were treated, penicillin was available and was used routinely, usually with a sulfonamide.

With these exceptions, methods of treatment of injuries of the hip joint were much the same as those described in American, English, and German records of World War I. These records, unfortunately, were not made generally available in World War II, nor was their value emphasized, with the result that many of the errors made in the First World War were repeated in the Second. The experience in these 29 cases is typical in that respect, and their record may prevent similar errors of omission and commission in some future conflict.

Initial Wound Surgery

A directive from The Surgeon General's Office,1 based on the accumulated experiences of World War I and experiences in World War II prior to D-day

1War Department Technical Bulletin (TB MED) 147, March 1945.


in the European Theater of Operations, provided for the management of wounds of the hip joint by debridement and early arthrotomy whenever circumstances permitted. Because of the more destructive character of the weapons used in World War II, extensive debridement was even more necessary than it had been in World War I. Wounds caused by land mines, hand grenades, shell casings, and particles of disrupted bullets might seem insignificant on the body surface, but underneath there would almost invariably be found extensive lacerations of the muscles and other deep tissues, and numerous blood clots.

In the presence of these injuries, it had to be assumed that bacterial contamination had occurred. The fact that bullets and some shell fragments had penetrated the body at white heat did not warrant the assumption that they were sterile. The phenomenon of the ricochet and the low velocity at which fragments of mine casings frequently entered the body made it inevitable that they would carry with them soil and fragments of clothing. This was not always evident when the foreign bodies were first lifted out of the tissues. When, however, blood clot, old fibrin, friable granulating tissue, or ensheathing scar tissue was soaked off, it was usually possible to identify wisps of woolen clothing, water-repellent raincoat material, and other fabrics which had not been apparent on the first inspection. Many observers thought the explanation of the apparently greater incidence of wound infection in winter was the greater amount of clothing worn in cold weather and the greater difficulties of personal hygiene under combat conditions. There seems little doubt that the original damage to soft parts, as well as the later ravages of infection, was almost directly proportional to the volume of cold steel, stone, wood, and other foreign material, including excreta inevitably present on clothing worn in combat, and to their retention in the tissues.

Fractures of the bony structures were the result not only of the force of the direct impact of the missile but also of its blast effect as it passed through the limb. The disorganization and disruption of the skeletal structure which followed, and the cutting effects of sharp fragments of bone, caused quite as much damage to muscles, nerves, blood vessels, and viscera as did the missile itself. Finally, foreign bodies which might have passed out of soft tissues were frequently retained in the hip joint because this pelvifemoral structure was powerful enough to arrest them.

All of these considerations explain why, in wounds of the hip joint as in other combat wounds, the first principle of management, regardless of location, was a bold incision. The excision of macerated tissue and retained foreign bodies was equally important; both served as excellent culture media for infection.

Primary arthrotomy, which is simply an extension of debridement to the joint, was regarded as a radical measure by many observers in World War I and by some surgeons in World War II. Pool,2 however, in reporting to The

2Pool, Eugene H.: Wounds of Joints. In The Medical Department of the United States Army in The World War. Washington: U. S. Government Printing Office, 1927, vol. XI, pt. 1, pp. 317-341.


Surgeon General on its use in World War I, termed it a "conservative" operation. A comparison of his report with reports by British,3 French,4 and German5 surgeons makes it apparent that all of them followed much the same policies and practices and that apparent differences can usually be explained by differences in nomenclature. The French, for instance, advocated what they termed extensive primary resection of the joint as soon as possible after wounding. What it amounted to was a particularly extensive debridement.

The basic principles of arthrotomy and joint debridement may therefore be said to represent the evolution of a large experience, accumulated over the same period of time by many different surgeons working under much the same circumstances in several different theaters of war. Furthermore, the criticisms of arthrotomy voiced by such experienced military surgeons as Jolly,6 Trueta,7 Wheeler,8 and Levit9 were all expressed prior to the introduction of penicillin and the subsequent decrease in the surgical risk.

Analysis of Cases

It is not evident from any report in the literature that arthrotomy, per se, has ever been responsible for the introduction of infection into the joint. That generalization holds for the 9 cases in this series in which it was employed. On the other hand, sepsis is frequent when primary arthrotomy is omitted. It was present in 9 of the 20 cases in this series in which the joint was not opened at operation, and Ellis10 reported that it occurred in more than half of the cases in his series under the same circumstances. The figures suggest that whatever may be the merits of a watching and waiting policy in some injuries, it is not sound in wounds of the hip joint.

The only possible exception to the routine application of early debridement and arthrotomy was the type of case described by Pool11 in the medical history of World War I, in which a high-velocity bullet perforates the hip joint cleanly, leaving punctate wounds of entrance and exit. In 2 such cases in this series, arthrotomy was omitted and only debridement was done. There were no septic complications, it is true, but it should also be pointed out that bone damage was much less severe than in the other cases.

In all, 20 of the 29 injuries of the hip joint in this series were treated initially by so-called conservative measures, consisting of cleansing the wound with soap,

3Frankau, C. H. S.: Gunshot Wounds of the Joints. In Great Britain War Office, Medical Services. General History, Surgery of The War, History of The Great War Based on Official Documents. London: His Majesty's Stationery Office, 1922, vol. 2, pp. 297-325.
4Fruchaud, H.: Wounds of the Joints. Lancet 2: 235-238, 19 August 1944.
5Franz, Carl: Lehrbuch der Kriegschirurgie. Verletzungen des Hüftgelenks. Berlin: Springer-Verlag, 3d ed., 1942, pp. 303-307.
6Jolly, Douglas W.: Field Surgery in Total War. New York: Paul B. Hoeber, Inc., 1941, pp. 123-128.
7Trueta, José: The Principles and Practice of War Surgery With Reference to the Biological Method of the Treatment of War Wounds and Fractures. St. Louis: The C. V. Mosby Co., 1943, pp. 170-173; 214-226; 352-355.
8Wheeler, W. I. de Courcy: Wounds Involving the Hip Joint. In Bailey, Hamilton: Surgery of Modern Warfare. Edinburgh: E. and S. Livingstone, 1st ed., 1941, pp. 474-480. 3d ed., 1944, pp. 541-549.
9Levit, V. C.: Concerning Gunshot Wounds of Joints and Their Treatment. Khirurgiya 14: 3-8, 1944. Abstracted in Bull. War Med. 5: 303-304, January 1945.
10Ellis, J. S.: Wounds in Region of Hip Joint. Lancet 2: 490-492, 20 October 1945.
11See footnote 2, p. 225.


water, and antiseptics; more or less complete debridement of the damaged soft parts, and final irrigation of the wound, without exposure of the joint or arthrotomy. The other 9 cases, as already noted, were treated within 12 hours of wounding by debridement of the soft parts, incision of the joint capsule, and removal of readily available foreign bodies and unattached fragments of bone and cartilage. The joint capsule was left open in 8 of the 9 cases, and the soft-tissue wounds were left open in all 29 cases.

Complete debridement was recognized, theoretically, as the ideal primary treatment for wounds of the hip joint, but there were many reasons why it could not always be applied:

1. Patients brought off the battlefield were frequently in shock and could not have withstood such extensive surgery as debridement and arthrotomy implied in wounds of the hip joint.

2. The injuries were often so extensive that complete debridement was not possible. The aim of early surgical treatment was to obtain gross mechanical cleanliness. To achieve it, wide excision of tissues was frequently necessary because small fragments of wood or metal, gritty substances, or bits of stone were so adherent that they could not be removed manually or by irrigation; sometimes their complete excision was impossible.

3. If injuries to bowel, bladder, or other structures were present and offered an immediate threat to life, their treatment had to take priority over treatment of a wound which threatened merely the vitality of a limb.

4. The tactical situation was often such that time-consuming operations on a major joint were not practical within an optimum, or even a reasonable time after wounding.

5. Surgery of the hip joint is a complex, difficult, and highly specialized procedure. Surgeons experienced in it and competent to undertake it were always in short supply and were seldom available in forward areas. Debridement of the soft tissues could be performed by inexperienced surgeons, but they could not undertake arthrotomy and debridement of the joint.

Nonetheless, the conservative measures employed in many cases in this series were deliberately undertaken, preliminary to a policy of so-called expectant observation or masterly inactivity. This was in direct conflict with instructions from the Office of The Surgeon General.12 With all due regard for the difficulties of complete debridement and arthrotomy, token treatment of surface wounds in fractures of the hip joint in which thorough debridement was impossible still could not be condoned. The least that should have been done was to remove accessible foreign bodies, together with fragments of tissue crushed and deprived of circulation. When ideal therapy proved impossible, the least that should have been done in any case in which sepsis was clearly inevitable or seemed likely was to employ a bold incision and saucerize the wound. When infection was not present or was not thought a possibility, the local forces of nature usually proved remarkably adequate to prevent it in the young, healthy men who made up the casualties of war. When the circulation

12See footnote 1, p. 224.


in the hip joint and the capacity for repair seemed insufficient, debridement should have included removal of parts whose death or loss might reasonably have been anticipated. These simple principles were violated in a number of cases in this series.

Case Histories

Poor results did not always follow inadequate debridement, as the first and second of the following case histories show. The poor results which may follow an inadequate procedure are, however, well illustrated in the third case.

Case 1.-An infantryman was shot in the right hip by a rifle bullet which entered the anteromedial aspect of the joint, ripped away a segment of the inferior margin of the epiphyseal line of the femoral head, and lodged in the tissues adjacent to the greater trochanter (fig. 59). After superficial debridement of the wound in a field hospital, the patient was transported in a plaster-of -paris hip spica to an evacuation hospital, which he reached 4 days after injury. At this time, there was considerable local tenderness, but no local or constitutional signs of infection were evident.

When the patient reached a general hospital in the United Kingdom Base, skeletal traction was instituted by means of a Kirschner wire at the level of the adductor tubercle of the femur in a Thomas splint with Pierson attachment, with 30-degree abduction, 150-degree extension, and 5-degree internal rotation in a Balkan frame. On the 16th day after wounding, exploration was carried out through a lateral incision; the gluteus maximus was split at right angles to the line of the skin incision. The bullet was located by palpation of the muscle bellies and was removed. It lay in a pool of purulent fluid, and the adjacent tissues were discolored. The wound was deepened to the level of the greater trochanter and was debrided of all contused, necrotic tissue. It was left open, and closure was not undertaken until the 15th postoperative day. At this time, the wound was clinically clean and healing was uneventful. A smear from the fluid surrounding the bullet had showed typical purulent exudate, but routine cultures were sterile.

At the end of 4 weeks, the patient was removed from skeletal traction and put up in Buck's extension. Active exercises were carried out in it for the next 4 weeks. Eight weeks after wounding, roentgenograms showed no evidence of avascular necrosis. Eight weeks later, there was moderate limitation of rotation in both directions, 15-degree abduction, 15-degree adduction, and from 18- to 60-degree flexion. The patient was in excellent condition when he was evacuated to the Zone of Interior 2 weeks later. He was eager to begin to walk, but weight bearing was not permitted until 6 months after injury. At the end of the year, there was no appreciable disability.

Case 2.-An airborne infantryman was injured by a machinegun bullet which caused a compound fracture in the intertrochanteric region, with fissure fractures of the neck and head of the femur (figs. 59 and 60). The wounds were thoroughly debrided in a field hospital a few hours later, and the bullet was removed from the substance of the muscular attachments on the greater trochanter. Although the capsule was torn and frayed anteriorly at the base of the neck of the femur, the joint was not opened. Supportive therapy, including transfusions, was required for 4 days after operation.

On the 11th postoperative day the patient was transported in a plaster-of-paris hip spica to a general hospital in the United Kingdom. His temperature had been normal for the preceding 6 days. Roentgenologic examination at this time showed an extreme coxa vara deformity. The fracture was manipulated on the fracture table, and the wound, which was clinically clean, was closed by delayed primary suture. The hip was suspended in skeletal traction in a Thomas splint, with the ischial half ring inverted and with a Pierson attachment. Excellent reduction was obtained.

Eight weeks after wounding, the fracture was united, and the wound was well healed. Four weeks later the patient was evacuated to the Zone of Interior in a double hip spica with


FIGURE 60 (case 2).-Roentgenogram showing compound intertrochanteric fracture. High-velocity machinegun bullet is seen in posterior soft parts. (Unless otherwise noted, all roentgenograms were taken within a few hours of wounding. This film and those reproduced in figures 66 and 67 were made with field equipment, under difficult conditions, which accounts for their poor quality.)

abduction, internal rotation, and flexion at the hip. At 9 months, he was reported to be walking in an ischial non-weight-bearing caliper brace, with 50 percent of the normal range of motion of the hip.

Case 3.-An infantryman was wounded in the abdomen by an artillery-shell fragment which perforated the pelvis and the right upper hip joint and emerged on the lateral aspect of the thigh. The upper end of the femur was shattered (figs. 59 and 61). Exploration of the abdomen at a field hospital a few hours later, after resuscitative measures which included a transfusion of 500 cc. of whole blood, revealed multiple lacerations of the large and small intestine. Enteroenterostomy and jejunostomy were performed. A 2-inch wound on the lateral aspect of the right thigh, through which the shell fragment had emerged, was enlarged to 7 inches, and debridement was done through it, but the joint was not opened. The saucerized wound was dressed with vaseline gauze, and a hip spica was applied.

Up to the 10th day, convalescence was uneventful, but in the hospital train, en route to the coast of France, a foul odor began to issue from the dressings, and the temperature rose to 101° F.

The hip spica was removed at a general hospital in England, and bilateral vertical skeletal traction was applied (fig. 62). The wound sloughed, and purulent material drained copiously from it. Cultures taken from the depths of the wound revealed a mixture of organisms, including both anaerobic and aerobic forms of streptococci and bacilli. In spite of systemic and local penicillin therapy (250 units per cubic centimeter of physiologic salt solution) drainage continued for 49 days. Closure of the wound was attempted at this time but was only partly successful, and a draining sinus persisted for several months longer. Over this whole period, there were daily temperature elevations to 99° and 100° F.


FIGURE 61 (case 3).-Roentgenogram showing extensive compound fracture of hip joint involving acetabulum and head of femur. Roentgenogram also shows shell fragments which produced the damage. This picture was made after fractures had been reduced and immobilized for 10 weeks in vertical skeletal traction.

FIGURE 62.-Vertical skeletal traction with overhead balanced suspension in compound fracture of hip joint. Note that patient can lift himself on bedpan and that active and passive exercise of knee joints is possible. This form of traction and suspension could therefore be maintained for 4 to 5 weeks without much danger of development of contractures.


Although the jejunostomy functioned well, the patient lost weight steadily. Laboratory studies showed a secondary anemia and slight hypoproteinemia. Supplementary feedings of a high-protein, high-caloric, high-vitamin diet were followed by a slow gain in weight.

At the end of 10 weeks, vertical skeletal traction was discontinued, and the patient was put up in suspension in a Rouvillois splint (fig. 63), with 45-degree flexion and 20-degree

FIGURE 63.-A. Use of Rouvillois lumbofemoral splint with skeletal traction and unilateral suspension in compound fracture of neck of femur. Note that patient can lift himself from bed and exercise while in this apparatus. B. Use of Rouvillois splint with bilateral suspension in bilateral fractures of pelvis and hips.


abduction. At this time, the fracture was in excellent position, and some motion could be obtained in the joint under anesthesia. At the end of 6 months, the roentgenograms showed a typical picture of avascular necrosis, with ossification of the superior capsule. Clinically, the joint was ankylosed and painful, and disability was almost complete.

Surgical Approach to the Hip Joint

In 2 of the 9 cases in this series in which arthrotomy was done, the approach was by a posterior (Kocher) incision, which was merely an enlargement of the original wound. In another instance, the approach was by a lateral (Watson-Jones) incision. In the other 6 cases it was by an anterior or Smith-Petersen incision. This approach, which was widely popular in the United States even before World War II, consists of anterior iliofemoral arthrotomy, with exarticulation of the head of the femur. In an occasional case, when the traumatic wound is very large, it is good judgment, as in this series, to use any other standard incision, such as the Kocher or the Watson-Jones incision, which best exploits the exposure already present. With these exceptions, the anterior iliofemoral route is the most expeditious of all approaches.

The Smith-Petersen incision has a number of advantages. It is easily modified. In 2 of the cases in which it was used in this series, the original technique was followed, except for omission of tenotomy of the rectus femoris. The incision was particularly useful in both cases because it permitted debridement and irrigation of the acetabulum under direct vision, which would not have been possible with any other approach. In both of these cases, elongation of the wound and development of the iliac portion of the approach, as in arthroplasty, provided sufficient exposure to permit exarticulation and resection of the head of the femur, including part of the acetabulum. In another case, L-shaped incision of the capsule and removal of numerous fragments of the head of the femur resulted in sufficient relaxation to permit exarticulation and irrigation of the joint and to make excision of the anterior ligaments unnecessary. It was possible, as the next step, to close the joint and instill penicillin. Another advantage of the Smith-Petersen approach is that the exposure can be made to correspond with the degree of damage to the joint. In 3 cases in this series in which there was a fracture of the neck or only minor damage to the circumference of the head of the femur, exposure of the joint in the sulcus between the sartorius and rectus femoris permitted thorough debridement without exarticulation.

The following case history illustrates the advantages of the Smith-Petersen approach:

Case 4.-When an infantryman was examined at an evacuation hospital a few hours after he had sustained a perforating wound of the left hip from a machinegun bullet, a small puncture wound was found on the anterior aspect of the upper portion of the injured thigh. The wound of exit, 2 inches long, was in the left buttock. Roentgenograms made under field conditions showed a stellate fracture of the neck of the femur which extended into the head and split the upper portion of the greater trochanter (fig. 59). After resuscitative


measures, including a transfusion of 500 cc. of whole blood, the joint was exposed by the anterior iliofemoral approach and was thoroughly cleansed of all loose bone and blood clots by debridement and irrigation. The posterior wound was enlarged to 3 inches and was carried down to the bone by splitting the muscle along the track of the missile. Closure of the joint capsule could not be effected. Penicillin solution (10,000 units) was injected into and around the articular cavity. The fractures were reduced by manipulation, and the hip was temporarily immobilized by skin traction in a Thomas splint.

Seven days later, the patient was evacuated to a general hospital in the United Kingdom in a plaster hip spica, in Whitman's position. Three days later, the wounds were closed by delayed primary suture, and the patient was suspended in skeletal traction in a lumbofemoral Rouvillois splint, in which he remained for the next 8 weeks. At the end of this time, there was partial union. The patient was evacuated to the Zone of Interior 3 weeks later (11 weeks after wounding) in Whitman's position. There the hip was again suspended, this time in a Thomas splint with Pierson attachment. An ischial caliper non-weight-bearing brace was fitted at 20 weeks. At 10 months, although the fracture was still incompletely united, there was no limitation of adduction or abduction, and 90-degree flexion was possible. A McMurray osteotomy was performed at the end of a year because of insufficient bone structure in the neck of the femur.

Management of Retained Foreign Bodies

The foreign bodies encountered in these 29 injuries about the hip joint consisted of fragments of steel from hand grenades, high-velocity bullets, casings from high-explosive artillery shells and land mines, and similar objects. In 9 cases the hip joint was perforated, and in 2 instances infection followed. In 1 of these cases the bullet passed through the bladder, did great damage to the head of the femur, and left a track of metallic dust behind it.

In the other 20 cases, 1 or more fragments penetrated the hip joint and lodged in it or about it. In 6 of these cases, no attempt at removal was made. In 9 cases, complete removal of the foreign bodies was possible; infection followed in 1 instance. Infection also followed 5 operations in which only partial removal of the foreign bodies was accomplished.

Operations for removal of foreign bodies from the hip joint were undertaken only through standard surgical approaches. These permitted sufficient visualization of vital nerves and blood vessels in the region to prevent further trauma. Blind extraction of a sharp missile from its bed was extremely dangerous and was not attempted. In actual practice, the most efficient method of detecting foreign bodies was found to be by digital exploration, oriented by true anteroposterior and lateral roentgenograms. Two gloves were always worn, because of the risk of tearing the glove on the sharp fragment.

In 4 cases, the foreign bodies were found at operation in lakes of green, purulent fluid. In 2 such instances, the fluid was sterile on culture. In the other 2 cases, various organisms were identified, including Bacillus proteus and Bacterium coli. One of these 4 patients developed fulminating pyoarthrosis. The other 3 recovered uneventfully, and wound healing was satisfactory in each instance.


Wound Closure

Closure of the wound of the joint was always the desideratum after arthrotomy, but in most fractures of the neck or head of the femur the ligaments were so badly damaged that complete closure of the capsule was usually impossible.

When the incision was anterior, closure of the wound of the soft tissues offered no difficulty; the muscles fell together with remarkable ease. When there had been considerable loss of substance in the soft parts, it was the practice to remove various amounts of iliac crest with a sharp osteotome, to obtain close approximation of bone and soft parts, and to eliminate dead spaces or close the incision without tension, according to the circumstances of the individual case. Black silk, which was used for deep stay sutures, was tied loosely, so as not to interfere with drainage of the wound along the natural anatomic planes. Closure of the skin was delayed for 5 to 7 days, until the critical period of possible infection had passed.

Delayed primary closure of the soft-tissue wound was carried out successfully in 9 clean cases in this series and was partly successful in 2 others. The operations were usually done within 4 to 10 days after wounding. Drainage was not employed in any of these cases. Nine other clean cases which were closed by the same plan healed by secondary intention or were treated by skin grafting. In the remaining 9 cases, wound closure was never seriously considered, because infection was present or was obviously impending when the patients first came under observation. In several of these cases, multiple foreign bodies were retained; delayed primary closure was, quite correctly, never employed in such cases.

Adjunct Fracture Management

Postoperative transportation splinting-Twenty-seven of the twenty-nine casualties, after surgical treatment in the field or evacuation hospitals, were transported to rear installations in plaster-of-paris double hip spicas, with the plaster carried only to the knee on the uninjured side. The other 2 patients were moved in Tobruk splints.

The orthopedic surgeons in general hospitals who treated the 154 patients with compound fractures of the hip joint from which these 29 cases were selected for detailed analysis had no doubt of the relative advantages of these 2 splints. Almost without exception, they agreed that the men were more comfortable and their injuries were more adequately immobilized in the properly padded full plaster spica than in the Tobruk splint or any of its modifications. It is true that the initial application of a hip spica was slightly more troublesome and somewhat more time consuming than the application of the Tobruk splint, but the hip spica was the least difficult of all splints to maintain, and in the end time was saved by its use. Extension in the Tobruk splint was advantageous when there were no other wounds or when the wounds were so located that the padded bar of the splint could be placed against the ischial


tuberosity, but this advantage was more than counterbalanced by the ineffectiveness of the traction thus obtained, as well as by the rapidity with which whatever traction had been achieved was dissipated during transportation.

Ellis'13 suggestion that a Thomas splint be incorporated in the hip spica was made independently by several United States Army surgeons in evacuation hospitals about the time that he advanced it. This technique was employed in 2 cases (not included in the 29 analyzed in this chapter) but it proved no more effective than the original Tobruk splint.

Skeletal traction-Two patients in this series were retained in plaster spicas without traction. Two others were treated in extension, one by Russell's and the other by Buck's apparatus. The other 25 were treated by skeletal traction. This was a larger number of cases treated by this technique than had been recorded in any other single series of combat-incurred fractures of the hip joint up to the end of World War II.

Reduction of fractures of the hip can be obtained more easily, at least by comparison, in combat-incurred injuries than in similar civilian-type fractures, because of the soft consistency and flexibility of fractures resulting from combat. Maintenance of fragments in position, however, requires some form of continuous traction. The proper combination of manipulation and traction is essential for good results. Neither one of these methods is sufficient in itself. In this series, the advantages of combined manipulation and traction were clearly evident. The combination provided comfortable immobilization, while at the same time it permitted maximum exposure of the wounds for dressing and other treatment. It kept the joint surfaces separated while they were healing. It held the reduced fractures in proper position in cases which had originally showed extreme coxa vara. Finally, and by no means least important, it facilitated nursing care.

The advantages of traction for compound fractures of the hip joint were recognized in World War I, just as in World War II, though not all of the methods proposed were efficient. Wheeler14 in 1944 and Ellis15 in 1945 suggested that the Jones abduction frame, which Frankau16 had recommended in World War I, be used when patients could not tolerate countertraction on the ischial tuberosities or about the gluteal or perineal regions. This apparatus (an upholstered table with leg splints attached for traction) was, however, cumbersome and not entirely efficient. It did not provide for flexion of the hip, which is essential in bilateral injuries, or for exposure of wounds of the buttocks or sacrum, which are frequently associated with hip injuries. Efficient nursing care was also difficult in this apparatus. Some of these difficulties had been solved in World War I17 by the use of the pelvic elevator, as suggested by the United States Navy Medical Corps, but the weights and pulleys required were also cumbersome and difficult to handle.

13See footnote 10, p. 226.
14See footnote 8, p. 226.
15See footnote 10, p. 226.
16See footnote 3, p. 226.
17See footnote 2, p. 225.


The newer techniques of suspension used in overseas hospitals in World War II proved far more efficient than any of the methods employed in World War I. Among them were the following:

1. Vertical traction (fig. 62). Although this is not a universally applicable method, it proved more efficient than the Thomas splint in every case in which it could be used.

2. The hip spica. This was the most efficient method for the immediate postoperative management of suppurating hip joints with missing bone substance. It was equally efficient as a transportation splint for patients to be returned to the Zone of Interior.

3. Pelvic suspension. This technique, which was devised by German surgeons,18 was carried out by means of wires placed through the anterior superior spines of the pubis. It could be used alone or in combination with vertical skeletal traction. American orthopedic surgeons did not use it widely.

4. The Rouvillois splint.19 This splint (fig. 63), which had been introduced in 1925, was standard equipment in the French Army. Considerable quantities were captured by the Germans at the fall of France in 1940, but they apparently did not use it, as large numbers were found by American forces going through France in 1944. A supply was issued to the 22d General Hospital, by the senior consultant in orthopedic surgery, for preliminary trial, and the splint soon became very popular among the surgeons who had access to it, though it remained unknown to other surgeons until the end of the war. This was unfortunate, for it provided more comfort for the patient than any other method of traction currently in use, and it also simplified nursing care. Not a single instance of sacral decubitus ulcer was observed in any case in which this method was used.

The Rouvillois splint (fig. 63) is essentially a modification of the Blake splint. It consists of a metal frame in which the leg and thigh are suspended on slings and attached to which is another large curved metal frame which extends anteriorly over the lower torso. This second frame holds the attachments of the supporting hammock placed under the lumbar spine. The weight of the frame, aided by the hammock, provides the necessary countertraction. Universal joints at the knee and hip permit any angle of flexion, abduction, and rotation. The whole apparatus can be suspended in balanced traction, and the region of the hip and sacrum can swing freely, without pressure on the bed at any point. The patient can also exercise in this splint.

5. The revolving orthopedic frame. This frame, which was devised at the 22d General Hospital in World War II,20 is basically a Bradford frame revolved by mechanical means. One of its advantages was that it could easily be constructed from salvage material available in any Army ordnance depot. This frame solved many of the problems which were encountered

18Westhues, H.: Die knöcherne Halbschwebelagerung des Beckens. Chirurg. 14: 489-493, 1942.
19Rouvillois, M. H.: Appareils universels ŕ suspension pour le traitement des fractures du membre inférieur. Bull. et mém. Soc. nat. de chir. 51: 846-849, 1925.
20Urist, M. R.; Maxwell, E. A.; and Ferguson, R.: Revolving Orthopedic Frames Made From Salvaged Material. Bull. U. S. Army M. Dept. 6: 628-630, November 1946.


FIGURE 64.-Active exercises in bed 8 weeks after open reduction and internal fixation of pertrochanteric fracture of femur. Old scars of bullet wounds of entrance and exit can be seen, as well as healed incision of lateral approach to upper femur between these wounds.

when fractures of the hip joint were associated with wounds of the intestine which required colostomy and wounds of the bladder which required cystostomy. It greatly simplified nursing care in such cases and was particularly useful when patients with severe, complicated injuries of the hip joint had to be transported in plaster casts for long periods of time.

Early motion.-Moderate passive exercises could be carried out while the patient was in suspension and were usually begun during the first 8 weeks after injury (fig. 64). Quadriceps-knee exercises were always supervised by physical therapists.

Active exercises were seldom possible or practical during the first 12 weeks of healing in compound fractures of the head or neck of the femur. They were, however, instituted guardedly and gradually in 2 cases in this series within this period, with very good results. One of these cases (case 1) has been described in detail elsewhere in this chapter.

Adjunct Therapy

Supportive therapy-The constitutional and nutritional status of many patients in this series, especially those with pyoarthrosis, was extremely poor and furnished extremely serious problems of management. Eight of the 9 septic patients became greatly emaciated within 4 to 8 weeks after wounding,


their weight loss being progressive. Many of the septic patients, as well as a number of others, presented serious secondary anemia, and most of the 29 patients had low plasma protein levels.

Men with extensive wounds were routinely given supplementary feedings of high-caloric, high-protein substances in addition to their regular meals. They also received supplementary vitamins and iron salts. The correction of anemia in the septic patients was more of a problem. Some of them had already had so many transfusions that matching them with donors was difficult when more blood was needed to fortify them for additional surgery on the joint. In a number of instances in which multiple small transfusions would have been highly desirable, this plan, for the reason just stated, could not be employed.

Chemotherapy and antibiotic therapy-Every patient in this series was treated with penicillin and sulfadiazine for at least 10 days from the day of injury. In many instances, chemotherapy was continued for weeks, the duration of treatment depending upon the temperature response. The early use of chemotherapeutic agents constituted one of the essential differences between the management of casualties with injuries of the hip joint in World War I and their management in World War II. Chemotherapy proved an effective measure, though the reliance apparently placed upon penicillin in the early days of its availability, when it was sometimes regarded as a possible substitute for surgery, was never justified. It should have been evident from the beginning that damaged bony tissues lack circulation to distribute a drug given systemically and lack interstices for its diffusion when it is given locally.

When therapy was systemic, as it usually was, penicillin was given in doses of 20,000 units at 4-hour intervals and sulfadiazine in doses of 1 gm. by the same schedule. Chemotherapy was used locally as a routine measure in the early phases of American participation in World War II and was occasionally employed by this route almost until the end of hostilities. Penicillin was used locally much less frequently.

Fourteen of the twenty-nine patients in this series, chiefly those injured in France within the first weeks after D-day, received local sulfonamide therapy, crystalline sulfanilamide being implanted in the wound in doses of 2 to 5 gm. In 5 instances, the sulfanilamide (10 gm.) was combined with 20,000 to 50,000 units of penicillin. In 2 other cases, penicillin solution (20,000 units) was injected into the joint after arthrotomy.

Seven of the nine septic cases were treated by one or another of these methods. There is no reason to believe that the infection was modified by their use in any instance. There is also no reason to believe that local chemotherapy had anything to do with the absence of sepsis in the 20 nonseptic cases in the series.

Complications of Hip Injuries

Suppurative Arthritis

Suppurative arthritis was a major problem in wounds of the hip joint in World War II, just as it had been in previous wars, but the incidence was much


lower than it had been in the War of the Rebellion21 and in World War I.22 The fearful infections so commonly observed in military hospitals during those wars were seldom observed in World War II, probably as the result of the almost routine use of the sulfonamides and penicillin from the time the patient was injured until his wound was completely healed. In the 154 wounds of the hip joint observed between D-day and V-E Day at the 802d Hospital Center, there was not a single instance of the type of rapidly spreading hemolytic streptococcic or staphylococcic infection described by Frankau23 in World War I and characterized by hyperpyrexia, tachycardia, and septicemia. Infections of this sort were usually promptly fatal.

The incidence of infection in this series, 9 of 29 injuries, is slightly less than that reported in similar series by Ellis24 and by Franz.25 There was 1 instance of infection in the 5 fractures involving the neck of the femur and 8 instances in the 20 cases in which the femoral head was fractured or completely exploded. Most writers, unfortunately, have failed to specify the region of the fracture in reports of the development of infection, though this is an important consideration. The femoral head has low vitality and high susceptibility to infection, and injury to it alone may largely determine the end result.

The infection was mixed in all 9 cases in this series. Bacterium coli and Bacillus proteus predominated in 5 cases and could be cultured from the deep spongiosa of the femoral head in a sixth case.

Experience proved that it was extremely important to keep close watch on the charts of patients receiving chemotherapy, to pick up low-grade fevers, anorexia, and other evidences of infection which might be masked by the drugs being administered. In 1 case in this group, aspiration, and later incision, revealed pyoarthrosis and a gluteal abscess, and, in another, pyoarthrosis was associated with a pelvic abscess. In both cases, the clinical localizing signs-rubor, calor, and dolor, as well as the usual systemic responses to infection-were strangely obscure. The explanation probably is that the more virulent organisms originally present had been controlled by penicillin and sulfadiazine, while the organisms undoubtedly introduced by fecal contamination on the buttocks and the clothing, although they did not produce the usual acute manifestations of suppurative arthritis, were not suppressed by chemotherapy and flourished in the deep necrotic tissues.

Suppurative arthritis developed in all 7 cases in which intestinal wounds were also present, just as in 6 similar cases observed by Collom and Hampton26 in the Mediterranean theater. In 2 cases, the anterior wounds of the hip

21Otis, George A., and Huntington, D. L.: Wounds and Injuries of the Hip Joint. In The Medical and Surgical History of the War of the Rebellion. Washington: U. S. Government Printing Office, 1883, vol. 2, pt. III, pp. 61-168.
22See footnote 2, p. 225.
23See footnote 3, p. 226.
24See footnote 10, p. 226.
25Franz, C.: Zur Frage der Gelenkresektionen. Deut. Militärarzt. 7: 266-272, April 1942. Abstracted in Bull. War Med. 3: 322-324, February 1943.
26Collom and Hampton. Unpublished data.


joint were only a few inches from the colostomy stoma, and in spite of the utmost care in dressing the intestinal wounds, as well as repeated attempts to isolate them with collodion gauze and adhesive tape, contamination of the wounds of the hip joint was repeated and constant. The juxtaposition of the intestinal wound to the hip joint was so obvious a source of infection that it was an error, as the following case shows, not to place the colostomy well up on the abdominal wall, even if it had been necessary to create it in a more proximal portion of the bowel.

Case 5.-A paratrooper suffered lacerated wounds of the right lower abdomen and a penetrating wound of the right hip joint from high-explosive shell fragment. Debridement, removal of a large intra-articular foreign body, and cecostomy were carried out shortly afterward at a field hospital. The hip was put in a spica. During the first week after wounding, the patient received 2,000 cc. of whole blood and 7 units of plasma. Although every effort was made to prevent contamination from the colostomy, the hip spica became soiled. On the seventh day, the temperature rose to 102° F., and the hip wound began to discharge large quantities of pus with an unmistakably fecal odor.

When the patient was received by air in the United Kingdom Base on the 11th day after wounding, the original cast was replaced by a fresh, reenforced, well-padded hip spica which was constructed with fenestrations lined with waterproofed material to permit dressing of the hip wound and the cecostomy. The temperature continued to rise daily to 102° F., and, by the 18th day after wounding, there had been a considerable and progressive loss of weight and bedsores had appeared over the sacrum. By the 32d day, sloughing of the wound had resulted in the loss of so much substance on the anterior aspect of the hip joint that the fracture site was clearly exposed.

At this time, the patient was removed from the spica and put up in suspension in skeletal traction, by means of a Kirschner wire inserted through the tibial tubercle. On the 50th day, in an attempt to simplify the management of the hip wound, an abdominal surgeon closed the colostomy and partly repaired the wound of the abdominal wall. Evacuation of the bowels through the rectum after this operation eliminated many of the previous problems of nursing care. Supplementary high-protein, high-caloric feedings were instituted. On the 60th day, the hip wound was widely opened, and the entire anterior aspect of the joint was saucerized. Drainage continued to be copious. On the 63d day, the patient became severely jaundiced, and weight loss was even more rapid than it had been previously. Death occurred 12 days later, on the 75th day after wounding. Homologous serum hepatitis was regarded as the immediate cause. The necropsy also revealed, as frequently happened in battle-incurred injuries, that the bone was damaged much more extensively than the roentgenograms had indicated. The inferior portion of the fracture line was healing, but the femoral head was necrotic.

The only other death in the series, also in a serious infection of the hip joint, was also attributable to homologous serum hepatitis. This patient had been given a transfusion of dried plasma immediately after wounding, 3 months earlier.

A followup of the 29 cases in this series for a minimum of 6 months and usually for longer periods of time showed that, if joint infection were to occur, it would become evident within 12 to 16 weeks after wounding. Suppuration did not occur primarily in any case in this series during the period of hospitalization overseas, which usually was 3 months, if delayed closure had been accomplished successfully, though abscess formation was reported in all kinds of wounds as late as 4 or 5 months after the patient was evacuated to the United


States. Suppuration was probably also a hazard in closed wounds of the hip joint, but it must have been extremely infrequent, if it occurred at all. At any rate, the possibility was not regarded as a reason for disregarding the important surgical principle of delayed primary suture of war wounds.

The immediate results in these 29 cases, and the similar results in the (unpublished) series studied by Collom and Hampton,27 show clearly what can be expected when thorough debridement is omitted or delayed. Even when debridement was omitted for the valid reason that men who had been submitted to major intra-abdominal surgery appeared incapable of tolerating surgery of the hip joint immediately afterward, the omission was followed by sepsis in a large proportion of cases.

It was not possible to determine in advance whether a wound of the hip joint would be contaminated in the course of intestinal surgery. On the surface, there seemed no reason why contamination should occur at this time unless there was a direct communication, either external or intrapelvic, between the two wounds. The difficulties of preventing contamination afterward when colostomy had been done have already been described.

Methods of management of suppurative arthritis in World War II included incision and drainage, resection of the femoral head, and amputation or disarticulation.

Incision and drainage-Incision and drainage were carried out in 6 of the 9 septic cases, through a liberal anterior or anterolateral incision, or through both approaches. The wound was kept open by Penrose tubing, rubber drains, or vaseline-impregnated gauze applied to various aspects of the drainage tract. Posterior incision, which was attempted in 1 additional case, proved just as inadequate in World War II, in which it was supplemented by chemotherapy, as it had proved in World War I. In this case, the tract closed within a week, as the result of contraction of the gluteal muscles, but the joint went on to destruction, just as in the World War I experience, and a second operation was necessary to excise the head of the femur.

The following case history illustrates the stormy course which a patient with suppurative arthritis following inadequate debridement was likely to pursue.

Case 6.-An infantryman sustained penetrating wounds of the pelvis and left hip (fig. 59) from a high-explosive shell. Shortly afterward, a portion of the ileum was resected, with end-to-end anastomosis; a large laceration of the rectum was repaired; and a Mikulicz sigmoidostomy was performed. The man was in such poor condition at the end of these procedures that the wound in the left hip was merely cleansed with soap, water, and antiseptic. Other injuries included a large lacerated wound of the left calf, fractures of the right elbow, forearm, wrist, and hand, and minor lacerated wounds of the right upper extremity.

Roentgenograms (fig. 65) taken at the field hospital showed a fragment of shell, approximately 1 cm. in each dimension, lodged squarely in the middle of the body of the ischium at the point at which it forms the acetabulum. The assumption was that the fragment had penetrated the joint from the pelvis outward after lacerating the upper rectum.



FIGURE 65 (case 6).-Roentgenogram showing fragment of 88-mm. artillery shell lodged in left anteroinferior acetabulum after traversing lower abdomen, penetrating rectum, and penetrating body of ischium. Septic arthritis, necrosis of head of femur, and luxation of hip joint followed within 3 months.

It was not possible at this time to determine which of four wounds of the hip and sacrum led down to the hip joint.

At the end of 3 weeks, the patient was nourishing well and had maintained a reasonable weight. At this time, he was evacuated to the United Kingdom, where the sigmoidostomy was closed 8 weeks later.

The management of the wounds of the hip had been unavoidable in the circumstances, and the results, as had been expected, were not satisfactory. There was no gross evidence of abscess formation, but a continuous discharge of seropurulent fluid occurred. After the patient had been evacuated from the Continent, he had considerable pain in the hip, which was intensified when he moved about in bed. The temperature rose to 101° and 102° F. each afternoon, and he was obviously toxic. Roentgenograms showed that the joint margins were becoming irregular and obscure. Drainage was instituted on the 33d day after wounding, through a posterior incision; 100 cc. of greenish-white, purulent fluid was evacuated. Culture revealed mixed organisms, chiefly Bacterium coli and Bacillus proteus.

Seven weeks after wounding, roentgenograms showed resorption of a large part of the surface of the femoral head, with erosion of the superior portion of the acetabulum. The head was beginning to rise out of the joint. By the 18th week after wounding, drainage had ceased, and all the wounds in the region of the hip had healed by secondary intention. Roentgenograms at this time showed further resorption of the femoral head, with definite subluxation. During the maneuvers necessary for securing the last set of films, the foreign body, which had been present in the hip joint in all previous roentgenograms, apparently migrated into the pelvis from its original bed. Active motion consisted of 50 degrees of flexion and 15 degrees of abduction and adduction, with limitation of rotation in both directions.


At the end of 6 months, because of complications related to the abdominal wounds and the additional surgery which they had required, this patient was still on full bed rest. He had received sulfadiazine for 3 weeks and penicillin for 8 weeks after wounding.

Resection of the femoral head-Resection of the femoral head was performed in 3 cases of suppurative arthritis, including 1 case in which incision and drainage through a posterior incision had failed to effect a cure. The operation was performed through an anterior incision in 2 cases and through a posterior incision, derived from enlargement of the original wound, in the third. Opinions differed as to the relative value of these two incisions in military surgery. In World War I,28 United States Army surgeons preferred the posterior route, while British29 and Continental30 surgeons preferred the anterior incision. In World War II, posterolateral and lateral incisions were also used for resection of the femoral head.

This operation was first performed for pyoarthrosis during the wars of the 19th century. It has naturally become a safer procedure as time has passed, but even when the case fatality rate was prohibitively high, there was universal agreement as to its value. Its prompt employment is amply justified for a number of reasons: Whether or not reconstructive procedures are contemplated the femoral head, when it is badly damaged, is a functionless structure and a nidus of infection, as well as a poor aid for joint fusion. After it is eliminated, drainage ceases and, when the greater trochanter has healed in traction, a functional articulation remains, which permits weight bearing and has a useful range of motion. There is resultant shortening, it is true, but this is easily remedied by an elevation of suitable height attached to the shoe. It is also true that for young men the result may not be as satisfactory in some ways as fusion of the hip joint, but it is still desirable to give resection of the femoral head a fair trial, preferably for a year or longer. If at the end of that period the outcome is not satisfactory, arthrodesis can then be performed.

In 1943, during the course of the war, Girdlestone31 described a technique for pyoarthrosis which permits saucerization of the entire hip joint on the lateral aspect by resection of the greater trochanter and the head of the femur, with minimum interference with muscle and minimum risk to the sciatic nerve and femoral blood vessels. Wheeler32 and Ellis33 both reported excellent results with it, and it is perhaps unfortunate that American surgeons did not give this method more recognition and a wider trial. It seems the logical procedure in cases in which the hip joint and gluteal region are so peppered with foreign bodies that their removal is impossible and there is nothing to do but wait for exfoliation of the infected areas. Girdlestone's operation also

28See footnote 2, p. 225.
29See footnote 3, p. 226.
30(1) See footnote 25, p. 239. (2) Auvray, M.: The Immediate Results of Surgical Intervention in 111 Cases of Purulent Arthritis of the Large Articulations. Bull. et mém. Soc. nat. de chir. 43:683-690, 1917. In Abstracts of War Surgery; An Abstract of War Literature of General Surgery That Has Been Published Since the Declaration of War in 1914. Prepared by the Division of Surgery, The Surgeon General's Office. St. Louis: The C. V. Mosby Co., 1918, pp. 304-305.
31Girdlestone, G. R.: Acute Pyogenic Arthritis of the Hip. An Operation Giving Free Access and Effective Drainage. Lancet 1: 419-421, 3 April 1943.
32See footnote 8, p. 226.
33See footnote 10, p. 226.


has other advantages. For one thing, it circumvents pressure by the weight of the body on the bed, which is a problem when posterior incisions are used. For another, it prevents the puddling of exudates frequently observed when anterior incisions are used. When it is employed in cases in which both the hip joint and the sciatic nerve have been destroyed by infection, it really amounts to a first-stage disarticulation of the hip joint.

Surprisingly good results, as the following case shows, are sometimes secured with resection of the femoral head.

Case 7.-An infantryman sustained a perforating wound in the region of the right hip (fig. 59) from a bullet which entered the inferior medial aspect of the right buttock and made its exit on the upper anterolateral aspect of the right thigh. Roentgenograms taken at a field hospital 3 hours later showed a comminuted fracture of the hip joint, with displacement of bone into the anterior muscles and soft parts. The wound was cleansed superficially with antiseptics, and arthrotomy was omitted.

Three days later, the wounds were debrided superficially at a general hospital in Belgium. Before the patient was evacuated to the United Kingdom Base by air, his temperature had risen to 101.6° F. (it had been 100.8° F. when he was first seen), and he had begun to complain of pain in the hip and the inner aspect of the thigh and knee. At the general hospital, a diagnosis of "acute catarrhal asthmatic bronchitis" was recorded.

The patient continued to feel and look very ill, and the temperature elevations persisted. On the 10th day after wounding, a tender, swollen, fluctuant area in the right buttock was incised, by enlarging the original wound, and about 75 cc. of greenish-white pus was evacuated. Culture showed a mixture of organisms, predominantly Bacillus proteus and Bacterium coli. Skeletal traction was instituted by means of a Kirschner wire inserted through the tibial tubercle, and the extremity was put up in a Rouvillois splint. On the 15th day after injury, the temperature was normal; the chest was clear; and the patient felt comfortable. Sulfadiazine was discontinued on the 20th day after wounding, but penicillin was continued through the 5th week.

Drainage continued to be profuse through the eighth week. Then it began to diminish, and for several days the temperature again rose to 101.8° F. daily. There was no tenderness on palpation of the buttock, but deep, rather indefinite fluctuation could be made out. When a probe was introduced into the wound, it went directly down to the bone. No active motion was possible in the joint, and attempts at passive motion, which were accompanied by considerable pain, resulted in only slight flexion.

On the 50th day after injury, exploration was carried out through an anterior iliofemoral incision. The head of the femur was found shattered into multiple fragments by fracture lines, but the pieces were well glued together and the head could be removed en masse. Several fragments of bone were found in small pools of pus in the substance of the adductor muscles anterior and medial to the joint.

Purulent drainage continued for another 3 months through the posterior incision. At the end of 6 months, the wounds were completely healed. The patient was wearing a nonweight-bearing caliper brace and had 30 percent motion in various directions in the hip joint. At the end of 28 months after the last operation, he was able to walk with a cane and had a useful range of motion. Although he suffered from some pain and instability, he preferred his present status to the possible improvement which might be accomplished by arthrodesis.

Amputation (disarticulation)-Neither amputation nor disarticulation was performed in any case in this series. Amputation was considered in 1 case, in which a series of pelvic abscesses with coxarectal fistulas developed after performation of the hip joint and rectum, but excisional surgery proved ade-


quate. In another case, however, in which the followup was indirect, the patient died 2 years after injury, of chronic sepsis and other causes. According to a local newspaper, he had had 40 operations in various hospitals in the Zone of Interior.

It may be that this man would have fared better if amputation had been performed promptly after injury. It is easy to understand the desire to be conservative in such cases, particularly if the limb distal to the injured joint is intact. There are, however, a number of arguments against this policy:

1. When there has been great mechanical destruction of the bone and soft parts and when retained foreign bodies carrying fragments of clothing cannot be removed, foci of infection are maintained for indefinite periods of time.

2. A prolonged delay before amputation merely results in exhaustion of the patient, so that, when the operation is eventually performed, it often poses a serious threat to life. Successful results are sometimes attained with late amputation, but they are by no means the rule. It must be assumed that patients with large areas of mixed, penicillin-resistant infection deteriorate every day that they live and that their chances of survival after major surgery become progressively less as time passes. All the experience of World War II was to the effect that penicillin can prevent septicemia and perhaps limit local extension of sepsis, but that it has no appreciable effect in a persistent mixed infection sown in the substance of tissue and associated with massive necrosis. Toxins are dispersed through the body, regardless of chemotherapeutic measures, and cachexia is the inevitable result and is very often fatal.

3. Amyloidosis has been reported in destructive injuries of the hip joint in which infection persisted even though drainage was satisfactory. In these cases, the patients were thought to be living in symbiosis with, or even to be slowly conquering, the local and general infection by virtue of their own immune resources. The assumption was unwarranted.

4. Observation of numerous instances of pyoarthrosis of the hip joint at United States Army amputation centers made it clear that when the sciatic nerve is lacerated the indication for early disarticulation of the hip is particularly strong.

Avascular Necrosis

Avascular necrosis is a possibility in every instance of battle-incurred injury of the hip joint. When the substance of the head of the femur has suffered a direct impact fracture, the irregularities in the bone structure and alterations in bony density suggest that this process is developing in this area, exactly as it may develop at the line of fracture of cancellous bone anywhere else in the skeleton. The difference between the reaction in the head of the femur and in joints with less intra-articular bone is quantitative, not qualitative. Another obvious reason for the appearance of avascular necrosis in injuries of the hip joint is anatomic: in adults this joint has insufficient collateral circulation in the femoral head, if indeed it has any at all, and the injury may destroy its end arteries temporarily or permanently.


Avascular necrosis occurred in 15 cases in this series. This number includes almost every instance in which the head of the femur was struck by a bullet or shell fragment, as well as some cases in which the damage was chiefly to the neck of the femur. In one or two of these cases, it seemed reasonable to speculate that both the bony injury and the loss of circulation were the result of blast or concussion from a bullet which passed through the limb.

The roentgenologic picture of avascular necrosis varied. Subchondral fractures and undisplaced fractures were sometimes so neatly contained in position in the acetabulum that they did not show up on routine roentgenograms and became apparent only at arthrotomy and debridement, or, even later, when avascular necrosis developed. In 3 cases in which the fractures were limited to the inferior portion of the femoral head, roentgenograms showed no evidence of this change until between 6 and 12 months after injury. Roentgenograms, when the condition became evident, always revealed rarefaction, resorption, and sclerosis of the head of the femur. When the fracture involved the femoral neck, stellate, incomplete fracture lines extended to the head of the bone. With healing, the roentgenograms showed increased density or patchy sclerosis. Since all of these patients were receiving penicillin in large dosages, it was often difficult to exclude the possibility of previous or controlled infection in certain cases in which resorption of parts of the femoral head had occurred. One point of distinction was that when there was clinical evidence of established suppuration, the bone substance was absorbed much more rapidly and more extensively than in clean cases. In the latter type of case, the pattern of bone rarefaction more nearly corresponded to the typical picture of creeping replacement seen in civilian fractures of the neck of the femur. In clinically clean cases, these changes appeared late. When infection was present, on the contrary, bone destruction was evident within a few weeks after injury.

In several cases, suppurative arthritis was followed by spontaneous subluxation or resorption of almost the entire femoral head. These patients were treated early, in skeletal traction, on the theory that this method supplied mechanical protection to the necrotic head during the active stages of infection. Although traction was maintained as long as possible during the period of healing and creeping replacement, the joint surfaces were almost invariably destroyed or irreversibly damaged within a period of a few weeks.

When the infection was mixed, particularly when B. coli was present in fractures associated with extensive necrosis of the entire head of the femur, the constitutional reaction was extremely severe, and the patients (as in case 5) deteriorated rapidly.

The following case is an excellent illustration of avascular necrosis in a compound fracture of the hip joint.

Case 8.-An infantryman, hit by a high-velocity bullet, sustained a fracture of the neck of the femur, with fissures into the head of the bone (fig. 59). The bullet entered on the lateral aspect of the left hip, perforating the joint, ischium, and pubic bones, and tearing the bladder and prostate. It then made its exit laterally below the greater trochanter on the right. The bladder was repaired and the space of Retzius drained in a field hospital


FIGURE 66 (case 8).-Roentgenogram [retouched] showing fracture of neck of femur extending into head, with associated fractures of ischial and pubic rami and coxa vara deformity.

on the same day. The wound of the hip was treated by debridement, and a hip spica was applied, in which the patient was evacuated to a general hospital, in the United Kingdom on the fifth day. Progress had been satisfactory, except for daily afternoon temperature elevations to 100.8° F.

Roentgenologic examination at the hospital in England showed a fracture of the base of the femoral neck extending up into the head, with coxa vara deformity (fig. 66). The fracture was reduced on the sixth day after injury by manipulation on a fracture table. It was then immobilized in skeletal traction with balanced suspension in the Rouvillois splint. The patient developed a urinary-tract infection which required drainage by an indwelling urethral catheter for 5 weeks. The hip joint showed no evidence of sepsis at any time and healing was satisfactory.

Roentgenologic examination 6 weeks after wounding showed the fracture of the neck of the femur in excellent position and beginning to heal. There was, however, a uniform increase in the density of the femoral head.

Two weeks later (8 weeks after injury), the patient was evacuated to the Zone of Interior in Whitman's position in a hip spica. At 6 months, he was walking in a non-weight-bearing ischial caliper brace. Two years later, although the head of the femur was nearly revitalized, the joint space was lost. The normal range of motion was limited by 50 percent, and the man was still on a total-disability pension.

Early Reconstructive Surgery

Attempts to salvage function in compound fractures of the femoral head by the use of reconstructive procedures soon after injury were not encouraged in military practice, because of the ever-present risk of actual or potential infection. In this type of case, avascular necrosis was frequent, and the tissues exposed had a minimum amount of resistance to infection. When fractures were limited to the trochanteric and subtrochanteric regions, conditions were different, and early open operation was regarded as justified if the associated


FIGURE 67 (case 9) .-A. Roentgenogram [retouched] showing subtrochanteric fracture of femur caused by accidental, self-inflicted pistol shot wound. B. Roentgenogram of subtrochanteric fracture of femur shown in figure 67A, 7 weeks after wounding, 6 weeks after successful delayed primary closure of wound, and 1 week after open reduction and internal fixation with McKee nail and single transfixing screw.

soft-tissue wounds had been successfully closed by delayed primary suture. In some cases in this series, it was possible to aline subtrochanteric fractures by internal rotation, 90-degree flexion, and wide abduction of the hip in skeletal traction. In other instances, however, the need for open reduction became clear soon after conservative therapy had been instituted.

This policy was followed in only 4 cases in this series, all of which were treated by internal fixation by means of a Smith-Petersen, a McKee, or a Liverpool nail 4 to 6 weeks after delayed primary suture of the wound. The McKee and Liverpool nails are basically Smith-Petersen nails with plates attached. In the Liverpool nail, ordinary plates are attached at the angle of the average neck of the femur. In the McKee nail, wide curved plates are similarly attached (fig. 67B).

In all 4 cases treated by nailing, extra-articular methods were used. When a plate was employed, it was attached at the same time, through a lateral incision. The results in these cases, in which weight bearing and approximately 90 percent of function were achieved within a year of wounding, encourage the attempt at early functional salvage by the use of reconstructive measures in carefully selected cases of fractures of the trochanteric and subtrochanteric regions.

The following case is an illustration of the good results of an early reconstructive procedure in a properly selected patient.

Case 9.-A soldier from an armored unit accidentally shot himself while returning his pistol to the holster. The bullet penetrated the anterior capsule of the left hip joint, fractur-


ing the femur below the lesser trochanter and lodging in the posterior margin of the fascia lata at the middle third of the thigh (figs. 59 and 67). Arthrotomy and debridement, with removal of the bullet, were performed in a field hospital a few hours later. Penicillin (10,000 units) was instilled into the joint at the conclusion of the operation. Five days later, delayed primary closure of the wound of the soft parts was carried out at a general hospital in the United Kingdom. The patient was then immobilized in skeletal traction in a Balkan frame by means of Kirschner wire through the tibial tubercle.

The fracture was reduced by manipulation twice within the next 3 weeks, but position could not be maintained for any length of time. Open reduction and internal fixation with the McKee nail (fig. 67B) were therefore carried out 7 weeks after wounding. Weight-bearing exercises were begun at 6 months. When the nail plate was removed a month later, the patient had 75 percent of normal motion in the joint. At 9 months, the fracture was solidly united, and disability was minimal.

An Ideal Plan of Management

The experience of World War I and World War II suggest the following routine as an ideal plan of management for battle-incurred injuries of the hip joint:

1. Initial surgery should consist of debridement, arthrotomy, exploration of the joint, irrigation if gross suppuration is not present, and primary closure of the joint if loss of substance does not prevent it.

2. The approach is preferably the anterior iliofemoral incision described by Smith-Petersen or the posterior approach described by Kocher.

3. The operation should be carried out as promptly as possible after wounding, but neither impending nor established sepsis contraindicates its delayed performance, regardless of the length of time since injury.

4. Debridement should consist of complete excision of blood clots, devitalized muscle, fatty tissue, bone fragments, foreign bodies, and other foreign material along the track of the wound. This is the general principle. In the special case, the surgeon must use his judgment in deciding how much chiseling, scraping, and curetting are necessary to secure clean, bleeding bone surfaces. In the hip joint almost more than in other areas, the wounds of the soft parts, the maceration of muscle, and the retention of shell-borne particles of clothing and other foreign material may be even more important than the articular damage. All tissue, including bone, which has no remaining circulation must be excised, to eliminate soil for infection.

5. The deeper soft parts should be closed over the joint, but the skin wound should be left open, to be closed within 4 to 10 days by delayed primary suture.

6. The extremity should be extended and suspended, preferably by skeletal traction in the Rouvillois splint, for 8 to 12 weeks.

7. Active motion should be begun in traction as early as this is practical and can be accomplished without undue risks or pain.

8. In cases in which sepsis is established when the patient is first seen, open operation by the Smith-Petersen approach, with thorough debridement, is the method of choice. If it fails, excision of the joint by Girdlestone's


method, combined with drainage or disarticulation, should be done. The operation should not be unduly delayed if sepsis is severe, since the patient's condition deteriorates rapidly in these circumstances.

9. Whole blood should be used in liberal amounts as long as it is indicated. Other supportive measures, including measures to correct protein deficits, should also be used according to the indications.

10. Penicillin and sulfadiazine should be used parenterally before and after surgery. Local chemotherapy is not indicated. Both drugs are adjuncts to surgery, not a substitute for it.

11. In occasional carefully selected cases, reconstructive procedures with internal fixation may be employed 4 to 6 weeks after successful delayed primary suture of the wound.