TOTAL HIP REPLACEMENT IN CHILDREN WITH ARTHRITIS

Transcription

1 ARTHRITIS 40 1 & RHEUMATISM OFFICIAL JOURNAL OF THE AMERICAN RHEUMATISM ASSOCIATION SECTION OF THE ARTHRITIS FOUNDATION TOTAL HIP REPLACEMENT IN CHILDREN WITH ARTHRITIS B. H. SINGSEN, A. S. ISAACSON, B. H. BERNSTEIN, M. J. PATZAKIS, H. K. KORNREICH, K. K. KING, and V. HANSON Fourteen children with juvenile rheumatoid arthritis and two with ankylosing spondylitis received 29 total hip replacements (THR). The ages at THR were 12 to 18 years, the median duration of prior hip disease was 7.1 years, and the minimum followup was one year (range 1-4 years, median 2.2 years). The primary indication for THR was pain in 5 hips and severe malposition or flexion contractures in 24 hips. (Sixteen of these also had some degree of pain.) All 29 hips demonstrated improved postoperative From the Divison of Rheumatology and Rehabilitation, Childrens Hospital of Los Angeles, and Rancho Los Amigos Hospital, Downey, California. Supported in part by a grant from the Southern California Chapter of the Arthritis Foundation. B. H. Singsen, M.D.: Assistant Clinical Professor of Pediatrics, Childrens Hospital of Los Angeles, Rancho Los Amigos Hospital, and the University of Southern California, School of Medicine; A. S. Isaacson M.D.: Clinical Professor of Orthopedic Surgery, Childrens Hospital of Los Angeles, and the University of Southern California, School of Medicine; B. H. Bernstein, M.D.: Associate Clinical Professor of Pediatrics, Childrens Hospital of Los Angeles, and the University of Southern California, School of Medicine; M. J. Patzakis M.D.: Associate Professor of Orthopedic Surgery, Childrens Hospital of Los Angeles, Rancho Los Amigos Hospital, and the University of Southern California, School of Medicine; H. K. Kornreich M.D.: Associate Professor of Pediatrics, Childrens Hospital of Los Angeles, and the University of Southern California, School of Medicine; K. K. King, M.D.: Associate Clinical Professor of Pediatrics, Childrens Hospital of Los Angeles, and the University of Southern California, School of Medicine; V. Hanson M.D.: Professor of Pediatrics, Childrens Hospital of Los Angeles, and the University of Southern California, School of Medicine: Address reprint requests to B. H. Singsen, M.D., Childrens Hospital of Los Angeles, 4650 Sunset Boulevard, Los Angeles, California, Submitted for publication October 11, 1977; accepted December 30, range of motion, and all were free of pain. In children, active rheumatic disease in other joints, prosthesis longevity, and possible lack of adequate motivation all merit special consideration prior to THR, but the results are encouraging. The indications for, and applications of, total hip replacement (THR) surgery have constantly expanded since its introduction by Charnley (1) and others (2,3) 15 years ago. An acceptable lower age limit for the use of THR still has not been determined (4). Several reports, however, now suggest that THR for selected patients with juvenile rheumatoid arthritis (JRA) or ankylosing spondylitis (AS) in their third decade of life may be successful and highly beneficial (5-10). Children in their teens were only rarely noted in these reports. Appropriate caution about impairing growth and concern about prosthetic component life have limited the advocation of THR for young patients. However, severe disability, resulting from advanced involvement of the weight bearing joints in teenagers with juvenile rheumatoid arthritis or ankylosing spondylitis, may hinder education, socialization, and employment. In the present report, we describe 16 children less than 18 years old who received 29 THR because of severe destructive hip disease due to JRA or AS. The objectives of this study were to assess the indications for, value of, and frequency of complications following THR in children less than 18 years of age who have active JRA or AS. Arthritis and Rheumatism, Vol. 21, No. 4 (May 1978)

2 402 SINGSEN ET AL Patients PATIENTS AND METHODS Between 1972 and May 1977, 22 patients with juvenile onset arthritis from the pediatric rheumatic disease clinics of the Childrens Hospital of Los Angeles and Rancho Los Amigos Hospital in Downey, California received 40 total hip replacements. The two criteria for inclusion in this study were age less than 18 years old and a minimum of one year postoperative followup. Sixteen children who received 29 THR remained available for study. There were 8 girls and 6 boys with severe, disabling juvenile rheumatoid arthritis; 9 of these children had experienced a systemic onset and 5 a polyarticular onset. Two boys had ankylosing spondylitis. The age range at disease onset, for the entire group, was 1 to 13 years old (median and mean both 6.8 years). The duration of overall chronic arthritis was 3.5 to 13.9 years (median 8.4 years, mean 8.6 years), and the duration of clinically or roentgenographically evident hip disease to the time of total hip replacement was 2 to 12 years (median and mean both 7.1 years). The age range at the time of the procedure was 12 to 18 years (median age 15.2 years). Thirteen children received bilateral THR; 3 boys, 2 with JRA and 1 with AS, received unilateral THR. Each child in our clinics receives frequent, comprehensive evaluation from a team which includes: a pediatric rheumatologist, an orthopedic surgeon, nurses, a social workiir, physical and occupational therapists, and psychologist when indicated. Degree of pain and disability, suitability of the bony structures to accommodate prosthetic elements, growth and epiphyseal status, age, and motivation are all the subject of review. Some children have been found not to be suitable candidates for total hip replacement because of lack of adequate bone to seat the prosthetic elements, because of upper extremity deformity, pain, or weakness that would not permit assistance of weight bearing during the postoperative rehabilitation period, or because the team believes a child demonstrates inadequate motivation. These children are deferred for reconsideration at a later time. Each child had received extensive medical, physical, and occupational therapy during the median 7.1 years of active hip disease prior to THR. The indication for THR in each case was marked disability due to some combination of the following: I) pain, 2) disability from severe deformity or malposition of the hip, 3) loss of motion from extended nonambulation which restricted self care activities. Postoperative evaluation of hip function was performed according to the method suggested by Harris (1 1); because of the retrospective nature of this study, formal preoperative evaluation was not available. Operative Technique Intravenous antibiotics were given for 24 hours preoperatively. The hip joints were exposed through a posterolateral or lateral approach. Because of significant alteration in the rotatory anatomy of the femur, trochanteric osteotomy was usually performed (26 of 29 hips). Charnley or Charnley-Mueller type prostheses were used; however, these were frequently custom made because of small femoral size in those children with marked growth failure. When severe flexion contractures were present the iliopsoas tendon was transected. Similarly, adductor tenotomy was performed if preoperative abduction or abduction following reduction of the replaced hip was less than 15". All 29 cases received 3 days of postoperative intravenous antibiotics followed by 5 days of oral antibiotics. An abduction pillow remained in place until control of the leg was regained. Brief periods of standing were usually achieved by 48 hours after surgery and walking with support by 72 hours. Anticoagulation was not employed, and there were no clinically evident instances of thrombosis or embolization. RESULTS Five of the children had had a total of 15 prior hip operations. These included 8 synovectomies, 6 soft tissue releases, and I cup arthroplasty. The 14 children with juvenile rheumatoid arthritis had also had more than 50 other previous operations; most commonly these were synovectomies of the knees, wrists, metacarpophalangeal, and proximal interphalangeal joints. A number of antiinflammatory or immunosuppressive medications were administered to the 16 children during the median 8.4 years of chronic arthritis (median 7.1 years of hip disease) prior to total hip replacement. These included aspirin in all 16 and gold sodium thiomalate in 14 (excluding the 2 boys with ankylosing spondylitis). Corticosteroids had previously been used in 12 cases, but only 3 children were receiving them at the time of this operation (prednisone in doses of 13 mg, 10 mg, and 3 mg administered every other day). Cyclophosphamide had been used in 5 children, and one child received azothioprine; neither was in use at the time of this procedure. Roentgenographic assessment of the 29 hips prior to THR revealed that all had proximal femoral epiphyseal closure or advanced destruction, indicating lack of further growth potential at that site. However, epiphyses frequently were open elsewhere, depending upon the degree of inflammation in other involved joints. Narrowing of the hip joint space, subchondral cysts, erosions, subluxation, thinning of the acetabular roof, and coxa valga were present to varying degrees in all of the hips. The primary indication for total hip replacement was pain alone in 5 of 29 cases. In 16 cases the operation was indicated because of some combination of pain and severe malposition, and in 8 cases it was performed primarily because of severe malposition and loss of function at the hip. Twenty of the 29 operations were followed by brief, low grade fever (1383 C); blood cultures grew no

3 ,.*' HIP REPLACEMENT IN CHILDREN 403 pathogens. Only 3 patients developed superficial wound infections which responded to oral antibiotics and local wound care. There have been no deep wound infections nor evidence of late loosening of the prosthetic components. Late complications have included bilateral fe; moral shaft stress fractures ("greenstick" type, site at lower one-fourth of femoral component shaft) which occurred 8 months postoperatively in one boy with ankylosing spondylitis who had been nonweight bearing for 2 years and then had a marked increase in activity following bilateral total hip replacement. These healed easily and without sequelae. An acetabular cement fracture and a femoral shaft stress fracture occurred 18 and 24 months postoperatively, respectively, in the other boy with ankylosing spondylitis; both were also related to excessive exercise. The cement fracture is roentgenographically visible but has caused no persistent pain or component subluxation. The femoral fracture, at the tip of the prosthesis, has healed very slowly and with only fair alignment. One girl with 60" flexion contractures had an acetabular component subluxation immediately after surgery; she is pain free, functionally independent, and does not wish revision. One other girl has intermittently complained of a painful trochanteric wire. The preoperative ambulatory status of the 16 children is displayed in Figure 1 and compared to the postoperative ambulatory status after a median followup duration of 2.2 years (range: 1 to 4 years). All 13 children with pain (21 hips) experienced complete relief after surgery, and this appeared to strongly contribute to the excellent improvement in ambulatory status. Seven children were complete nonambulators prior to surgery. The one 17-year-old girl who remained nonambulatory had been wheelchair bound for 3 years before surgery; her replacements were performed primarily to improve her transfer ability. The most dramatic improvement was the advance of 8 additional children into limited or full community ambulatory status. A comparison of the pre- and postoperative hip ranges of motion is shown in Table 1. All 29 replaced hips had preoperative flexion contractures which ranged from 10" to 85" (median 35"). The median postoperative improvement in flexion contracture was 15" and included 7 hips where a reduction to zero degrees of contracture was obtained. Before surgery, 10 hips had at least 45" of flexion contracture; at followup, only one hip (in a patient who remained wheelchair bound) had this degree of contracture. The median preoperative hip flexion arc was 25", FULL COMMUNITY LIMITED COMMUNITY WITH SUPPORT HOUSEHOLD AMB WITH SUPPORT NON-AMB -,/',/'.a* /',/*,/',4' 40.,/' /,a',/.' /* d 80 a /',/,,a' 8 80 a,.*'.* I 1 1 I

4 404 SINGSEN ET AL Table 1. Pre- and Postoperative Hip Ranges of Motion and the need for at least a full year of intensive post- Range of Motion Preoperative Postoperative* operative physical therapy. Many children with arthritis have a very limited view of the long-term consequences Flexion contracture Median 35" Median 20" of their disease process. This age-appropriate lack of Range 10" to 85" Range 0" to 45" Hip flexion arc Median 25" Median 61" AbduCtion Range 0" to 80" Median 8" Range 35' to 105" Median 20" Range -15" to 30" Range -5" to 35" * Median at 2.2 years. including 3 hips with 0" of range of motion, and 2 hips with 80" of flexion arc. The median postoperative flexion arc was more than doubled to 61". All 29 hips had equal to or greater than 45" of flexion arc, with one exception of 35". Only 5 of the 29 hips had preoperative abduction of 15" or more; the median for the 29 hips was 8". Postoperatively, the median abduction was increased two-and-one-half times to 20"; only 6 hips had less than 15" of abduction. Harris hip evaluation, which assesses a combination of pain, functional ability, and hip range of motion, was performed in each child at followup (range 1-4 years) and gave a median score of 71 (mean: 70). DISCUSSION Four groups of investigators have previously reported total hip replacement in 109 patients with juvenile rheumatoid arthritis or ankylosing spondylitis under the age of 30 (mean age 25.1 years) (7-10). There were very few teenage patients in these series and the number with active rheumatic disease was not uniformly stated. The patients who comprise the present report were all less than 18 years old and each had active joint disease elsewhere, as well as in the involved hip, at the time of the operation. An assessment of the patients' willingness to cooperate must be attempted prior to total hip replacement in teenagers with active joint disease. Several authors have noted that in patients under 30 the success of THR depends to a great degree on the motivation of the patient (5,6,9). Our experience was similar, but the problem was magnified in the adolescent age group. Young patients, and to a lesser degree their parents, may quickly grasp at this procedure as a solution to many difficulties. This is particularly true if, as suggested by Sledge (lo), the children are desperate to be like their peers. However, the children frequently do not comprehend the limited benefits that may be gained from THR maturity may obstruct the realization that THR is only one small part of the total process of rehabilitation. Conversely, relief from severe pain or deformity during the formative teenage years may be a unique advantage of early hip replacement and allow more appropriate socialization, improved education, and enhanced employment opportunities. For these reasons, it is clear that the health team caring for the juvenile arthritic must exercise unusual care and restraint in deciding which child should be a candidate for total hip replacement. A significant, but largely unknown, percentage of children with juvenile rheumatoid arthritis or ankylosing spondylitis beginning in the first decade of life will develop rapidly progressive hip joint pain, destruction, and deformity such that weight bearing ceases before completion of growth or cessation of disease activity (12,13). In such patients, our experience and that of others (9) suggests that total hip replacement should be considered, if medical and physical therapy are not successful in sustaining ambulation, even in the presence of open epiphyses at the hips. Between the ages of 12 and 18 years the approximate contribution (mean for boys and girls) of normal proximal femoral epiphyses to total linear growth is 30% (14) of 4.5 cm (19, or 1.35 cm. This latter figure is 0.8% of the mean expected height for normal boys and girls at age 18 years (16). Discussion with the majority of our patients indicated that this possible small loss in ultimate height was irrelevant compared to functional improvement. In addition, the proximal femoral epiphysis and acetabular triradiate cartilage removed are rarely normal; both usually demonstrate little or no potential for further growth. Clinically, up to the present time none of the children have had further growth at the hip joint or evidence of resultant prosthetic loosening. In young patients with multiple deformities, total hip replacement may also be of value, as suggested by Bisla and associates (9), because improved weight bearing on knees and ankles with open epiphyses will result in less deformity and improved growth. Several significant technical problems should be considered prior to total hip replacement in adolescents with arthritis. The first is the potential for severe alterations in bone growth at the hip of the juvenile arthritic. If growth failure occurs, custom made prostheses, designed from roentgenographic measurements, may be

5 HIP REPLACEMENT IN CHILDREN 405 required. One-half of our patients received modified Charnley-Mueller prostheses for this reason: Other children have had this operation deferred in hopes of better medical control and further bone growth, or because there was evidence of hip joint restoration (17). Several previous reports of total hip replacement in young patients have described the indication for operation in up to 75% of cases to be extreme hip stiffness (8,9), in contrast to the more common indication of severe hip pain in adult rheumatoid arthritis. Among these 16 children studied, 24 of the 29 hips were replaced primarily because of markedly decreased range of motion or malposition. In our experience the magnitude of preoperative soft tissue contractures has been a limiting factor in the amount of postoperative increase in range of motion that can be expected. The contribution of adductor and ileopsoas tentomy and capsulectomy to improved postoperative range of motion in these children has been disappointing in our experience and in others (6,7). Therefore, we have placed emphasis on an extensive program of pre- and postoperative physical therapy to increase range and strength. A third consideration is implant and acrylic cement life expectancy, and possible long-term side effects (18). In older patients the components of a total hip replacement have been shown to have acceptable wear characteristics at least for the first 10 years (19). A more recent study of young adults suggested that, after a mean followup of 39 months, the rate of component wear was similar to that for older patients and the authors hypothesized years of function before wear might present a problem (8). Most probably, the restricted activity of children with active polyarticular joint disease and soft tissue limitations will also reduce component wear rates. The 2 patients with ankylosing spondylitis had the best result from THR among the 16 children, in contrast to other observations of only poor to fair results following THR in young patients with ankylosing spondylitis (5,6,9). The marked improvement in these 2 children represents a small number but suggests that THR in young males with ankylosing spondylitis and severe hip disease requires continued evaluation. The complete absence of postoperative deep wound infections in these patients has been gratifying. In agreement with observations of other authors (6, lo), our experience suggests that past immune suppressant therapy and current low dose corticosteroids probably do not jeopardize the success of THR in children, provided that meticulous scrutiny is maintained for signs of infection. In addition, postoperative antibiotic pro- phylaxis at the time of dental extractions, or other situations with a risk of bacteremia, is advisable for many years. Undoubtedly, more extended followup will be required to assure these points. The retrospective nature of this study precluded a formal comparison of pre-and postoperative hip function of the type described by several authors (8,20). Harris hip evaluation (1 1) scores were, however, comparable to those measured by other methods in several series of young adults who received THR because of juvenile rheumatoid arthritis or ankylosing spondylitis (5,6,8,9). The uniform relief of hip pain, moderate improvement in hip ranges, and expansion of ambulatory capacity led to a significant improvement in the quality of life for our patients. These findings suggest that total hip replacement should be considered for carefully selected teenage patients with active juvenile rheumatoid arthritis or ankylosing spondylitis. ACKNOWLEDGMENTS Grateful appreciation is expressed to the many orthopedic surgeons and pediatricians who participated in the care of our patients; particularly, James D. Mays, M. Mark Hoffer, John C. Wilson, and Donna M. Barras. The secretarial assistance of Ms Valerie Valle is greatfully acknowledged. REFERENCES 1. Charnley J: Total hip replacement by low-friction arthroplasty. Clin Orthop 72:7-21, McKee GK, Watson-Farrar J: Replacement of arthritic hips by the McKee-Farrar prosthesis. J Bone Joint Surg 48B , Ring PA: Replacement of the hip joint. Ann R Coll Surg Engl48: , Coventry MB: Surgery for arthritis of the hips. Bull Rheum Dis 22: , Arden GP, Taylor AR, Ansell BM: Total hip replacement using the McKee-Farrar prosthesis in rheumatoid arthritis, Stills disease and ankylosing spondylitis. Ann Rheum Dis 29:14, Arden GP, Ansell BM, Hunter MJ: Total hip replacement in juvenile chronic polyarthritis and ankylosing spondylitis. Clin Orthop 84: , Arden GP Surgical treatment of Stills disease. Ann Roy Coll Surg Engl 53: , Halley DK, Charnley J: Results of low-friction arthroplasty in patients thirty years of age or younger. Clin Orthop 112: , Bisla RS, Inglis AE, Ranawat CS: Joint replacement surgery in patients under thirty. J Bone Joint Surg 58A: , 1976

6 SINGSEN ET AL 10. Sledge CB: Joint replacement surgery in juvenile rheumatoid arthritis. Arthritis Rheum 20: , Harris WH: Traumatic arthritis of the hip after dislocation and acetabular fractures: Treatment by mold arthroplasty: An end-result study using a new method of result evaluation. J Bone Joint Surg 51A: , Jacqueline F, Boujot A, Canet L: Involvement of the hips in juvenile rheumatoid arthritis. Arthritis Rheum 4: , Isdale IC: Hip disease in juvenile rheumatoid arthritis. Ann Rheum Dis 29: , Caffey J: Pediatric X-ray diagnosis. Vol 2. Sixth Edition. Chicago, Year Book Medical Publishers, 1972, p Anderson M, Green WT, Messner MB: Growth and pre- dictions of growth in the lower extremities. J Bone Joint Surg 45A:1-14, Reed RB, Stuart HC: Patterns of growth in height and weight from birth to eighteen years of age. Pediatrics 24: I, Bernstein B, Forrester D, Singsen B, King KK, Kornreich H, Hanson V: Hip joint restoration in juvenile rheumatoid arthritis. Arthritis Rheum , Urist MR: Acrylic cement stabilized joint replacements. Curr Probl Surg November: 1-54, Charnley J, Cupic Z: The nine and ten year results of the low friction arthroplasty of the hip. Clin orthop 95:9-25, Salvati EA, Wilson PD, Jr: Long-term results of femoral head replacement. J Bone Joint Surg 55A: , 1973