|OFFICE OF MEDICAL HISTORY AMEDD REGIMENT AMEDD MUSEUM|
HISTORY OF THE OFFICE OF MEDICAL HISTORY
Jeep Injuries of the Hip Joint
Marshall R. Urist, M. D.
This chapter is based on an analysis of 58 injuries of the hip joint originating under generally similar circumstances of military traffic. They were observed over a period of 18 months in the 802d Hospital Center in the European Theater of Operations, which, over the same period, admitted approximately 25,000 combat and noncombat casualties. Both before and after the institution of definitive treatment, these men, as is inevitable in time of war, were treated by various medical officers, who ranged from general practitioners trained in the emergency to handle military casualties to experienced orthopedic surgeons whose views and practices, however, were frequently at wide variance concerning the management of such injuries.
Although this series consists of only a small number of cases, they furnish an amply sufficient background for this chapter for several reasons:
1. This series includes as many instances of this type of injury as would usually be admitted to any large urban hospital over a period of 10 to 15 years.
2. No single surgeon in civilian life would be likely to have as large an experience as this in the course of a professional lifetime.
3. It is possible to derive from a detailed clinical analysis of these cases enough material for an evaluation of the teachings and practices concerning these injuries when the United States entered World War II. There was then no uniform teaching concerning them, and no Army directives for their management were issued.
4. Because of the many different medical officers who participated in one phase or another of the treatment of these casualties, the series is of historic interest and may be accepted as an overall picture of the management of traffic injuries of the hip in World War II.
For convenience of analysis, these 58 cases have been divided into 3 groups (tables 11 through 15), as follows: (1) Dislocations of the hip joint (15 cases); (2) fractures of the acetabulum without dislocations (16 cases); and (3) fracture-dislocations of the hip joint (27 cases). The dislocations, in effect, serve as a control group for the fracture-dislocations, while the fractures of the acetabulum similarly serve as a control group for the injuries to the articular surfaces in the group of fracture-dislocations.
These three groups of cases have been analyzed from a variety of points of
view, with particular emphasis on the nature of the traumatic lesion, methods of treatment, complications, and followup studies. Followup reports were obtained in various ways, including personal examinations, letters and roentgenograms submitted by patients and their local physicians, and reports of examinations in Army and veterans' hospitals.
The end results in these 58 military injuries were compared with the end results of 18 similar injuries sustained in civil life and treated at the Massachusetts General Hospital. The rarity of this type of injury in civilian practice is evident in the fact that these 18 cases represent a 12- to 16-year experience in a thousand-bed civilian hospital.
Circumstances of Injury
All of these casualties sustained their injuries in traffic accidents, and 40 of the 58 received them while they were riding in jeeps. Curiously, although jeeps were used in all theaters of operations, the impression was gained from communications with other orthopedic surgeons working in other theaters that this type of injury was frequent only in the European theater. This is perhaps to be explained by the excellent concrete highways which are a part of the European scene but which are less frequent or nonexistent in other parts of the world. These highways permitted and encouraged travel at high rates of speed and help to explain the high incidence of traffic injuries in this theater.
Other contributing factors were also operative. The excitement and recklessness of men at war, most of whom were young men, were responsible for a high incidence of all types of severe injury. In this series, one patient was 51 years of age, but all the others were between 19 and 33 years, the period of life in which trauma always exacts its greatest toll.
Soldiers were constantly involved in accidents in trucks, tanks, and other types of mechanized motor vehicles, under conditions which seemed just as hazardous as those which prevailed when they were riding in jeeps. Nonetheless, the fact that the jeep was the most commonly used of all Army vehicles cannot be overlooked as an explanation of the relatively high incidence of the special injuries to be discussed in this section, particularly the dislocations, which were present in 42 of the 58 cases.
The United States Army truck, one-fourth ton, more popularly called the jeep, seemed to possess special structural characteristics which put its riders into a favorable anatomic position for the production of this special type of injury. A review of recorded experimental studies, beginning with Allis'1 in 1896 and Bigelow's2 in 1900, shows that it provided the exact conditions necessary for the production of dislocations of the hip joint (fig. 68). It is constructed like an old-fashioned wagon, with a minimum of protective framework, and passengers readily fall out of it. Its seats are so low that when passengers are seated their hips are flexed more acutely than in other automobiles.
FIGURE 68.-Artist's interpretation of patients' positions and attitudes at time of injury, derived from their own descriptions. The accident was most often a head-on collision. A. A rider in this position could fall out, striking the pelvis and incurring a fracture-dislocation or a fracture. B. A rider sitting with legs abducted could sustain a fracture of the acetabulum or a fracture-dislocation from a dashboard blow on the knee. C. A rider sitting with hips acutely flexed and adducted could sustain a dislocation of the head of the femur with minimal damage to the bony rim of the acetabulum.
While the riders are in this position, the head of the femur is in internal rotation, directed posteriorly and disposed toward posterior dislocation. The instrument board, or the back of the front seat, depending upon where the rider is sitting, is so close to this flexed knees that in case of accident it is difficult to avoid the so-called dashboard type of blow, which has become a classical injury in this modern mechanized age. There seems no doubt, therefore, that the jeep, because of its special structure, must be assumed to be primarily responsible for the appearance of as many dislocations of the hip over an 18-month period in military experience as would occur over a 10-year period, or longer, in civilian practice.
Fractures and fracture-dislocations of the acetabulum cannot be properly discussed, any more than they can be properly evaluated and treated, without a recollection of certain anatomic considerations. The acetabulum, viewed from within the joint cavity, consists of a semilunar or horseshoe-shaped platform of bone covered with articular cartilage. This platform, which is the weight-bearing surface, is formed mainly from the body of the ischium, which receives the attachment of the posterior capsule of the hip joint on its outer surface, several millimeters from the edge and well separated from, and proximal to, the cotyloid ligament.
The anterior portion of the horseshoe, which is formed from the body of the pubis, is about half as wide as the superior portion. The capsule on the anterior, superior, and inferior aspects is attached close to the edge of the rim, in contrast to its attachment on the posterior aspect, and blends closely with the fibers of the cotyloid ligament. Only areolar and synovial tissue covers the central portion of the acetabulum.
The integrity of the posterior rim of the acetabulum is essential for the stability of the joint. The rim of the anterior portion may be excised, as in arthroplastic operations, without any serious harm to the weight-bearing function, but the superior shelving portion, which is the thickest part of the innominate bone, is also vital if function is to be preserved.
The retinacula are three flattened bands of loose fibrous tissue which are covered with synovial membrane and which lie partly within and partly outside the reflected capsule on the inferior, posteroinferior, and posterosuperior aspects of the joint. They contain many blood vessels and serve literally as aqueducts in that they carry the blood supply to the head of the femur from the numerous branches of the medial femoral circumflex vessels in the intertrochanteric region. They pass through the superior and inferior pads of fat at the margin of the articular cartilage and form a great many of the terminal arterial branches to the head of the femur. A plexus of veins which drain the superior capital epiphysis is also found in the retinacula. The clinical importance of the retinacula, which is considerable, is frequently overlooked in injuries of the hip joint (pp. 275, 293).
Nature of the Lesion
As has already been pointed out, the hips of the rider in a jeep are flexed more acutely than in any other type of car, because jeep seats are so low. In this position, the head of the femur is in internal rotation, directed posteriorly and disposed toward posterior dislocation. Trauma could be received by the greater trochanter or transmitted through the femur from the knee flexed against the dashboard, or from the foot on the floorboard, or could be received posteriorly and transmitted to the hip by the sacrum. Cases have been recorded of injuries received under all of these conditions. Specific inquiries into
FIGURE 69.-Typical posterior dislocation of hip as result of jeep accident. Note unequal length of legs, internal rotation of knee, and adduction of hip on affected side. Note also severe bruises in both prepatellar regions.
the circumstances of injury in this series showed that in many instances the head left the joint with minimal damage to the bony structure when the hip was adducted and flexed, as it is when a man is riding in a jeep, perhaps with his legs crossed. The result in these cases was an ordinary dislocation without a gross fracture (table 11).
Traumatic lesions of the soft parts were evident in many of these cases (fig. 69), but otherwise, since the method of treatment was always closed manipulation, there were no opportunities for direct observation of the damage caused to the joint structures by the dislocations. From the standpoint of accumulation of knowledge this was unfortunate. Observations in fresh injuries might have supplied information which would indicate why some dislocations are followed by avascular necrosis of the head of the femur while others are not.
Roentgenologic studies, however, supplied information which, combined with accumulated experience with experimental dislocations, seems significant and relevant. These observations were as follows:
1. Prereduction roentgenograms revealed serrations of the rim or avulsion-chip fractures in three cases (fig. 70, case 11; cases 2 and 6). The location of these injuries suggested that the joint capsule might have been torn away at its acetabular attachment. Avulsion fractures in the posterior intertrochanteric region in two other cases (fig. 71, case 7; case 9) indicated that the femoral attachment of the capsule was torn. Allis3 described both of these capsular lesions in his experimental study of dislocations, but further investigation is needed to clarify the pathologic process. The location of the capsular tear may play some role in the damage suffered by the local blood supply, especially by the branches of the medial circumflex femoral artery which enter the hip joint in the intertrochanteric area posteriorly, and may explain the later development of avascular necrosis of the bony structures.
TABLE 11.-Dislocations of hip joint without significant fractures
FIGURE 70 (case 11).-A. Emergency postero-oblique roentgenogram showing posterior dislocation of hip joint. Film is of average quality for roentgenograms made under field conditions. In these circumstances, bone detail was often lacking, and minor fractures were not suspected. B. Anteroposterior roentgenogram of posterior dislocation shown in figure 70A after manipulation and reduction of dislocation by Stimson's maneuver. Note improvement in quality of film (made in fixed hospital), which now reveals avulsion-chip fracture, probably of posteroinferior rim of acetabulum. The acetabular attachment of the joint capsule has presumably been torn away.
FIGURE 71 (case 7).-A. Emergency postero-oblique roentgenogram showing posterio dislocation of hip joint, with possible fracture of femur. B. Anteroposterior roentgenogram of posterior dislocation shown in figure 71A after manipulation and reduction by Stimson's maneuver. Improvement in the quality of the film now permits the demonstration of an avulsion fracture of the superior portion of the intertrochanteric ridge, in the line of the femoral attachment of the joint capsule posteriorly.
2. Prereduction and postreduction roentgenograms also revealed other minor fractures in the form of incomplete fissures of the posterior acetabular rim. There were, in fact, only 8 of the 42 dislocations and fracture-dislocations in which there was no roentgenologic evidence whatsoever of bony damage. Since roentgenograms of the hip joint usually show only gross defects, it seems reasonable to assume that even in these 8 cases there was some damage to the rim of the acetabulum and the head of the femur and to assume, further, that a fracture, which may be major or minor, is an almost invariable accompaniment of a dislocation of the hip joint.
Dollinger's4 description of the traumatic lesion in 12 instances of inveterate dislocation of the hip joint managed by surgery, although it was published in 1911, still remains the most complete on record. In these cases, tears were found in all the posterior muscles, including the quadratus femoris. A few other notes on findings at operation for irreducible dislocations and a few autopsy observations, all in the older literature, suggest that in these injuries the head of the femur most often escapes between the piriformis and obturator internus and then tears the fragile gemelli muscles. When it finally comes to rest, the more durable obturator tendons are stretched over and under the protruding femoral neck. Damage to the intertrochanteric attachment of the posterior capsule at the insertion of these muscles, the point at which the largest part of the blood supply enters the hip, seems never to have been described. Yet avulsion fractures are repeatedly observed in this region in postreduction roentgenograms and furnish proof that the damage exists and that the fractures should be searched for, by repeated roentgenograms, if they are not immediately evident.
Before the era of roentgenology, there were a number of elaborate pathologic classifications of traumatic dislocations of the hip, all based on the position of the head of the femur as determined by the position of the lower extremity. Clinical diagnosis was made on the same basis. The advent of roentgenologic examination made it clear that many of the types of injury described were chiefly theoretic possibilities, and that the only really important consideration is whether the dislocated head lies anterior or posterior to the acetabulum.
Posterior dislocations (Bigelow's regular dislocations) are the most frequent variety. They were present in 14 of the 15 cases in this series. When no fracture occurs in a posterior dislocation, the head of the femur probably does not displace farther than just over the rim of the acetabulum, which was its position in all 14 cases in this series. In the single anterior dislocation, the head rested opposite the obturator foramen, but its position could not be determined by rectal examination. There were no instances in the series of Bigelow's irregular dislocations in which the head of the femur is displaced into the scrotum or is found in other bizarre locations.
A case reported by Kleinberg5 in 1923 suggests that the excursion of the
head may be less than the length of the ligamentum teres, which may not be torn if the head escapes by way of the cotyloid notch. This possibility is easily verified by a fresh anatomic preparation of a hip joint after circumcision of the capsule. In most of the cases (fracture-dislocations) in this series which were examined at open operation, this was not true. The ligamentum teres was stretched and was partially torn if not completely severed.
There were no instances of bilateral dislocation of the hip in the series. There was also no apparent damage to the vascular structures of the joint capsule in any of the 15 cases. Theoretically, this variety of damage would occur only in dislocations and fracture-dislocations.
First aid and transportation.-First aid was well handled in all 15 cases. The clinical picture is so striking in this condition, and the deformity (fig. 69) is so obvious, that most of the dislocations were promptly diagnosed in battalion aid stations and field hospitals. In most instances, preventive treatment for shock was instituted, and the leg was bandaged to the litter or the patient was otherwise protected against further trauma until an installation in the rear was reached. In one or two cases, a Thomas splint was used for immobilization. Six patients had severe injuries, which of themselves required immobilization, of other parts of the ipsilateral extremity.
Reduction.-Although the records are not complete in all respects and it is not clear in some cases how many attempts at reduction were made or how long a time they took, considerable information is still available on these points. In 1 instance, spontaneous reduction occurred, without anesthesia; this patient, who had sustained injuries of all 4 extremities, was in shock. In 1 instance, the first attempt at reduction was unsuccessful, and a second surgeon later accomplished it. In 12 of the remaining 13 cases only Pentothal Sodium (thiopental sodium) was used for anesthesia, which suggests that reduction was accomplished without special difficulty, in spite of the relative inexperience of most of the surgeons who were obliged to undertake it.
In some cases, reduction was accomplished by gravity alone after the joint had been properly flexed and relaxation obtained under anesthesia. The records indicate that 5 posterior dislocations were successfully reduced at the first attempt by Stimson's maneuver, which was also employed secondarily in 2 other cases, after the Bigelow method of circumduction and the Allis supine method had failed. The technique was not difficult if Stimson's description was closely followed: nothing was done until the muscles were fully relaxed under anesthesia. Then, with the patient prone, the thigh was rocked slightly to and fro to accomplish gentle internal and external rotation. The weight of the hanging limb, supplemented by slight downward pressure on the upper posterior tibia, supplied the needed traction. In every instance in which this method was used, the head of the femur promptly slid down in the acetabulum.
The single anterior dislocation was easily reduced, with the patient supine, by the reverse circumduction method of Bigelow. The essential feature of the
circumduction technique is flexion of the hip to relax the Y-ligament, which then can be used as a fulcrum. Bigelow's post mortem studies showed that this ligament is usually intact in posterior dislocations. After his demonstration of this method, traction, which had been regularly employed up to that time, came to be regarded as unnecessary and actually harmful.
However it was brought about, dislocation of the hip joint furnished no particular difficulties in reduction if there were no obstacles in the pathway of the displaced femoral head and if the acetabulum had not been distorted. As this series demonstrates, Stimson's method, or some one of its variants, is the simplest and gentlest method of obtaining traction by manipulation. The satisfactory results achieved by it, as this series also demonstrates, suggest that inexperienced surgeons would always do well to use it in preference to any other technique. Experienced orthopedic surgeons naturally based their procedures in the special case on the clinical and roentgenologic findings and did not adhere to any standardized technique. The modifications of the Allis and Bigelow methods are not well known and most of them were not tested in World War II.
Immobilization-When these patients were received at general hospitals, at periods varying from a few hours to 10 days after injury and reduction of the dislocation, it was found that a variety of methods had been used to immobilize them during transportation from evacuation hospitals. Five, chiefly those who had suffered fractures and other injuries of distal joints of the injured extremity, were evacuated in plaster-of-paris hip spicas. One patient was transported with his legs bound together in a swathe. The others were not immobilized by any special methods but were required to maintain complete recumbency.
Definitive treatment-Treatment at general hospitals was limited to traction, in the form of (1) skin traction, accomplished by bilateral Buck's extension or Russell's suspension, or (2) vertical skeletal traction, accomplished by means of a Kirschner wire through the tibial tubercle, with 10 pounds' pull. Traction was continued from 4 to 8 weeks. The optimum time would seem to be 6 weeks, which is the period theoretically required for complete healing of the joint capsule and its ligaments. The wide variation in this series was not intentional but was unavoidable because of irregularities in the evacuation schedule caused, in turn, by the weather and the tactical situation in the theater. Every patient with a dislocation of the hip was a candidate for evacuation; treatment was necessary for at least a year after injury, and often longer, which far transcended the holding period permitted for the theater.
Traction was employed arbitrarily in the management of these cases because it seemed logical. In view of the complications which could arise from injury to the vascular structures of the joint and which, at least theoretically, could be avoided by maintenance of the circulation and healing of the blood vessels, rest in traction seemed highly desirable. The literature,
however, offers no evidence to support this point of view and supplies no control studies. At the start of the war, methods recommended for the management of dislocations of the hip varied widely, as follows: That the limb be immobilized in a hip spica for 2 to 4 weeks; that traction be applied for 2 to 4 weeks; and that the patient be kept at bed rest but otherwise be permitted complete freedom of motion. There is no evidence that any of these methods, including the techniques of traction used in this military series of injuries, is superior to any other method.
Weight bearing-All patients, when they were evacuated to the Zone of Interior, were instructed not to put weight on the injured extremity until at least 6 months after injury without specific permission from a medical officer who had examined recent roentgenograms. They were also strongly urged not to dispense with crutches until at least 6 months had passed. They were told, in very plain words, that failure to follow these instructions might be detrimental to recovery. In about half the cases, as might have been expected, weight bearing was practiced as early as 3 months after injury, either because of the patients' own restlessness or because of advice from medical officers in Zone of Interior hospitals. Only patients with associated fractures of the knee or the foot are known to have refrained from weight bearing for 6 months or longer.
Again, the literature was of no help in formulating policy. In the reported cases, weight bearing was arbitrarily forbidden for periods ranging from 6 weeks to 6 months. Neither the cases recorded in the literature nor any cases in this series suggest that early weight bearing was harmful or that abstinence from weight bearing contributed to recovery.
Recovery was uncomplicated in every case in this series, and there were no sequelae in the immediate period of observation. Avascular necrosis and degenerative arthritis appeared in a number of cases of fracture-dislocation of the hip joint (p. 293), and whether their absence in this group of dislocations is significant or accidental it is not possible to say. The literature suggests that avascular necrosis is almost as common in simple dislocations as in fracture-dislocations, but the presence of associated fractures is frequently not mentioned, and there is some doubt, therefore, as to the validity of the observation.
Certainly, as already intimated, the absence of complications in this series cannot be attributed to the method of treatment employed, which was entirely arbitrary. The literature suggests that when these complications, as well as ossification of the joint capsule, have developed, they are symptomatically aggravated by early weight bearing. No comment is possible from the standpoint of this series.
Restriction of weight bearing is generally believed to produce disuse atrophy of the base of the neck of the femur, which accentuates the density
of the dead parts of the head and therefore aids in earlier diagnosis of avascular necrosis. Again, no comment is possible from the standpoint of this series, since disuse atrophy was not evident in any case in it.
Fractures of the Acetabulum Without Dislocation
The 16 fractures of the acetabulum (table 12) which occurred in these 58 injuries of the hip joint fall into 3 classes:
1. Fracture of the rim of the acetabulum, 5 cases, including 3 of the superior and 2 of the posterior rim.
2. Central fracture, 8 cases, including 7 of the body of the pubis and inner table of the pelvis, and 1 of the anterior rim and superior ramus of the pubis, with intrapelvic protrusion of the head of the femur.
3. Comminuted or bursting fracture, with disorganization of the entire joint cavity, 3 cases.
This classification is practical, not theoretical, and is therefore not open to the objection that is attached to the purely morphologic classifications. It takes cognizance of the morbid and roentgenologic anatomy of the region, as well as the physiologic considerations having to do with the localization of function of various parts of the acetabulum (p. 254). It also takes cognizance of therapeutic problems.
The injury in every case was of the so-called dashboard type, with consequent incongruity of the joint surfaces. As is pointed out elsewhere (p. 276):
1. Fractures limited to the anterior portion of the articular surfaces are suitable for conservative management, in traction.
2. Fractures which destroy a significant part of the superior and posterior rims of the acetabulum require open operation.
3. More extensive fractures, which destroy the entire joint surface, do not respond well to any form of treatment and usually result in great disability, frequently with associated ankylosis and degenerative arthritis (cases 29 through 31).
Nature of the Lesion
Fractures of the posterior rim of the acetabulum, of which there were 2 in this series, are not uncommon. They vary in magnitude and in the particular portion of the lunate articulate surface involved. Fractures of the superior rim, of which there were 3 in this series, are extremely uncommon, presumably because of the volume of the bone and the adaptation of the stress lines of the trabeculae in this region to the burden of the weight of almost the entire torso. In 1 instance of superior rim fracture in this series (case 18, table 12), in which open operation was required, 0.5 cm. of the full width of the articular cartilage was attached to the fragment, while the entire iliofemoral ligament was attached to the outer cortex of its rim.
The 8 fractures of the central portion of the acetabulum showed an infinite number of patterns and degrees of displacement. The single instance of intrapelvic protrusion of the head of the femur (case 28) was, like most reported cases, only partial, which was fortunate, as the highly comminuted, bursting type of fracture is almost irreparable. In the majority of reported cases, as in this case, there was a fracture of the anterior rim and body of the pubis, associated with a second fracture of the superior ramus of the pubis. This fracture acted as a hinge and was responsible for the displacement of the inner table of the pubis. In such cases, the head of the femur escapes into the fracture site (that is, the anterior portion of the acetabulum) and the vital superior and posterior lunate cartilages sustain relatively little damage.
It might be expected that intrapelvic protrusion of the femur would occur in jeep accidents. The injury can be produced experimentally and accidentally by a blow on the flexed knee transmitted to the head of the femur, and it has been reported, by Haines,6 during convulsions from shock therapy, possibly as the result of indirect force produced by abnormal muscle action. Most often, however, it follows a blow on the greater trochanter. The degree of abduction of the limb at the time of the accident, as well as the amount of internal and external rotation of the head of the femur, determines the pattern of the lesion and the portion of the rim of the acetabulum involved. If the head is in external rotation, a blow on the greater trochanter would drive it into the anterior portion of the acetabulum, where, as in the case in this series, intrapelvic protrusion is possible with minimal damage to the weight-bearing, lunate-shaped articular cartilage.
In each of the 3 cases (cases 29, 30, and 31) in which comminuted fractures involved all portions of the acetabulum, the joint was completely disorganized and incongruent. In 2 of these cases, there was also an associated fracture of the ipsilateral femur.
Fractures of the rim of the acetabulum-The 5 patients with fractures of the rim of the acetabulum received no treatment in forward installations other than such first-aid measures as were indicated. All were immobilized for transportation in hip spicas, in 30-degree abduction, 20-degree flexion, and neutral rotation.
The treatment employed in the fixed hospital depended upon the degree of displacement of the fracture and the size of the fragment of rim. In every case, the diagnosis had been made by emergency roentgenograms taken in field or evacuation hospitals, but repeated examinations, with more elaborate equipment, revealed further details. Postero-oblique views were particularly informative. In 2 instances the fractures were found displaced. In 2 others there was either no displacement at all or very slight displacement. In the
TABLE 12.-Fractures of acetabulum without dislocation
1This patient had no treatment for 2 months, while
he was a prisoner of war.
FIGURE 72 (case 19).-A. Anteroposterior roentgenogram showing fracture of posterior rim of acetabulum shortly after injury. In this view, the displacement of the fracture does not seem significant. B. Postero-oblique roentgenogram of fracture shown in figure 72A, showing displacement. When the hip was examined in flexion in this case, the joint was stable, but little more than half of the original lunate cartilage of the acetabulum articulated with the head of the femur.
remaining case (fig. 72, case 19) the patient had been a prisoner of war, and the existence of the displacement was not realized until 6 weeks after injury because of failure of enemy surgeons to study the important postero-oblique roentgenogram.
Four of the 5 patients with fractures of the rim of the acetabulum (cases 16, 17, 19, and 20) were treated conservatively, in traction suspension, for periods varying from 3 to 8 weeks, the duration of traction depending upon the extent of the lesion. Operative treatment would have been desirable, because of the extent of the displacement, in 1 of these cases (case 20), but it was contraindicated because of serious head and chest injuries.
Open reduction and internal fixation were carried out primarily in the remaining case (fig. 73, case 18) because a vital portion of the acetabulum had been injured. The anterior iliofemoral approach, which gave excellent exposure, revealed the joint capsule in the sulcus between the sartorius, tensor fascia femoris, and rectus femoris. After irrigation and curettage, to remove the blood and blood clots which obscured the fracture lines, it was possible to return the main fragment to its normal anatomic position, with its attachments to the capsule and anterior iliofemoral ligament intact. The surgical approach did not permit the insertion of the transfixing screw, and a lateral stab wound was therefore made in the gluteus medius, large enough to permit the introduction of a long drill-point screw driver at right angles to the middle of the fracture line. The parts dovetailed perfectly, and the fragment could be
FIGURE 73 (case 18).-A. Anteroposterior roentgenogram showing fracture of supero-posterior rim of acetabulum before joint repair. B. Anteroposterior roentgenogram showing fracture shown in figure 73A a week after anatomic repair of joint surface and internal fixation of fracture. Note that screw penetrates inner table of pelvis; if it does not, immobilization of the fragment will not be reliable.
compressed into position so accurately that the fracture line was scarcely visible in the postoperative roentgenogram (fig. 73B). Since the superior rim is the strongest part of the acetabulum, fractures in it are correspondingly uncommon, and it is not surprising that no case similar to this one seems to be on record in the literature.
Central fractures of the acetabulum.-The 8 patients with central fractures of the acetabulum received only necessary first-aid measures in forward installations. All were immobilized in either a Thomas splint or a plaster spica, with the joint in neutral position. A second roentgenologic examination at the fixed hospital revealed that in 4 fractures there was no displacement. In 3, all involving the dome of the acetabulum, the inner table of the pelvis was displaced for a distance about equal to the width of the cortex. The remaining fracture (fig. 74, case 28) was, as already noted, a typical fracture of the anterior acetabulum and body of the pubis, with intrapelvic protrusion of the femoral head and wide displacement of the inner table of the pelvis.
All 8 patients in this group were treated conservatively, by suspension and skeletal traction, for 6 to 8 weeks. The fracture complicated by intrapelvic protrusion of the head of the femur required special management. Reduction was accomplished by means of a vertical Kirschner wire placed just inside the lateral cortex of the trochanter. The head of the femur was then restored to its normal relationship with the rim of the acetabulum by the use of the large-sized wire-tautening spreader. The hip was suspended in a Balkan frame with 5 pounds' lateral traction and 10 pounds' bilateral skin traction through the lower extremities. The ipsilateral side of the bed was elevated slightly to provide lateral countertraction. Suspension traction was maintained for 8
weeks. The failure of the fragments to follow the femoral head after its extraction (fig. 74B) was not unexpected; this observation has been made in almost every case recorded in the literature. The displacement of the inner table of the cortex was not modified by the technique employed, but the fracture site and the anterior portion of the joint had been filled in with abundant callus and healed with the formation of new bone and fibrocartilage before the patient was evacuated to the Zone of Interior, 4 months after wounding.
FIGURE 74 (case 28).-A. Anteroposterior roentgenogram showing central fracture of acetabulum with intrapelvic displacement of anterior acetabular rim and inner table of pelvis. Lowest arrow shows how the fracture of the superior ramus of the pubis acts as a hinge and causes displacement of the inner pelvic table. Other arrows indicate the main fragments. B. Anteroposterior roentgenogram of fracture shown in figure 74A after withdrawal of head of femur from fracture site by lateral skeletal traction through a vertical Kirschner wire in the greater trochanter. In this type of fracture, the main fragments of the fracture do not follow the femoral head.
Extensive comminuted fractures of the entire acetabulum.-The emergency management of the 3 patients with extensive comminuted fractures of the entire acetabulum (cases 29, 30, and 31) was limited to measures to forestall shock. They were evacuated from the combat zone as promptly as possible, with the limbs immobilized in hip spicas in slight abduction.
Definitive treatment was conservative, by traction, in all 3 cases. Skeletal traction by means of a Kirschner wire through the supracondylar region brought the fractures of the shaft of the femur into alinement but had little effect upon the fractures of the acetabulum. It was realistically assumed, from the time these patients were first examined, that arthroplasty or fusion of the joint would be necessary at a later date in Zone of Interior hospitals. In each case, at the time of the 2-year followup, the joint was ankylosed and painful.
The 27 cases of fracture-dislocation in these 58 chiefly jeep injuries of the hip joint fall into 3 groups when they are classified according to the standard textbook categories:
1. Fifteen cases of posterior dislocation, with fractures of the posterior rim of the acetabulum (table 13).
2. Eight cases of fracture-dislocation of various types, with the common denominator of irreducibility by closed manipulation (table 14).
3. Four cases of posterior dislocation associated with fracture of the head of the femur (table 15).
Nature of the Lesion
The necessity of performing open operation in 15 of the 27 fracture-dislocations in this group of cases provided opportunities for direct examination of the traumatic lesion from both the anterior and the posterior aspect of the hip joint. The description which follows is based upon the observations made at operation:
The major damage was sustained by the acetabulum, periarticular ligaments, tendons and muscles, ligamentum teres, and retinacula of the joint (figs. 75 and 76).
Damage to the acetabulum could be assumed to represent the first stage of a fracture-dislocation of the hip joint. It was particularly likely to occur if the hip was in abduction and the knee was flexed, so that the knee received the full force of the dashboard blow and transmitted it proximally to the acetabulum. The second stage of the injury, the excursion of the head of the femur, explained the tearing of the soft parts of the joint and the laceration of the periarticular structures.
Damage to the posterior rim of the acetabulum ranged in magnitude from insignificant bruises or chip fractures, which can be assumed to be present in practically all ordinary dislocations (p. 258), to fractures in which small fragments of the edge of the cotyloid labrum were separated and more extensive fractures, with disruption of large, crescent-shaped pieces of the articular cartilage and wide areas of the posterior and superior portions of the acetabulum associated with massive disruption of the whole width of the horseshoe-shaped articular cartilage. Since less than half of the head of the femur normally lies outside the bony acetabulum, the integrity of the posterior and superior portions of the rim appeared to be of increased importance to the stability of the hip after many of the supporting structures in the capsule had been ruptured.
Small fragments from the edge of the rim of the acetabulum usually contained little or no articular cartilage but were coated with cotyloid ligament on two sides. Frequently, they lay free in a hematoma in or around the joint cavity or were attached only by a few fibers of periosteum or ligament. Larger fragments, measuring a centimeter in diameter on one aspect or the
TABLE 13.-Fractures of posterior rim of acetabulum with reducible posterior dislocations of hip
TABLE 14.-Comminuted fractures of acetabulum with irreducible dislocations of hip joint
1Two unsuccessful closed manipulations.
TABLE 15.-Fractures of head of femur with posterior dislocation of hip joint
FIGURE 75.-Arthrotomy for fracture of posterior rim of acetabulum. Limits of exposure in posterior approach to injured joint. Sketch shows fracture in posterior rim of left acetabulum and posterior joint capsule. [From The Journal of Bone and Joint Surgery, vol. 30-A, pp. 699-727, July 1948.]
other, were usually attached to strips of capsule, and thin arcs of articular cartilage could be seen on their inner aspects. Only large fragments, approximately 4 to 7 cm. long, were broadly attached to the capsule. These bled on curettage and could be assumed to have sufficient blood supply to survive when they were replaced. The larger the fragment appeared in the roentgenogram, the more articular cartilage could one expect to find upon it at operation.
In addition to the changes listed, comminuted fractures of the rim of the acetabulum were always accompanied by extensive, contused, lacerated wounds of the posterior soft parts.
In almost every injury in this series, the parietal or external portion of the capsule, rather than the visceral or reflected portion, sustained the major damage. This is a significant observation, since it is the visceral portion of the
FIGURE 76.-A. Superficial anatomic relationships on posterior aspect of hip joint. The reflection of a segment of the trochanteric insertion of the gluteus maximus provides the relaxation necessary to develop a large window in this muscle. B. Demonstration of small, deep muscles and relationships of sciatic nerve. The piriformis, when reflected medially, envelops the sciatic nerve, and all further dissection is deep and lateral to the short external rotator muscles of the hip. C. Demonstration of skeletal relationships, fracture site, short external rotator muscles reflected medially, and rent frequently found after injury in inferoposterior portion of joint capsule. [Parts A and B, from Campbell, Willis C.: Operative Orthopedics, The C. V. Mosby Company, as redrawn and modified for The Journal of Bone and Joint Surgery, vol. 30-A, pp. 699-727, July 1948. Part C from The Journal of Bone and Joint Surgery, vol. 30-A, pp. 699-727, July 1948.]
capsule which carries the greater part of the blood supply to the head of the femur. When a large segment of bone was detached from the rim, the capsule was essentially intact except for a single tear in the posterior portion. The short external rotator muscles, however, sustained considerable damage; the belly and tendon of the piriformis were stretched over the head of the femur, which was dislocated into the fracture site. The tendon of the obturator internus was stretched over the neck. The gemelli were torn and shredded. All of these structures were so infiltrated with blood and so matted together with fibrinous exudate and granulation tissue when the patients were seen at general hospitals that dissection was difficult, if more than 2 weeks had elapsed since injury.
The head of the femur could be examined thoroughly at operation by rotating it through its range of motion in all directions at the fracture site. Almost every inspection revealed some damage to the articular surface of the head, though many of the lesions were small and not visible roentgenographically. Some fractures of even greater magnitude had not been demonstrated in anteroposterior views alone and, of course, were not demonstrated in films made by routine or imperfect techniques.
In 5 cases in which the rim had been fractured, lunate-shaped indentations were seen in the anterior aspect of the head of the femur and presumably represented cleavages caused by the edge of the rim. In a subchondral compression fracture, a circular, flat depression 2 cm. in diameter was noted on the anterior aspect of the femoral head, almost as if it had been tapped by a mallet. The surface of the bruised cartilage was dotted with spots the size of a pinhead and of uniform distribution, which represented the tufts of granulation tissue known to form early in the process of repair.
In 3 cases in which the posterior portion of the acetabulum had been removed, the ligamentum teres was found stretched and frayed. Complete visualization was possible in only 2 of these, in both of which disarticulation of the hip was necessary. In both of the cases, the ligamentum teres had been torn across. In the third case, in which complete visualization was not possible, it may have been only stretched. In many instances, the excursion of the femoral head was not as great as the roentgenograms had suggested.
Blood clot and organizing hemorrhage were observed in the retinacula in 2 cases, in 1 of which this structure was completely infiltrated with a large, subsynovial, dark-red blood clot. The significance of these findings was not realized until a year later, when the records were reviewed in preparation for this presentation, and the roentgenograms in both cases showed avascular necrosis of the head of the femur.
In distinct contrast to dislocations without fractures, which responded readily to conservative measures, the management of fracture-dislocations was often complicated and difficult. The dislocation presented one set of problems and the fracture another. Each required different management. The presence
of the fracture always prevented reduction of the dislocation. When the dislocation was reduced by closed manipulations, the fracture remained displaced. If the fracture was extensive, open reduction and internal fixation were required before an acceptable repair could be secured. Obviously, the end results in these cases of fracture-dislocation could not be expected to be better than the end results of dislocations without fractures or of fractures without dislocations.
Some cases in each category of fracture-dislocations were treated conservatively and some by open operation, the decision depending upon (1) the surgeon's clinical judgment and experience, (2) his evaluation of the individual cases, (3) the physical and tactical circumstances in which treatment had to be carried out, and (4) the results obtained from conservative treatment, which was usually, though not invariably, given a trial. As experience was accumulated, the decision to resort to surgical treatment promptly in certain types of fractures was arrived at more readily and with less anxiety than early in the war.
Posterior dislocation with fracture of posterior rim of acetabulum-In all posterior dislocations associated with fractures of the rim of the acetabulum, the dislocation was reduced by closed manipulation, just as the 15 simple dislocations in this series had been reduced, and with as little difficulty.
The management of the fracture was another matter. In 4 instances, its existence does not seem to have been recognized when the reduction was done. In several other instances, its extent was not appreciated by the surgeons who administered emergency treatment and manipulated the dislocation. This is easy to understand. Unless crepitation was present or the rim fragment was very large and consisted of the entire posterior aspect of the joint, as it occasionally did, physical examination did not reveal the presence of the fracture. Furthermore, roentgenologic interpretation was often difficult because no one had had experience with this type of injury. Unless one knew the roentgenologic anatomy of the hip joint intimately and in detail, the fracture was likely to be obscured by the head of the femur in the roentgenograms taken before reduction. Even when it was evident in postreduction films, a false idea of the apposition of the fracture lines was sometimes conveyed, if only the anteroposterior roentgenograms were examined. A true picture of the displacement of the rim fragments was best obtained in a postero-oblique view, with the patient lying supine on the cassette and the injured side tilted 60 degrees. In this position, the posterior portion of the acetabulum could be visualized in profile.
In 5 of the 15 posterior fracture-dislocations, the hip was suspended in traction for 8 weeks, in a Thomas splint with Pierson attachment. In 3 of these cases, the method of treatment was dictated by circumstances. The patient had head or abdominal injuries which took precedence over the hip injuries. In the 2 remaining cases, the fractures of the rim, although they were larger than chip fractures or avulsion fractures, were still not extensive enough to affect the stability or the range of motion of the joint.
Quadriceps, hip flexion, and abduction exercises were begun 2 weeks after injury. Weight bearing was not attempted during the approximately 6 weeks
these patieints were under observation overseas. At the end of this period, when they were evacuated to the Zone of Interior as litter patients, they had no special complaints to indicate joint irritation, but there was, in each case, approximately 25-percent limitation of motion, and internal rotation of the hip was associated with some muscle spasm.
Ten other patients with dislocations associated with fractures of the rim of the acetabulum were treated by arthrotomy, after the dislocation had been reduced by closed methods. In seven cases in which the fragment of the rim was large, viable, and attached to the parietal portion of the capsule, the fracture was reduced and immobilized by internal fixation with a screw (figs. 77, case 34, and 78, case 37). In the three remaining cases, (cases 44, 45, and 46) the injury was irreparable; the main fragments were detached from their blood supply or the articular cartilage was so severely damaged that removal of the posterior rim was thought advisable (fig. 79).
FIGURE 77 (case 34).-A. Postero-oblique roentgenogram showing posterior dislocation of hip with fracture of rim of acetabulum and displacement of fragments. A large fragment from the rim, with its capsular attachment, lies above the head of the femur. B. Anteroposterior roentgenogram of fracture and dislocation shown in figure 77A, 2 years after open reduction and internal fixation. Arrow indicates ossification of attachments of joint capsule and small muscles.
The modern literature, including modern textbooks, was not in agreement at the outbreak of World War II about the best approach for operations on old fracture-dislocations of the hip. This sometimes confused the clearly defined problem of open reduction or the necessity for removal of extra-articular borne fragments following fresh fractures of the posterior rim of the acetabulum. In all of these cases in this series, the indications were clear cut. When a large single fragment of the posterior rim was attached to the posterior intertrochanteric region by a large flap of parietal capsule, the flap served as a hinge, and the
FIGURE 78 (case 37).-Postero-oblique roentgenogram of fracture-dislocation 2 weeks after closed reduction of dislocation and 3 days after open reduction of fracture of rim of acetabulum. Note penetration of screw through both cortices and accuracy of repair. Both of these criteria must be met for sound fixation and a good end result.
bone could readily be swung back and dovetailed into place. In 1 such case, although the fragment was large, the articular cartilage and cotyloid ligament had been ripped apart and were so roughened that replacement of the fragment would obviously have resulted only in an incongruent and perhaps painful joint, and it was therefore removed. When multiple small fragments were present, tlney were practically always lying loose or very loosely attached, and for the most part they had no blood supply. They therefore had to be removed to prevent ossification of the periarticular structures, migration of loose bone, and probably, in later years, a painful joint. In short, the decision as to replacement and fixation of the fragments or their excision had to be made at operation and was made chiefly upon the presumed vitality of the bone fragments.
The operation was performed with the patient prone, with small pillows placed along the anterolateral aspect of the chest. The operating table was equipped with a dropleaf under the hip, so that the surgeon could use flexion in his manipulations. The posterior approach described by Kocher,7 the similar approach described earlier by Von Langenbeck,8 or Osborne's9 modification of these incisions was used in all cases (fig. 75). Although the number of reported surgical repairs of fractures of the acetabulum was still less than a hundred in 1944-45, several of the more recent case reports had indicated that this approach is particularly well suited to posterior dislocations associated with fractures of the rim of the acetabulum, and it proved entirely satisfactory in these cases.
The incision was begun along a line drawn between the posterosuperior spine of the ilium and the inferior portion of the greater trochanter. At the latter point, it was turned distally for 2 inches along the posterolateral aspect of the thigh. The gluteus maximus was split along its fibers for its whole length above the line of the neck of the femur. Through this opening, the deep fascia covering the short external rotator muscles of the hip was widely exposed with the aid of a Balfour self-retaining retractor. A window 2 inches square was thus provided in the gluteus maximus. If larger exposure was desired, a part of the insertion of the muscle was also divided.
At this stage of the incision, the traumatic lesion in the short external rotator muscles, the capsule, and the joint were usually visible. The joint, as a rule, was found filled with blood clots, the volume sometimes amounting to as much as 75 cc. Removal of the hematoma exposed the fracture line and the glistening white head of the femur.
Both sides of the fracture line were next fully exposed by section of the tendinous insertions of the piriformis, obturator internus, gemellus superior, obturator externus, and gemellus inferior. Stubs of all tendinous insertions were left in situ on the trochanter for later repair. Heavy black silk sutures were placed in each end of each transected tendon to facilitate retraction and reflection of the short external rotators. In each one of these cases, the sciatic nerve passed across the medial aspect of the operative field, in the plane between the piriformis and superior gemellus, about 4 cm. medial to the superior edge of the acetabulum and 1 cm. medial to its inferior rim. Reflection of the muscles medially (fig. 76C) afforded sufficient protective covering for it. Reported variations in the course of the sciatic nerve in this region were not observed in any of these cases.
Further dissection was limited to the region anterior or deep to the external rotator muscles and was carried out subperiosteally, on the pelvic side of the fracture line. In order to prevent injury to the blood supply, neither the periosteum nor the joint capsule attached to the rim fragments was lifted from them.
FIGURE 79 (case 46).-Posterior dislocation of hip joint with fracture of rim of acetabulum. A. Anteroposterior roentgenogram taken immediately after injury. B. Anteroposterior roentgenogram following manipulation and reduction of dislocation. Rim fragment remains displaced inferiorly. This position, rather than a position superior and lateral to the head, which is the usual location in displacements of the posterior rim, indicates that in this case the capsular attachments of the bony fragment have either been subjected to extensive injury or have been destroyed. C. Specimen of posterior rim of acetabulum removed at operation. The fragment, which was avascular and completely devoid of blood supply, was attached to a torn and twisted pedicle of joint capsule. In this case, the lost segment amounted to almost the complete width of the acetabular cartilage. A hemorrhagic lesion of the retinaculum was also observed at operation.
After loose chips of bone and flakes of articular cartilege had been removed, the main fragment could always be fitted accurately into place by apposing the dorsal edges as closely as possible. A position of adduction, 10 degrees of flexion, and slight external rotation provided maximum relaxation of the
FIGURE 79.-Continued. D. Anteroposterior roentgenogram 2 years after original injury, showing necrosis of superior segment of femoral head. E. Lateral roentgenogram showing avascular necrosis of head of femur. Note cleavage line between necrotic and viable portions of head. Dead bone is in process of separation and replacement by new bone. F. Anteroposterior roentgenogram 2½ years after injury, showing additional absorption, separation, and replacement of necrotic portion of femoral head. G. Lateral roentgenogram 3 years after injury, showing sheering of dead portion of head away from living bone, posteriorly, at line of absorption and regeneration. In retrospect, the significance of the hemorrhagic lesion of the retinaculum observed at operation became apparent. It had been ignored at the time but undoubtedly furnished the genesis of the avascular necrotic process which later occurred in the femoral head.
posterior capsule and eliminated pressure of the head of the femur on the posterior rim, so that the circumstances for maintaining the reduction while internal fixation was applied were optimum.
The location for the application of the screw was carefully calculated, so that it would not be placed too near the articular cartilage. It was inserted almost horizontally, at right angles to the plane of the fracture and was directed somewhat superiorly. A depth finder was used to make certain that it had penetrated the cortex of the inner table of the pelvis. Bicortical penetration (fig. 78) was necessary to secure fixation capable of withstanding the test of cautious rotation and flexion, which were carried out at the conclusion of the operation. Reduction of the fragments also closed the defect in the posterior capsule. The operation was concluded by the excision of small amounts of shredded muscle tissue, the repair of the tendons of the small deep muscles, and closure of the wound in layers with black silk sutures.
Immediate postoperative care was instituted with the patient prone and the hip flexed 20 degrees over a pillow. Gluteal setting-up exercises were begun as soon as the sutures were removed. At the end of 5 to 10 days, the patient was turned to the recumbent position, and the hip was suspended in 10 pounds' skin traction in a Thomas splint with Pierson attachment. Uncomplicated wound healing resulted in every instance in this series.
Irreducible fracture-dislocations-Before roentgenograms were available, the diagnosis of unreduced or irreducible dislocations of the hip was usually established by assumption, based on failure of multiple attempts at reduction. In most of the recorded cases, so few data are supplied that it is difficult to determine why conservative therapy failed, beyond the fact that the injuries are uncommon and treatment was carried out by physicians who had had little or no experience with them.
The eight patients in this series with irreducible fracture-dislocations (figs. 80, 81, and 82, cases 47 through 54, table 14), all arrived in general hospitals after at least two unsuccessful attempts at reduction had been made in field and evacuation hospitals. Improved roentgenograms, including stereoscopic films, at once revealed the difficulty: either the replacement of the femoral head was obstructed by bony fragments detached from the rim or the joint cavity had become so deformed and distorted by the injury that it could no longer receive the head of the femur. For anatomic reasons, it was therefore impossible to effect reduction by closed manipulation in any of these cases. Under the circumstances, conservative therapy in traction was of limited value, but it was employed consistently, partly for the patient's comfort and partly because it was necessary to retain anatomic approximation between the head of the femur and the fractured pelvis.
In three cases (cases 47, 48, and 49) bone fragments had lodged between the head of the femur and the joint cavity and provided an obstacle to reduction. In one of these cases (fig. 80, case 49) an attempt was made to disengage the fragments by pulling the head of the femur down over them and thrusting it inward with grinding and scooping movements. These maneuvers, in addition
to being futile, were obviously harmful to the joint surfaces. They were therefore not employed in any other similar case. Instead, immediate open reduction was resorted to.
The anterior iliofemoral approach was employed for arthrotomy in all 3 cases. In each instance, the anterior portion of the capsule was found intact, but the superior and posterior portions were stretched, frayed, and torn. Small fragments of bone were removed, but all fragments 1 sq. cm. or larger which were found attached to pedicles of torn capsule or strips of the short external rotator muscles were left in situ. In order to replace the head of the femur in the joint, it was necessary, in addition to employing traction, to pry loose interposed capsule, muscle, and torn tendon with the aid of a smooth, flat instrument. In 1 case (case 49) a large fragment of bone became lodged in the acetabular fossa during the manipulations, and, to remove it, it was necessary to make an incision in the intact anterior portion of the capsule and disarticulate the hip. In all of these cases, however, the obstruction to replacement of the head of the femur in the joint was not fragments of bone, but the interposition of the soft part attachments of these fragments.
In the five other cases of irreducible fracture-dislocation (cases 50 to 54) the problem was not only the dislocation of the hip but also the extensive fractures involving the entire acetabulum and the adjacent pelvis. In two cases (fig. 81, case 52, and case 53) manipulations were undertaken to restore the length of the extremity and place the head of the femur in a more suitable anatomic relationship to the pelvis. They were carried out with great caution because the risk of injury to the sciatic nerve was fully realized. The result did not justify the hazard.
In 4 of the 5 patients with disorganized hip joints, traction and suspension were employed for several weeks, until improvement in the general condition permitted either a major reconstructive procedure or transportation to a hospital for prolonged convalescent care. In the remaining case (fig. 82, case 54) peroneal palsy developed while the patient was still in the fixed hospital. Exploration was clearly indicated and was performed by the following technique:
The hip joint was exposed through the posterior approach already described. The sciatic nerve, which was intact, was discolored by ecchymosis for about an inch. It was stretched over a large fragment of bone which represented the entire posterior portion of the acetabulum, and it seemed reasonable to assume that the peroneal palsy was the result of compression of the nerve, because of traction on the femur in the soft parts between the head of the femur and the fragment of rim. The nerve was retracted medially in order to expose the joint cavity. The head of the femur was adducted and internally rotated beyond the normal range, and the rim fragment was retracted laterally with a bone hook. The acetabulum was obliterated by a displaced, impacted fragment of the body of the ischium. A second fracture of the superior ramus of the ischium was then found. In order to disimpact and reduce these fragments, it was necessary to mobilize and reflect the origin of the quadratus femoris. The body of the ischium could then be pulled down, and this maneu-
ver immediately reestablished the joint cavity and permitted reposition of the femoral head in the joint. The rim fragment was fixed in position with a single screw. Function of the hip was regained in this case, but in many similar injuries, although the immediate surgical result may be good, fusion of the joint or arthroplasty is necessary within a few years of operation.
Posterior dislocation with fracture of the head of the femur-Gross fractures of the head of the femur, in contrast to minor bruises, fissures, and subchondral fractures, are uncommon in dislocations of the hip. There were four in this series (table 15). In two impacted fractures (cases 56 and 58)
FIGURE 80 (case 49).-Posterior dislocation of hip joint with comminuted fracture of posterior rim of acetabulum. A. Postero-oblique roentgenogram shortly after injury. B. Antero-oblique roentgenogram after closed reduction of dislocation, which was accomplished immediately after injury. Note fragment of bone carried into acetabular fossa during manipulations. This view of the joint provided the best demonstration of the fragment and of its relation to the head of the femur. C. Lateral roentgenogram after removal of bone fragment at open reduction, 3 weeks after injury.
FIGURE 80.-Continued. D. Bone fragments removed from joint by anterior iliofemoral arthrotomy. The large fragment on the right was removed from the acetabular fossa. The three smaller fragments on the left were embedded in the posterior soft parts and were held in situ only by shreds of capsule or cotyloid ligament. The size of the fragments removed is in striking contrast to the apparently slight loss of articular surface demonstrated in figure 80C. E. Anteroposterior roentgenogram 20 months after arthrotomy, showing avascular necrosis of head of femur. The patient had used crutches ever since he became ambulatory, but this precaution did not prevent collapse of the bone structure.
there was little displacement. In the third case (fig. 83, case 55) the anteroinferior portion of the head, including the fovea capitis femoris and its attachments to the ligamentum teres, remained in the acetabulum. In the fourth case (fig. 84, case 57) there was appreciable limitation of motion of the joint under anesthesia, an observation not made in any of the other cases in this group. Open reduction was therefore performed at once to reduce the size and diameter of the deformed head and establish free motion as soon as possible. The following technique was employed:
The entire joint was exposed by the anterior iliofemoral approach and disarticulation was carried out as in a Smith-Petersen arthroplasty. The head
FIGURE 81 (case 52).-Extensive fracture of acetabulum with dislocation of hip joint. A. Anteroposterior roentgenogram [retouched] showing fracture and dislocation, with disorganization of hip joint and separation of symphysis pubis. B. Anteroposterior roentgenogram showing patient in vertical skeletal traction after reduction of dislocation by closed manipulation.
FIGURE 81.-Continued. C. Patient in vertical skeletal traction. This was the only position which would prevent redislocation of the head, with pressure on the sciatic nerve and severe pain radiating down the leg. The virtue of this method of suspension was the patient's own comfort and the simplification of nursing care. He could actively lift himself off the bed. Note the immobilization of the right arm and forearm in skin traction, for the management of other fractures. The patient also had an injury to the posterior portion of the urethra, for which urologic surgery and treatment were necessary.
of the femur was the site of an impacted stellate fracture, and the articular cartilage was split into six segments which were separated by intermediate areas of raw cancellous bone. The anteroinferior portion of the head protruded from the acetabulum and was crushed against the rim. The ligamentum teres was torn apart and was pressed into the substance of the crushed bone in the fovea. The fractured, impacted bone which protruded from the head of the femur was removed with a sharp osteotome, and the head was thus reduced, as nearly as possible, to its normal size and shape. The profuse bleeding which occurred during this step of the procedure suggested that the greater part of the circulation, or at least its capsular source, was still intact. After hemostasis had been obtained by compresses soaked in warm physiologic salt solution, the head was replaced and was now found to revolve freely in the acetabulum.
The management of these 4 fractures of the head of the femur with posterior dislocation was in accord with current trends. Such fractures must, properly, all be regarded as serious, but the treatment instituted and the results obtained necessarily vary with the exact type of lesion. Conservative management is
obviously preferable whenever it is possible, as it was in 3 of these 4 cases. Arthrotomy with removal of displaced bone fragments is likely to be successful only when the fracture is limited to the anteroinferior portion of the head, as it was in the fourth case. Had the weight-bearing surface of the head been involved, failure would have been inevitable.
FIGURE 82 (case 54).-Irreducible fracture-dislocation of hip joint. A. Anteroposterior roentgenogram showing fracture-dislocation. The body of the ischium is impacted into the body of the ilium, and there is an associated fracture of the superior ramus of the ischium. The joint cavity is totally obliterated. B. Anteroposterior roentgenogram 3 weeks after open reduction of dislocation and internal fixation of fracture of acetabulum. Note that a single screw holds the posterior portion of the acetabulum in accurate position. C. Lateral roentgenogram showing repair.
FIGURE 83 (case 55).-Fracture-dislocation of hip. A. Anteroposterior roentgenogram showing posterior dislocation of hip and fracture of head of femur. The separated segment of the inferior portion of the head includes the fovea. It is attached to the ligamentum teres, in the acetabulum. B. Anteroposterior roentgenogram 2½ years after original injury. The proximal fragment, which is demonstrable only with the femoral head in extreme internal rotation, has become reattached to the femoral head and does not interfere with the normal function of the joint.
Complications of fracture-dislocations of the hip may be either early or late. In these 27 cases, early complications, some of which were actually part of the original injury, included urinary-tract injuries, retroperitoneal hemorrhage, thrombosis of the hemorrhoidal veins, thrombophlebitis, sciatic neuritis, and sciatic-nerve palsy. Late complications encountered within a period of 6 weeks to 2 years after injury included ossification of the joint capsule, synovitis, traumatic arthritis, and avascular necrosis of the head of the femur. From the standpoint of possible late complications, injury to the soft parts proved fully as important as injury to the articular cartilage.
Urinary-tract injuries-Nine of the twenty-seven fracture-dislocations of the hip joint were associated with multiple fractures and other injuries of the pelvis, including displaced fractures of the superior and inferior pubic rami, separation of the symphysis pubis, and dislocations of the sacroiliac joint. In several of these cases, the urethra was obstructed or lacerated. A number of patients had symptoms evidently caused by irritation of the bladder, and, in 1 instance, gross bladder injury required suprapubic cystostomy. In all these cases, treatment of the urinary-tract injury took precedence over treatment of the skeletal injury.
Retroperitoneal hemorrhage-Blood is probably extravasated into the retroperitoneal space in all comminuted fractures of the acetabulum and may diffuse upward to the under surface of the diaphragm. This observation was
FIGURE 84 (case 57).-Fracture of head of femur with posterior dislocation of hip joint. A. Anteroposterior roentgenogram showing fracture-dislocation 2 weeks after closed manipulation. Rotation of the hip was painful and the total range of motion was less than 75 percent of normal. B. Hip joint after exposure and disarticulation through anterior iliofemoral approach. Note the impacted stellate fracture of the head of the femur, with segmentation of the articular cartilage. The photograph has been retouched for clarity. C. Anteroposterior roentgenogram of hip joint after anteroinferior portion of mushroom-shaped femoral head has been sculptured down to normal size at open operation, 2 weeks after original injury. D. Anteroposterior roentgenogram of both hips, 2 years and 8 months after open operation. The head of the right femur shows normal density, but the articular cortex is slightly irregular, and the old epiphyseal line is partially obliterated. Avascular necrosis of the entire head of the femur probably has not occurred in this case.
made at autopsy in 1 case in this series (case 51), in which death occurred several days after wounding, from a concurrent head injury. The possibility of retroperitoneal hemorrhage is one of the reasons why a period in traction for 5 to 10 days is always advisable before surgery on the hip joint is undertaken. If distention, vomiting, and fever are present in the first few days after extensive fractures of the acetabulum have occurred, the possibility of retroperitoneal hemorrhage should be considered in the management of the case.
Thrombosis of the hemorrhoidal veins.-This complication, which does not seem to have been reported previously in association with fracture-dislocations of the hip, occurred in 3 cases in this series. In 2 cases, including the fatal case just mentioned, it was transient and minor. In the third case (case 52), massive edema of the rectum and anal orifice developed a few days after the patient had sustained an extensive fracture of the left acetabulum, complicated by dislocation of the sacroiliac joint on the same side. There was no previous history of hemorrhoids, and the manifestations were far more severe than would ordinarily be observed in that condition. The edema diminished after treatment with hot compresses and disappeared entirely within 3 weeks. There was no recurrence within a 2-year period of observation.
Thrombophlebitis-There was only 1 instance (case 54) of thrombophlebitis in the series. It was severe but was treated successfully with Dicumarol (bishydroxycoumarin). This is a complication which is a possibility in all injuries of the lower extremity but is probably no more likely to occur in injuries of the hip than in any other injuries.
Sciatic neuritis-Four patients complained of severe pain over the lateral aspect of the lower extremity. In each case, it began shortly after wounding and persisted for several months. There was no motor weakness, and the changes of sensation which were observed were inconsistent and followed no pattern. Repeated neurologic examinations furnished no indication for exploration, and it was assumed that the sciatic nerve had suffered minor damage from the dislocated femoral head or from displaced fragments of the rim of the acetabulum.
Sciatic-nerve palsy-Attention has already been called (p. 283) to the single case (case 54) in which the sciatic nerve was explored during open operation for an irreducible fracture-dislocation, and to the remarkable fact, in view of the close relationship between the nerve and the displaced fragments of the rim, that sciatic-nerve palsy did not occur almost uniformly in comminuted fractures of the posterior portion of the acetabulum. In all such cases, the shelf of bone which protects the nerve from the trauma of motion of the head of the femur is missing. In this series there were no instances of permanent damage to the sciatic nerve in the 15 simple dislocations and only 4 instances in the 27 fracture-dislocations. In all 4 cases, the dislocations were irreducible by closed manipulations (table 14).
In all 4 cases in which it was present, the palsy was a delayed phenomenon. One patient, who had had good motor function and sensation in the foot immediately after injury, began to complain of pain and foot drop after he had been en route for several hours from a forward hospital to a fixed installation.
In another instance, the foot drop became apparent after several hours of pain which radiated down the extremity following manipulation of the dislocated hip. In the 2 remaining cases, the palsy appeared only after a second manipulation of the dislocated hip. All 4 cases illustrate the well-known, but as yet inexplicable, fact that in all forms of trauma to the sciatic nerve at the level of the hip joint, the peroneal portion is apparently more vulnerable than the remainder of the nerve. This portion is more commonly affected than the posterior tibial portion, and paralysis becomes evident sooner after injury.
Only 1 of the 3 surviving patients had any return of function of the peroneal nerve. The other 2, at the end of 2 years, presented extreme atrophy of the affected extremity and still had complete paralysis of the extensors and evertors of the foot. The discouraging functional results are in correspondence with those recorded in the literature, which show that return of function is always slow and is usually incomplete if it occurs at all.
The significant feature of these 4 instances of peroneal palsy is that in every case it had been preceded by one or more attempts at manipulative reduction of the dislocation. Obviously, the risk of nerve injury is so great in these circumstances that manipulations should be avoided and dislocations reduced under full vision at primary open operation, whenever the acetabulum is too distorted by the injury to receive the head of the femur or there is other evidence that manipulation will be difficult or unsatisfactory. Careful examination of even emergency roentgenograms will usually provide enough information to determine these points.
Calcification of the joint capsule-Ossification of the joint capsule, which occurred in only 2 simple dislocations without fracture, occurred in 14 of the 27 fracture-dislocations. In 12 of the 14 cases, the fracture was in the rim of the acetabulum, and in 2 instances it involved the femoral head. Roentgenograms revealed new bone formation in the posterior and superior portions of the capsule, but in every instance the clinical limitation of motion of the joint was slight. Almost all of the other patients with fracture-dislocations also showed deposits of new bone in either the femoral or the acetabular attachment of the capsule, but they were small and had no detectable effect on the range of motion of the joint.
Synovitis.-A number of patients with simple dislocations of the hip joint complained of minor aches and pains, discomfort in inclement weather, clicking or snapping sensations in the joint, or severe pain when heavy objects were lifted (p. 297). Similar complaints were made by a number of the patients with displaced fractures of the acetabulum (p. 302). These complaints were almost universal in the 27 patients with fracture-dislocations, even in those who had a normal range of motion. Slight muscle spasm was noted in some cases, even when the roentgenograms showed flawless joint surfaces and no bone atrophy. These symptoms and signs were regarded as indicative of traumatic synovitis, and the tentative diagnosis was substantiated by the fact that in a number of cases they disappeared within a year or two after injury. Whether the other patients who still complained of them at the 2-year followup will
eventually be rid of them, or will present traumatic arthritis 5 or 10 years after injury, is not possible to say. The literature is not helpful, as a search of it revealed no cases followed for these periods of time.
Traumatic arthritis.-Degenerative joint changes secondary to trauma were observed in 12 of the 27 fracture-dislocations of the hip, sometimes within 6 weeks of injury and always within the 2-year followup period. Clinically, it was not possible to distinguish this complication from traumatic synovitis or from the early stages of avascular necrosis of the femoral head. Roentgenograms clarified the diagnosis by revealing evidences of abnormal wear and tear on the defective joint surfaces, in the form of irregularities on the articular surface opposite the defect on the injured side.
In 6 of the 12 patients with traumatic arthritis, the defects revealed roentgenologically in the posterior rim of the acetabulum could be explained by unreduced fractures (cases 39 and 40), operative removal of bone fragments (fig. 85, case 44), and healed fractures of the head of the femur (cases 56, 57, and 58). All of these patients complained of painful crepitations and clicking sensations and had special discomfort when they climbed stairs or lifted heavy objects. Some also complained of morning stiffness, pain during inclement weather, and inability to stand for long periods without discomfort. These complaints were not easy to evaluate and in some instances seemed more closely related to the compensation aspects of the injury than to the pathologic process.
The other 6 patients with traumatic arthritis began to complain of severe pain in the joint, with increasing limitation of motion, within a year after injury. All had extensive fractures of the acetabulum (table 14). Roentgenograms in every instance demonstrated irregularities and subsequent loss of the entire joint space. All of these patients were regarded as immediate candidates for arthrodesis or mold arthroplasty, depending upon their individual requirements. In 4 instances (cases 49, 50, 52, and 53) it is known that these operations were done at Army or Veterans' Administration hospitals.
Avascular necrosis of the head of the femur-There were no instances of avascular necrosis of the femoral head in the 15 simple dislocations in this series (p. 261). This complication appeared, however, in 2 patients with fracture-dislocations. In one instance (case 46, fig. 79) a line of demarcation separated living bone tissue of the femoral head and neck from an area of increased density and necrosis in the superior portion. In this case, as already noted (p. 275), the significance of the large blood clot present in the retinaculum at the margin of the articular cartilage at the head of the femur was not appreciated when it was observed shortly after injury. In the other (case 49, fig. 80), ecchymosis, new granulation tissue, and thickening of the entire posterior portion of the reflected capsule had been observed at operation, as in the case just described, again without appreciation of the importance of the findings. In this case, there was nearly complete disintegration of the bony structure of the femoral head. The observations made at operation seem to supply sufficient evidence that avascular necrosis is associated with thrombosis of either the retinaculum or the intertrochanteric branches of the medial
FIGURE 85 (case 44).-Dislocation of head of femur with fracture of posterior rim of acetabulum. A. Postero-oblique roentgenogram showing fracture of rim of acetabulum after reduction of dislocation by manipulation. The posterior rim, as usual, has not been influenced by conservative treatment in traction and remains undisplaced. B. Anteroposterior roentgenogram of both hips showing loss of considerable amount of posterior rim of acetabulum on affected side, as result of irreparable damage and surgical excision. C. Fragment removed at open operation. The extensive damage to the acetabular cartilage and the impacted fracture contraindicated its replacement.
circumflex vessels. This proof is in line with the damage which can be done experimentally (p. 255) by traction and tearing of the capsule in the posterior intertrochanteric region.
Up to 1947, approximately 50 well-documented cases of avascular necrosis had been encountered in a total of 270 recorded posterior dislocations of the hip joint. The lesion was usually observed within 6 to 18 months after injury. In the 2 cases in this series it occurred within 6 to 12 months. This time span does not bear out the theories of some observers that avascular necrosis may be a late complication, occurring 5 years or more after injury, though serial
FIGURE 85.-Continued. D. Anteroposterior roentgenogram 2 years and 10 months after excision of posterior rim. Note irregularities and slight necrosis of surface of femoral head, presumably due to irregular wear and tear on the articular surfaces. E. Lateral roentgenogram showing area of degenerative changes in head of femur, corresponding to area of fracture defect of articular surface of acetabulum. As yet, there is little if any narrowing of the joint space. There is no evidence of avascular necrosis, and it is significant that retinacular hemorrhages were not observed at operation. These roentgenograms should be compared with those shown in figure 79, case 46, in which retinacular hemorrhages were observed at operation, and avascular necrosis followed.
roentgenograms have not been studied in a large enough group of cases to determine the true incidence of the complication in adults. It is known that it is relatively commoner in childhood and adolescence, probably because of the larger and more vulnerable vascular network in the reflected capsule at these periods of life.
Since avascular necrosis occurs in a variety of injuries of the hip joint, the microscopic changes associated with it have been described in detail. The literature concerning its pathogenesis is vast and highly controversial. A survey of 135 contributions published up to 1945 may be summarized about as follows:
1. The blood supply to the superior portion of the head of the femur comes a long distance, through the reflected capsule, which receives it from numerous branches of the medial femoral circumflex vessels in the posterior and posterosuperior aspects of the intertrochanteric region.
2 The pathologic and roentgenologic appearance of avascular necrosis in man differs from the process produced experimentally in rabbits, dogs, and goats by surgical severance of all the arteries and veins which enter the hip joint. In man, collapse of bone structure occurs relatively early, but revascularization of the area and replacement of dead bone begins relatively soon, so that there is only temporary interruption of the arterial circulation or venous return. Repair is, however, imperfect, and the joint is permanently defective.
3. The possibility that contusions and subchondral compression fractures may be etiologic factors has been entertained, but the evidence is circumstantial and inadequate.
4. The condition has been reported not only after posterior dislocations, after which it occurred in the 2 cases in this series, but also after anterior dislocation and central fracture of the acetabulum with prolonged intrapelvic protrusion.
5. Abstinence from weight bearing has been suggested as a method of prevention, but there is considerable doubt that it would be effective. One of the patients in this series began weight bearing within 3 months of injury, but it was not permitted in the other case because roentgenograms showed some loss of substance and irregularity of contour of the head of the femur. One is forced to assume, therefore, that a certain amount of necrotic bone is crushed, independently of weight bearing, by the action of the pelvifemoral muscles upon the head of the femur, just as happens in other ischemic and destructive lesions of the hip.
6. There is doubt that so-called creeping bone replacement can restore the joint structure in adults with avascular necrosis as effectively as it does in children with Legg-Calvé-Perthes disease, or in the occasional adult with an impacted fracture of the neck of the femur.
The only treatment now available for avascular necrosis of the head of the femur is restriction of full weight bearing. This is not a practical method for, as in these cases, there is a lack of premonitory symptoms, and the prohibition usually begins only when roentgenograms show the process in the articular surface of the femoral inead. Then the damage has already been done. Nonetheless, weight bearing should be prohibited, even after the process is established, to prevent torsion of the necrotic femoral head during the period of new bone replacement.
The value of the methods of treatment used in these 58 jeep injuries of the hip joint can be determined only by the end results. Even with all the resources available for tracing former military personnel, it was possible to secure results in only 42 cases (chart 1; tables 11 to 15, inclusive). Furthermore, the followup covers only a 2-year period, and there is much to be said for the position of the observers who maintain that the results of such injuries cannot be determined conclusively until 5 to 10 years have elapsed. On the other hand, although very few cases in the literature have been followed for longer than 2 years, as a practical consideration, only traumatic arthritis is likely to become evident after a longer period.
In order to compare the results obtained in these three groups of military injuries with those obtained in civilian practice, a comparison was made with patients treated on the fracture service of the Massachusetts General Hospital who had sustained similar injuries in civilian life and who had been followed up for 2 years or more (tables 16, 17 and 18). To secure the civilian cases for
comparison, it was necessary to go back for a period of 12 to 14 years. The chief difference in therapy in the two series is that patients in civilian life were immobilized for briefer periods than Army patients and were permitted earlier weight bearing.
The 9 (of 15) patients in the military series with simple dislocations of the hip who could be followed up were compared with 7 patients with similar injuries treated at the Massachusetts General Hospital. The results (tables 11 and 16) were essentially similar. No immediate benefits were derived from the extended period in traction and the extended restriction of weight bearing practiced in the military group. Five of the nine patients in this group and 2 of the 7 in the civilian group complained of slight aches and pains in the joints
TABLE 16.-Selected traumatic dislocations of hip joint, Massachusetts General Hospital, 1931-47
TABLE 17.-Selected fractures of acetabulum, Massachusetts General Hospital, 1932-46
TABLE 18.-Selected fracture-dislocations of hip joint, Massachusetts General Hospital, 1933-45
at irregular intervals. In both groups, patients who did hard labor or who worked on farms were uncomfortable when they had to lift heavy objects. During the first few months of weight bearing, 4 in each group complained of pains in the joint when the weather was rainy, when they were engaged in sports, or when they climbed stairs, but at the end of 2 years these symptoms had disappeared. One patient in the military group complained of night pain, for which roentgenograms furnished no explanation. The only patient with avascular necrosis in either series was a 14-year-old boy in the Massachusetts General Hospital group. This complication, as already noted, is more common in adolescence than in adult life. There was no evidence of degenerative arthritis in either group.
Fractures Without Dislocation
The 14 (of 16) patients with fractures of the acetabulum without dislocation who could be followed up in the military series were compared with 7 patients with similar injuries treated at the Massachusetts General Hospital (tables 12 and 17). When like fractures were compared, the results in both series were much the same.
At the end of 2 years, 11 of the 14 military patients had no serious disability, though many complained of mild pain, snapping or clicking sensations in the joint, stiffness early in the morning, and discomfort when lifting heavy objects. Those with displaced fractures of the superoposterior rim of the acetabulum appeared to have more symptoms and more disability than those with anterior rim or central fractures. One patient (case 19) with an unreduced fracture of the posterior rim had severe pain and some limitation of motion at the extremes of normal range in all directions. Another (case 18), whose joint was accurately repaired at operation, had a normal hip. The 2 cases were similar in that in both the displaced rim fragment was relatively large. The assumption is that when the fragments were thin and small and carried less articular cartilage with them as disruption occurred, the important function of the joint was not affected, regardless of whether the fragments were or were not replaced. The 3 patients with extensive fractures of the acetabulum and destruction of the joint cavity (cases 29, 30, and 31) had ankylosis and complete disability 2 years after injury and were thus candidates for arthrodesis or arthroplasty. Neither avascular necrosis nor degenerative arthritis was present in any instance of fracture without dislocation at the end of 2 years.
It is unfortunate that the single patient in the military series with an intrapelvic protrusion at the head of the femur could not be followed up. There were 2 such injuries in the comparative series from Massachusetts General Hospital, both of which were treated by the plan employed in the single case in the military series. Neither of the civilian patients had any disability at the end of 1 year and 3 years, respectively. It is difficult to understand why, in this type of central fracture, with intrapelvic protrusion of the head, healing could occur without functional disability, regardless of the displacement of the inner table of the pelvis. The explanation may be that in
these successful cases the fracture affects the anterior rim of the acetabulum, which is a nonessential part of the joint, while the posterior and superior portions of the horseshoe-shaped articular cartilage, which are vital to function, are spared. On the other hand, when intrapelvic protrusion of the femoral head is associated with extensive fractures of the acetabulum and bursting of the triradiate seam of ilium, ischium, and pubis, or with distortion of the rim of the acetabulum, the functional results, as might be expected, are poor. Classification of these injuries into these two categories would do away with the difficulty in explaining the discrepancies in results reported in the literature.
Fractures With Dislocation
The 19 (of 27) patients with fracture-dislocations of the hip who could be followed up for 2 years were compared with 4 patients with similar injuries treated by similar methods at the Massachusetts General Hospital (tables 13 to 15 and 18). It is interesting that, over a 12-year period at the civilian hospital, it was possible to locate only 4 patients with this sort of injury, while several times as many had been observed over an 18-month period in a single military-hospital center.
The 4 civilian patients showed, variously, excellent results, fair results, and, in 2 cases, poor results, with hip fusion necessary in one and mold arthroplasty in the other. In the military series, the results were also varied. Dislocations without displacement of fragments or associated with only minor fractures of the rim of the acetabulum showed, as would be expected, results which did not differ from the best and worst results observed in uncomplicated dislocations. Most of the patients, presumably because of scarification in the synovial membranes and the joint capsule, complained of aching pains, morning stiffness, and snapping of the hip, just as the patients with simple dislocations. In both groups of cases, the integrity of the lunate acetabular cartilage and of the rim of the acetabulum was presumably the critical factor in the end results. In matched cases of fractures treated conservatively and treated by open operation, good function and little or no disability were accomplished when the joint surfaces were restored as precisely as possible, but one could be certain that this had been achieved only when open operation had been done.
Dislocations which were reduced, associated with significant fractures of the rim of the acetabulum which were not reduced, gave, for the most part, fair to poor results after surgical removal of the entire posterior rim. Late degenerative changes in the joint were the rule. When, however, a single large fragment of the posterior or superior portion of the acetabulum had been detached and could be replaced precisely, with the capsular attachment intact, the results were excellent and were fully as satisfactory as the results of uncomplicated dislocations without fractures (fig. 86, case 32). Comminuted fractures which disrupted the acetabulum produced serious disability, and arthrodesis or arthroplasty was required in almost every case within 2 years.
In all 4 cases in which fracture-dislocations were associated with distortion of the femoral head, slowly progressive degenerative changes occurred over the
2-year followup period (table 15). One patient, whose injury was limited to the anteroinferior portion of the head, had practically normal function and in the 3 other cases, in which the entire head was involved, the range of motion was still useful.
Traumatic arthritis occurred in 12 patients during the period of observation (p. 293) and avascular necrosis in 2 others (p. 293).
FIGURE 86 (case 32).-A. Anteroposterior roentgenogram showing central fracture of posterior rim of acetabulum and inner table, with posterior dislocation of hip joint. B. Anteroposterior roentgenogram of fracture and dislocation shown in figure 86A, after manipulation and reduction of dislocation on day of injury. Note that posterior rim fragment remains widely displaced and unaltered by treatment of dislocation. C. Anteroposterior roentgenogram of fracture and dislocation shown in figure 86A and B, 2 weeks after injury, after open reduction of fracture and internal fixation with a single screw.
At the end of 2 years of observation, the results in the 42 Army patients with dislocations, fractures, and fracture-dislocations of the acetabulum who could be traced corresponded closely with the results in the 18 similar patients studied comparatively from the fracture service of the Massachusetts General Hospital. The conclusion seems warranted that except in fractures of the posterior rim treated promptly by primary open reduction and internal fixation,
the nature and magnitude of the traumatic lesion, and not the technical procedure, determines the final outcome.
The experience gained in the management of these cases supports the impression derived from a survey of the literature that the seriousness of a fracture of the acetabulum depends upon whether the superior or the posterior portion of the lunate articulate cartilage has been injured. In the past, the tendency was to accept the imperfect results of conservative therapy because of the formidable nature of surgical management and also because there was no understanding of late degenerative changes in the joints. Now that the Army experience in World War II has shown the possibility of these changes, as well as the better results possible with surgical management in properly selected cases, there should be more willingness to adopt the principle that reposition of the bony fragments is as necessary in injuries of the hip joint as it is in injuries of other weight-bearing joints.