Total hip replacement in patients with Down syndrome and degenerative osteoarthritis of the hip

Down syndrome was first described in 1866 by John Langdon Down in an essay describing a group of children with common features and mental retardation.¹ In 1959, the cause of Down syndrome was independently determined by Lejeune et al² and Jacobs et al³ to be a triplication (trisomy) of the 21st chromosome as a result of chromosomal non-disjunction, translocation, or mosaicism that occur during meiosis. It is the most common survivable chromosomal abnormality in humans. One in 691 live births in the United States is affected with Down syndrome.⁴ The incidence increases substantially with increasing maternal age.⁵ It is characterised by the variable presentation of several associated abnormalities including mental retardation, characteristic facial and body features (flat face with an upward slant to the eye, a short neck, small ears, a large tongue, short stature, a simian palmar crease, and clinodactyly), and numerous visual, neurological, cardiac, gastrointestinal and endocrine manifestations as well as haematological malignancies. Several musculoskeletal manifestations related to hypotonia, decreased bone density, and generalised ligamentous laxity have been described, including instability of the cervical spine^6-9 and hip,^10-13 pes planus,^14-15 and sandal gap deformity of the great toe.^14-15

Many factors may predispose the hip to degenerative joint disease in Down syndrome. Pathology affecting the hip is common, occurring in between 8% and 28% of patients with Down syndrome.¹⁶ There may be acetabular dysplasia and posterior acetabular deficiency.^10-13,16 Instability of the hip is seen in between 5% and 14% of patients with Down syndrome.^10-13,16 The presence of mental retardation, hypotonia, decreased bone density, generalised ligamentous laxity, and the deformity of the hip may increase the complexity of total hip replacement (THR) in these patients. Previous reports of THR in patients with Down syndrome describe good outcomes.^16,17 One recent multicentre series of 26 THRs in patients with Down syndrome reported 85% survival and a mean Harris hip score¹⁸ (HHS) of 84 at a mean follow-up of eight years.¹⁹ The purpose of our study is to report the experience of one centre with THR in patients with Down syndrome and degenerative joint disease with special emphasis on the clinical outcome and the reasons for -failure of THR.

Patients and Methods

Review of a registry that has prospectively -followed all patients who have undergone joint replacement at our institution from 1969 to 2009, identified a consecutive group of patients with Down syndrome who had undergone THR. Patients were scheduled for clinical evaluation at one, two, and five years post-operatively and every five years thereafter. In all, 14 patients with Down syndrome (six males and eight females) with 21 primary THRs performed for degenerative joint disease with a minimum follow-up of two years were included. The operations were performed by 11 surgeons with subspecialty training in THR. Patients with Down syndrome undergoing primary THR for pathology other than degenerative joint disease of the hip, such as a fracture of the femoral neck, were excluded. The mean age at operation was 41.7 years (standard deviation (sd) 12.9; 17 to 70). Their mean height was 151.9 cm (sd 5.8), the mean weight was 71.7 kg (sd 15.7) and mean body mass index was 31.1 kg/m² (sd 6.8). A total of 20 hips (13 patients) were dysplastic. According to the Crowe classification,²⁰ seven were grade I, six grade II, six grade III and one grade IV. The remaining patient (one hip) had prior slipped capital femoral epiphysis. In seven patients, bilateral THR was performed and the remainder underwent unilateral THR. In all, ten THRs were performed with an anterolateral approach, nine with a posterior approach and two with a transtrochanteric approach. The mean overall operating time was 158 minutes (sd 45). The acetabular components were modular cementless in 19 THRs and all-polyethylene cemented in two (Table I). A mean of two screws was used to fix the cementless acetabular components. Femoral head autograft secured with two cancellous screws was used in four hips, all of which were Crowe grade III. Six of the articulations were metal on conventional polyethylene, 14 were metal on cross-linked polyethylene, and one was ceramic on cross-linked polyethylene. In one patient, constrained acetabular liners were used bilaterally. No femoral shortening osteotomies were required. A total of three hips had cemented femoral components and the remaining 18 had cementless femoral components (Table I).

The diameter of the femoral head was 22 mm in eight, 26 mm in one, 28 mm in seven, 32 mm in four and 36 mm in one THR. A total of six of the eight 22 mm heads, the 26 mm head, and two of the 12 which were ≥ 28 mm diameter were used prior to 2003, while two of the eight 22 mm heads and ten of the 12 ≥ 28 mm heads were used after 2003. The intra- and post-operative outcomes including complications and revision surgery were recorded. The HHS was calculated from clinical examinations and patient surveys.¹⁸ The post-operative and latest radiographs of all hips were reviewed (DFA) for fixation according to De Lee and Charnley²¹ for acetabular components, Gruen et al²² for cemented femoral components, Engh et al²³ for un-cemented femoral components, and Stilling et al²⁴ for wear. The study had ethical approval.

Statistical analysis

All values are reported as the mean, sd and range. Comparisons of outcomes were performed using a two-tailed Student’s t-test or chi-squared analysis as appropriate. All calculations were performed in Microsoft Excel (Redmond, Washington). The primary endpoint was revision surgery and the reasons for failure were documented. A survival curve was generated using revision THR for any cause as the endpoint. Statistical significance was set at a p-value < 0.05.

Results

The mean clinical follow-up was 5.8 years (sd 4.7; 2 to 17) and the mean radiological follow-up was 4.3 years (sd 4.8). A total of four patients had died at final follow-up at 11, 14, 18, and 31 years post-operatively.

The mean HHS was 38.0 (sd 8.1; 27 to 51) pre-operatively and 91.4 (sd 6.9; 83 to 100) (t-test, p = 1 x 10^-9) at final follow-up, excluding three patients (four THRs) who underwent revision. One patient with Alzheimer’s disease had a persistently low HHS of 49 at final review despite radiologically well-fixed bilateral THRs.

During the period of the study, there was a significant increase in femoral head diameter after 2003 (χ² = 7.1, p = 0.029).

There were few intra- and post-operative complications. No patient suffered a pulmonary embolus, dislocation, periprosthetic fracture, or deep infection. There were two intra-operative fractures. One nondisplaced fracture of the greater trochanter was identified on the post-operative films and treated non-operatively. One fracture of the calcar required intra-operative cerclage wiring and limitation of weight-bearing post-operatively. It healed but the uncemented stem subsided and was revised after three years. One patient had a deep venous thrombosis nine weeks post-operatively which required anticoagulation.

A total of four THRs in three patients were revised for aseptic loosening (Fig. 1). The mean time to revision was 7.7 years (sd 6.3; 3 to 17). In addition to the femoral failure described above, one cemented all-polyethylene acetabular component failed after 17 years and required revision. In one patient, two vertically placed dual geometry Omnifit (Stryker, Kalamazoo, Michigan) acetabular components with conventional poly-ethylene failed after three and five years, requiring revision THR. Two of the three patients requiring revision THR had died at the time of this review. The remaining THRs were well fixed with intact cement mantles or evidence of bony ingrowth and with < 2 mm of acetabular wear at the most recent radiological follow-up. The overall patient survival rate was 81.0%, with four hips undergoing revision for any reason.

Discussion

Our results demonstrate a high rate of aseptic loosening after THR in patients with Down syndrome. However, the improved mean HHSs indicate that these patients benefit from THR and, when appropriate, should be given the opportunity to undergo this procedure.

We acknowledge the limitations of our study. The surgery was performed by 11 different surgeons, all fellowship trained in arthroplasty surgery but with differing approaches to the procedure, and it is likely that significant modifications of technique have occurred over time, as demonstrated by the use of larger diameter femoral heads after 2003. Additionally, the HHS is not validated for use in patients with Down syndrome. It may not be the best outcome measure in patients with many physical and mental problems. However, it has a low respondent and minimal examiner burden and has shown a better potential than other scores when assessing elderly patients with fractures of the hip who present with co-morbidities, such as dementia.²⁵ There was a significant improvement in the mean HHS following THR in patients with Down syndrome. Two other small retrospective case series report similar improvements with a mean follow--up of almost ten years.^17,19

We identified only one post-operative complication of an isolated deep vein thrombosis. A small retrospective series of nine THRs in nine patients with Down syndrome reported no dislocations,¹⁷ and another retrospective study of 26 THRs in 21 patients with Down syndrome had one dislocation.¹⁹There were no dislocations in our series.

While the rate of survival in this series is slightly lower than the previously published rates of 85% to 89% for these patients, it is consistent with the challenges of treating young patients with dysplasia of the hip. All four revision THRs were for aseptic loosening, three on the acetabular side (one cemented all-polyethylene component and two vertically placed press-fit components) and one on the femoral side in a patient with early subsidence following an intra-operative fracture of the calcar. Technical intra--operative problems seem to be the main cause of aseptic loosening in our series.

In conclusion, THR in patients with Down syndrome provides satisfactory clinical results at mid-term follow-up, with the failures in this series attributable to implant specific and intra-operative decisions.

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.

This article was primary edited by G. Scott and first proof edited by J. Scott.