Acetabular dysplasia that is associated with static encephalopathy in an older child is a common problem, but currently there is no well defined solution. Most authors agree that a combination of soft-tissue procedures and bracing is the most appropriate treatment for children who are less than five years old8,12,17. Severe acetabular dysplasia with established subluxation or dislocation of the hip poses a more difficult problem. Soft-tissue procedures alone are not adequate for an older child, and the restoration of the stability of the hip necessitates either femoral or pelvic procedures, or both, as well as appropriate soft-tissue and capsular procedures1,6,14-16. Recently, a number of studies have shown that a global acetabular deficiency and, in particular, a posterolateral defect of the acetabulum is present in patients who have perinatal static encephalopathy2,4,7,18. For this reason, the efficacy of the traditional pelvic osteotomies has been called into question because they do not improve posterolateral coverage of the femoral head, although some authors have proposed using the Dega osteotomy because it improves posterolateral coverage9,11.
The Pemberton (pericapsular) acetabuloplasty differs from both the common redirectional (Salter, Sutherland, and triple innominate) osteotomies and the common capsular augmentation (Chiari and shelf) procedures in two principal respects13. First, because the triradiate cartilage is used as the point of rotation, the Pemberton osteotomy substantially improves anterior and lateral coverage while little or no posterior coverage is sacrificed. The second unique feature, as Pemberton pointed out in his description of the technique, is that the direction of the improved coverage can be altered from almost purely anterior to almost purely lateral by adjustment of the level of the cuts on the inner and outer tables of the pelvis13.
Because of these unique properties, since 1984 we have used a single-stage combined procedure for the treatment of severe dysplasia of the hip secondary to static encephalopathy in older children. The combined procedure includes appropriate soft-tissue releases, an open reduction of the hip with femoral shortening when necessary, a varus rotational osteotomy of the proximal part of the femur to correct femoral valgus angulation and anteversion, and a Pemberton periacetabular osteotomy to correct acetabular dysplasia.
We performed the current review to determine the effectiveness of this single-stage combined procedure involving the Pemberton osteotomy for the correction of acetabular dysplasia in patients who have perinatal static encephalopathy. We also sought to determine whether this procedure provides clinically important posterior stability of the hip and if its benefits persist with time.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
†Shriners Hospital for Crippled Children, St. Louis Unit, 2001 South Lindbergh Boulevard, St. Louis, Missouri 63131. Please address requests for reprints to Dr. Gordon.
We retrospectively reviewed the records of all patients who had a Pemberton osteotomy performed between January 1, 1984, and December 31, 1992. Forty-five patients (fifty-three hips) who had static encephalopathy were identified. One patient was lost to follow-up less than two years postoperatively, leaving forty-four patients (fifty-two hips) available for study (Table I). The indications for the operation included a subluxated or dislocated hip associated with pain or limited motion leading to problems with perineal care, walking, transfers, or sitting. Severe deformity of the femoral head after long-term dislocation was a relative contraindication for this procedure.
Thirty-three (75 per cent) of the patients had severe quadriplegia and were not able to walk. The remaining eleven patients (25 per cent) had diplegia and were able to walk. Eight patients had pain in either one hip or both hips (Fig. 1-A). The ages of the patients at the time of the operation ranged from four years and five months to sixteen years and five months (mean, eight years and eight months). Thirty-eight patients (forty-five hips) had perinatal static encephalopathy, one (one hip) had trisomy-12, two (three hips) had leukodystrophy, one (one hip) had had perinatal meningitis, one (one hip) had had transverse myelitis, and one (one hip) had microcephaly as a result of prenatal infection with cytomegalovirus. Radiographs that were made before the combined procedure demonstrated subluxation of thirty-nine hips (77 per cent) and dislocation of thirteen (23 per cent) (Fig. 2-A). The preoperative center-edge angles ranged from 17 to -80 degrees (mean, -11 degrees). Although this angle may be difficult to measure in some patients, we believe that it provides the best prognostic information about the long-term stability of the hip. Mose circles were used to identify the center of the femoral head8. Some authors9 have used the center of the proximal femoral physis to measure the center-edge angle, but we believe that this falsely increases the size of the angle. Previous operative procedures had been performed in twenty-two of the involved hips. An adductor tenotomy alone had been done in ten hips; a proximal femoral varus rotational osteotomy, in five hips; an open reduction and varus rotational osteotomy, in one hip; a varus rotational osteotomy with a soft-tissue release, in four hips; a soft-tissue release only, in one hip; and an adductor tenotomy and a trochanteric transfer, in one hip.
Procedures that were done concomitantly with the Pemberton osteotomy included a varus rotational osteotomy in fifty hips (96 per cent), with shortening of the femur in fifteen hips (29 per cent), a soft-tissue release in thirty-seven hips (71 per cent), and an open reduction in thirteen hips (25 per cent). Both of the patients who did not have a varus rotational osteotomy concurrently with the Pemberton osteotomy had had this procedure previously.
Operative Procedure
The entire lower extremity is prepared and draped so as to allow free access to the groin, the lateral portion of the thigh, and the iliac crest area. Tenotomies of the adductor longus and gracilis muscles, without obturator neurectomy, are done with the goal of obtaining 45 degrees of abduction with minimum resistance. If sufficient abduction has not been obtained after these tenotomies, femoral shortening is indicated at the time of the varus rotational osteotomy. At this point, an anteroposterior radiograph with the hip in 25 degrees of flexion, 20 degrees of abduction, and 20 degrees of internal rotation is made with use of the image intensifier. If the hip does not reduce concentrically, an open reduction of the hip is performed through an anterior approach. An adequate medial release of the capsule and lengthening of the iliopsoas tendon are the keys to attaining a concentric reduction. Next, a varus rotational osteotomy is performed through a lateral approach, with an osteotomy of the femur performed between the middle portion and the inferior margin of the lesser trochanter. Care is taken not to perform the osteotomy proximal to the middle portion of the lesser trochanter because of the risk to the medial circumflex vessels if the osteotomy enters the basilar-neck area. The femur is shortened, as needed, by resection of an adequate amount (usually one to five centimeters) of the proximal end of the distal fragment to achieve the goal of 45 degrees of abduction. The site of the femoral osteotomy then is fixed with a Wagner plate (Aesculap, San Francisco, California), maintaining a neck-shaft angle of 115 to 125 degrees, which corrects the anteversion to 0 to 10 degrees. The stability of the hip should be assessed both manually and radiographically. When the hip is unstable or severe dysplasia is present and the triradiate cartilage is open, a Pemberton osteotomy is performed. Although the site of the Pemberton osteotomy is inherently stable, sometimes the bone graft cannot be lodged securely to it and Kirschner wires may be necessary to hold the graft in place.
If the triradiate cartilage is not open on the preoperative anteroposterior radiograph of the pelvis, the patient is not a candidate for the Pemberton osteotomy. We have observed closure of the triradiate cartilage in patients as young as seven years old and as old as twenty-two years.
Postoperatively, a hip-spica cast is worn for six to ten weeks, until the sites of the pelvic and femoral osteotomies have healed adequately. After the cast has been removed, three to four months of physical therapy are necessary to restore mobility of the hip.
Clinical
The duration of follow-up for the forty-four patients ranged from two years to eight years and eight months (mean, four years). At the time of the latest follow-up, no patient had worse function compared with the preoperative status. No patient who had quadriplegia gained the ability to walk postoperatively. The patients who had diplegia had subjective improvement, and none of them had a decrease in the ability to walk. All of the eight patients who had had pain in the hip preoperatively were pain-free at the time of the latest follow-up (Fig. 1-B and 1-C). However, one patient who had been pain-free preoperatively had pain in the hip postoperatively. Most parents were pleased with the result of the procedure, noting improved hygiene and decreased pain. Only the parents of the child who had pain at the most recent follow-up visit were displeased with the result.
Radiographic
Radiographically, the center-edge angle improved from a mean of -11 degrees preoperatively to a mean of 28 degrees immediately postoperatively. At the time of the latest follow-up, the center-edge angles ranged from 5 to 62 degrees (mean, 27 degrees) (Fig. 2-B and 2-C). The center-edge angle was less than 20 degrees in six hips; one of these six was subluxated and, in retrospect, had never been fully reduced, and five had had a substantially widened or v-shaped teardrop preoperatively (Fig. 3-A and 3-B). The center-edge angle did not deteriorate with time, but rather, it seemed to improve as the hip remodeled following the procedure. The age of the patient at the time of the operation did not have an appreciable effect on the radiographic result. The mean age was eight years and ten months. In the twenty-four patients (twenty-nine hips) who were younger than the mean age, the center-edge angle improved from a mean of -12 degrees preoperatively to a mean of 29 degrees postoperatively. In the twenty patients (twenty-three hips) who were older than the mean age, the center-edge angle improved from a mean of -10 degrees preoperatively to a mean of 27 degrees postoperatively.
There were no complications attributable to posterior uncovering of the hip. The patient (Case 29) in whom the hip was subluxated at the time of the latest follow-up had been managed with a varus rotational osteotomy and soft-tissue releases at the time of the Pemberton osteotomy. In retrospect, the hip was not concentrically reduced at the time of the operation and an open reduction should have been done. The patient remained pain-free at the time of the most recent follow-up, despite the continued subluxation. Avascular necrosis of the femoral head was not observed in any patient, and no hip was dislocated.
Subsequent Operations
Two patients had subsequent operative procedures on the hip. One patient (Case 11) had a trochanteric advancement because of a painless abductor lurch. After this operation, the patient remained pain-free and the gait was slightly improved. The second patient (Case 20), who had perinatal static encephalopathy with a major athetoid component, was five years and six months old at the time of the Pemberton osteotomy and the concomitant procedures. Three years and eight months later, the patient had an internal rotation and adduction contracture of the involved hip. The femoral head remained well covered, and there was no evidence of subluxation. Repeat adductor releases failed to adequately improve abduction intraoperatively, and a femoral shortening osteotomy and concomitant derotation of the femur were performed. Eighteen months after the subsequent operation, the patient had had no recurrence of the contractures.
Complications
No wound infections developed postoperatively. Although healing of the site of the femoral osteotomy was delayed (more than ten weeks) in two patients, all patients had final consolidation of the site by sixteen weeks. No patient needed a secondary procedure to attain union. Healing of the site of the pelvic osteotomy was not delayed in any patient. That site had been fixed with Kirschner wires in three patients because of concern during the operation about the stability of the bone graft. Pneumonia developed in one patient postoperatively. Skin ulcerations developed in two patients, both of whom had quadriplegia, while they were wearing the hip-spica cast.
Few reports have addressed the acetabular dysplasia that often develops in patients who have perinatal static encephalopathy1,6,11,14-16. Brunner and Baumann reported that pelvic procedures, including the Salter, Pemberton, Chiari, and shelf acetabuloplasties resulted in long-term stability of the hips of such patients. Pope et al. reported similar results with use of the Salter, Chiari, and Steel osteotomies. Root et al., in their comprehensive review of the results in thirty-one patients (thirty-five hips) who had perinatal static encephalopathy, also reported that combined open reduction and pelvic and femoral osteotomy predictably resulted in long-term improvement of stability and sitting balance as well as relief of pain. However, many of their patients were skeletally mature, and only one had had a pelvic osteotomy of the Pemberton type. Also, in contrast to our findings, radiographic evidence of avascular necrosis developed in eight of the thirty-five hips postoperatively. Mubarak et al. used a comprehensive approach that was similar to ours, but the pelvic osteotomy was a variant of the Dega type; the results were excellent in seventeen of eighteen hips.
Computerized tomographic studies have demontrated that patients with perinatal static encephalopathy have a global rather than a purely anterior acetabular deficiency18. This finding suggests that pelvic osteotomies other than the Pemberton type may be unsuitable for such patients because the procedures have the potential to induce posterior instability. For this reason, we believe that caution is warranted when recommending pelvic redirectional osteotomies and capsular arthroplasties for patients who have acetabular dysplasia and neuromuscular disease. However, this caution is not necessary for the Pemberton osteotomy.
For eight years, we have used the Pemberton acetabuloplasty as part of the comprehensive operative treatment of hip dysplasia secondary to perinatal static encephalopathy, and we have been very encouraged by the results. Surprisingly, we did not find that the younger patients in the current series benefited appreciably more from this procedure; the radiographic improvement in the older patients was as good as that in the younger ones. The Pemberton osteotomy corrects acetabular dysplasia in a patient who has neuromuscular disease and a stable hip and it increases the stability of a concentrically reduced hip, but it must not be used to reduce a dislocated hip. When the teardrop is substantially widened preoperatively, adequate coverage of the femoral head is difficult to attain with use of the Pemberton osteotomy. A shelf or a Chiari acetabuloplasty may produce better results in this subgroup of patients. Several authors2,4,7,11,18 have suggested that redirectional pelvic osteotomies may lead to a loss of posterior coverage of the femoral head in patients who have dysplasia of the hip secondary to neuromuscular disease. To date, we have not found evidence of this and we have not seen posterior dislocation or subluxation in any patient.
This study did not address the complex question of whether operative intervention is warranted in patients who have dysplasia of the hip secondary to neuromuscular disease1,3-5,10. However, we noted an improvement in the quality of life for our patients, with relief of pain and easier perineal care, as well as improved sitting, leading to greater mobility in a wheelchair. This improvement seems to be sustained, with no indication of deterioration over the duration of the study.
We have found that, when combined with appropriate soft-tissue releases, open reduction of the hip as needed, and correction of pathological femoral valgus angulation and anteversion, the Pemberton osteotomy is effective for the treatment of acetabular dysplasia in patients who have perinatal static encephalopathy. We found no appreciable adverse clinical effects in our review of the results of this comprehensive operative treatment of the fifty-two hips in this series.