Patients who have neuromuscular disease may need a hip replacement for two general reasons: either the disease process (for example, cerebral palsy or myelomeningocele) has led to dysplasia of the hip, which in turn has evolved into degenerative joint disease, or the degenerative joint disease has developed independently of the neurological condition (for example, Parkinson disease).
In general terms, neuromuscular conditions can be divided into two categories: (1) those that are accompanied by decreased muscle tone and (2) those that are associated with increased muscle tone as well as contractures or movement disorders. The first category includes conditions such as poliomyelitis, Down syndrome, and myelomeningocele, and the second group includes conditions such as cerebral palsy, Parkinson disease, and stroke.
*Printed with permission of the American Academy of Orthopaedic Surgeons. This article, as well as other lectures presented at the Academy's Annual Meeting, will be available in March 2000 in Instructional Course Lectures, Volume 49. The complete volume can be ordered online at www.aaos.org, or by calling 800-626-6726 (8 A.M.-5 P.M. Central 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.
‡Department of Orthopedic Surgery, Mayo Clinic, 200 First Street S.W., Rochester, Minnesota 55905.
§Department of Orthopedics, Kantonspital, University of Bern, 3010 Bern, Switzerland.
Poliomyelitis
We are not aware of any reports of hip replacement being performed in the involved extremity of a patient who had residuals from poliomyelitis and major paralysis. This might be related to the fact that forces across the joints in these patients are substantially decreased and, therefore, degenerative disease is relatively rare in these flaccid limbs. A review of the Joint Registry at the Mayo Clinic revealed no instance of total hip replacement being performed on the involved hip of a patient who had residuals of poliomyelitis. However, one of us (M. E. C.) performed five hip replacements contralateral to hips affected by poliomyelitis, and the results in these patients (after durations of follow-up of two to eight years) appeared to be no different from those in the general population of patients managed with hip replacement. Although the loads on these contralateral hips are somewhat abnormal, the patients generally are not very active on account of their residual infirmity.
Down Syndrome
Total hip replacement may be necessary in patients who have Down syndrome primarily because dysplasia of the hip, which can eventually lead to degenerative disease, frequently develops in these patients. With a search of the Mayo Clinic Joint Registry, we identified five hips in three patients with Down syndrome who were managed with total hip replacement. After a duration of follow-up ranging from five to fifteen years, the results were excellent after all three of the hip replacements that had been performed with cement, but the two sockets that had been implanted without cement had been revised. There were no dislocations or other complications. On the basis of this relatively small and anecdotal experience, it would appear that hip replacement in patients who have Down syndrome is justifiable if the patients are reasonably functional and sufficiently symptomatic.
Myelomeningocele
We are not aware of any study involving patients with myelomeningocele and dysplasia who were managed with joint replacement. Again, reviewing our Joint Registry, we identified three patients (three hips), all with a low-lumbar-level myelomeningocele, who had hip replacement because of severe pain. In this group of patients, all of the prostheses performed poorly primarily because of pain and instability after five to ten years of follow-up16. Reasonable muscle power of flexion, abduction, and perhaps extension is necessary for a satisfactory result of a hip arthroplasty. If muscle power is suboptimum, then alternatives other than arthroplasty should be considered for patients who have paralytic hips resulting from myelomeningocele. Total hip arthroplasty should be reserved for patients who have symptomatic arthritis of the hip but preferably without evidence of neuropathic arthropathy and with adequate muscles around the joint.
Cerebral Palsy
The reason to consider hip arthroplasty for a patient who has cerebral palsy is generally related to subluxation or dislocation of the hip, the prevalence of which varies widely in patients with this disease. In one of the largest published series that we know of, Samilson et al.11 managed 274 patients who had cerebral palsy and subluxation or dislocation of the hip. These patients accounted for 27 percent of a total group of 1013 patients who had cerebral palsy. Instability of the hip is thought to be due to muscle imbalance, contracture, coxa valga, femoral anteversion, and an increased acetabular index7. Treatment options for a painful degenerative hip joint in an adult include resection arthroplasty, arthrodesis, and replacement arthroplasty.
Resection arthroplasty1 needs to be very extensive to relieve pain and deformity reliably and, therefore, is reserved for patients who are unable to walk and whose functional needs might include improved hygiene and ease of positioning. Arthrodesis is an option that is indicated rarely and only for patients who cannot walk. Total hip arthroplasty is believed to offer the highest probability of pain relief and satisfactory motion. The concerns with hip arthroplasty include the patient's age, which is often very young; the abnormal muscle strength; the spasticity and contractures that are often present; and the cooperation on the part of the patient, which may be poor. On the other hand, patients who have cerebral palsy usually have a low level of activity and often use crutches full time; thus, loads across the hip joint might be decreased.
The outcome of total hip arthroplasty in patients who have cerebral palsy is not well represented by the commonly used hip-rating scores because many of the functional variables (such as walking distance, need for support, mobility, limp, and ability to climb stairs) are substantially affected by the cerebral palsy itself. The most valuable factors in assessing the result in a patient who has cerebral palsy are pain relief and walking status.
To our knowledge, the first report of total hip replacement in patients who had cerebral palsy was by Buly et al.2, in 1993, and included eighteen patients in whom a total of eighteen total hip arthroplasties had been performed with cement. At an average of ten years, seventeen of the eighteen patients had pain relief and the ten-year rate of prosthetic survival was 95 percent. However, sixteen of the eighteen patients had been managed with postoperative immobilization in a spica cast, and two patients had recurrent dislocations.
Our review of the Mayo Clinic experience with total hip arthroplasty in patients who had cerebral palsy identified sixteen patients who had been managed with a total of sixteen total hip replacements and had been followed for a minimum of two years17. There were ten men and six women, and the average age at the time of the operation was 48.5 years (range, twenty-two to seventy-nine years). The type of cerebral palsy was hemiplegia in seven patients, diplegia in four, quadriplegia in two, athetoid quadriplegia in two, and athetoid diplegia in one. The diagnosis leading to the total hip arthroplasty was primary degenerative joint disease in three patients; arthritis secondary to hip dysplasia in eight, three of whom had had previous operations; and failed treatment of a fracture of the femoral neck in five.
The preoperative pain was severe in five patients, moderate in nine, slight in one, and unknown in one. The preoperative functional assessment showed that three patients were unable to walk, five were able to walk indoors only, six were able to walk about the community, and one patient was able to walk an unlimited distance; the walking status of the last patient was unknown. The operation was performed through an anterolateral approach in eight patients, a transtrochanteric approach in seven, and a posterolateral approach in one.
Both components were fixed with cement in twelve patients, both were fixed without cement in two, and the socket was fixed without cement and the stem was fixed with cement (hybrid fixation) in two. An adductor tenotomy was performed in two patients. In two patients, the potential for instability was thought to necessitate the use of protective external stabilization: one of the patients was managed with immobilization in a spica cast, and the other patient used a hip guide brace.
The duration of clinical follow-up averaged almost ten years, with a range of 2.5 to twenty-one years. Fifteen patients were available for follow-up, and one had died. At the time of the latest follow-up, pain relief was complete in eleven patients, two patients had slight pain, and one patient reported moderate diffuse pain in the lower limb that was thought to be unrelated to the prosthesis. One patient had had a revision at thirteen years. Thus, overall, thirteen of the fifteen patients had pain relief.
Nine patients had an improvement in the walking status, five had no change, and the walking status of one patient was not known. Two of the five patients had had an initial improvement in the walking status, but it later returned to the preoperative level.
Intraoperative complications included one trochanteric fracture and one acetabular fracture on press-fit impaction of a component that was inserted without cement. There was also one trochanteric avulsion after a transtrochanteric approach, one ulcer on the heel of a patient who wore a cast, one instance of ileus with spontaneous resolution, and one case of urinary retention. Three patients had a reoperation: one of these operations was for fixation of the avulsed trochanter, one was a tenotomy that was performed because of adductor spasticity, and one was a revision because of aseptic loosening of both components thirteen years after the primary procedure.
Radiographic examination was performed on eleven patients at an average of eight years (range, 2.5 to twenty-one years) after the operation. Ten of the femoral components were solidly fixed, including the stem, which had been inserted without cement, in the patient who reported diffuse pain. One stem was associated with some lucency, but it did not cause symptoms. Nine of the acetabular components were stable; one, which had been implanted in a patient who had slight pain and was unable to walk, was associated with circumferential lucency; and one cup was loose and caused slight pain in a patient who was able to walk independently.
Thus, the outcome of hip replacement in our group of patients who had cerebral palsy was very satisfactory. A few conclusions can be made on the basis of our experience as well as that reported in the literature1,2. Total hip arthroplasty is a valuable option for patients who have incapacitating hip pain and cerebral palsy. Pain relief and an improvement in function can be expected in a large percentage of patients. Reasonable longevity of the implant can also be expected. Attention should be paid to adductor spasticity that might necessitate adductor tenotomy and to intraoperative instability necessitating postoperative support.
Protection from instability might be afforded by decreasing the inclination and increasing the anteversion of the socket. It also appears that it is better to insert the stem with cement in these patients, who often have very osteoporotic femora (Figs. 1-A, 1-B, and 1-C).
Parkinson Disease
Parkinson disease will develop in approximately 1 percent of people who are more than fifty years old in the general population of the United States5. The annual incidence is 20.5 per 1000, and in one study concomitant dementia was diagnosed three times more frequently in patients who had Parkinson disease than in a control group9. Modern medical treatment can effectively relieve the tremor, rigidity, and akinesia of Parkinson disease. However, the problems related to the impairment of the righting reflexes in the more advanced stages of the disease may be aggravated by the postural hypotension that is the side effect of some of the medications. High rates of postoperative complications and death as well as unsatisfactory results have been reported after the orthopaedic management of patients who had Parkinson disease, particularly after the treatment of fractures about the hip3,4,10,12-14. A mortality rate of 20 percent at six months and a high frequency of complications, including urinary tract infections and pneumonia, were reported after fifty hemiarthroplasties that were done for treatment of a fracture of the femoral neck in forty-five patients who had Parkinson disease13. The reported results of total knee arthroplasty in patients who had Parkinson disease have varied8,15.
We carried out a retrospective review of the cases of the ninety-eight patients with Parkinson disease who were managed with a total of 107 total hip replacements between 1970 and 1994 at the Mayo Clinic. There were seventy-four primary total hip arthroplasties. Fifty-eight of them were done because of osteoarthritis; seven, because of an acute fracture of the femoral neck; five, because of a nonunion of the femoral neck; and four, because of avascular necrosis after a fracture of the femoral neck. The remaining thirty-three total hip arthroplasties were revisions: nineteen were performed to revise a failed femoral endoprosthesis, ten were done because of aseptic loosening of a previous total hip replacement, two were revisions of implants that had failed because of a fracture of the femoral neck, one followed a failed cup arthroplasty, and one followed a failed resection arthroplasty.
To be included in the study, all patients had to have had the diagnosis of Parkinson disease confirmed by a neurologist. The level of clinical disability caused by the disease was classified into five stages according to the system of Hoehn and Yahr6. In stage I, there is unilateral involvement with little or no functional impairment. In stage II, there is bilateral or midline involvement but balance is not affected. In stage III, there is early loss of equilibrium with mild or moderate functional disability. Patients who have stage IV have severe disability and are barely able to stand or walk unaided. Patients who have stage V are confined to bed or a wheelchair unless aided. According to this classification system at the time of the arthroplasty, fourteen procedures were done in patients who had stage-I disability; fifty-two, stage-II; thirty-eight, stage-III; two, stage-IV; and none, stage-V. The preoperative stage of the Parkinson disease of one patient was not recorded.
The study included forty-nine men and forty-nine women. The operation was performed on the right hip of fifty-four patients. The average age at the time of the arthroplasty was seventy-two years, with a range of fifty-seven to eighty-seven years. Seventy-one patients were taking anti-Parkinsonian medications. The severity of pain and any limp, the use of walking aids, and the walking distance were recorded preoperatively, at one year postoperatively, and at the time of the latest follow-up (Table I). The operative approach was anterolateral in fifty-six hips, transtrochanteric in thirty-six, posterolateral in twelve, and direct lateral in three (revision operations). An adductor tenotomy was performed in seven hips and a psoas tenotomy, in one.
These 107 total hip arthroplasties accounted for 0.4 percent of the total hip arthroplasties performed at our institution during the same period. We analyzed the results with emphasis on the functional status, the rate of complications (particularly dislocation), mortality, and the need for additional operative procedures.
Complications: Thirty-eight complications, including dislocations, occurred. There were two pulmonary emboli, one of which was fatal, and one deep hip wound infection, which necessitated removal of the components. There were six dislocations, all of which occurred within the first three postoperative months. Thus, the total rate of dislocation for the entire group was 6 percent (six of 107 arthroplasties), although all but one of the dislocations occurred after revision arthroplasties. Four occurred after revision of a failed femoral endoprosthesis; one, after revision of an aseptically loose hip replacement; and one, after an arthroplasty performed for an acute fracture of the femoral neck after failure of internal fixation. One other patient had late instability that developed after a cerebrovascular accident. Closed reduction was performed in all of these patients, and one had a revision of the acetabular component. One patient was managed with a hip spica cast and one, with a hip guide brace. The other complications in this series included trochanteric nonunion (four), deep-vein thrombosis (three), hematoma (one), urinary tract infection (eight), pneumonia (three), transient peroneal nerve injury (two), cerebrovascular accident (two, including the patient who had late instability), postoperative confusion (four), ileus (one), and gastrointestinal hemorrhage (one).
Mortality: Four patients died postoperatively: two died of pneumonia; one, of a cerebrovascular accident; and one, of a massive pulmonary embolus. The rate of mortality at six months was 6 percent (six of ninety-eight patients) for the entire group. At the latest evaluation, fifty-one of the ninety-eight patients had died.
Reoperations: Eight patients had a total of nine reoperations. The reasons for the operations included a hip wound infection and subsequent removal of the components, periprosthetic fracture, trochanteric nonunion, removal of trochanteric wires, late instability, and aseptic loosening of the femoral or acetabular component, or both. The rate of prosthetic survival without a reoperation at five years was 93 percent (confidence interval, 87 to 99 percent).
Functional results: The results were evaluated with regard to pain relief and functional improvement. The findings with regard to pain at the preoperative examination differed significantly from those at the time of the latest follow-up (p < 0.05) in that 93 percent (sixty-two) of the sixty-seven patients (seventy-five hips) who had been followed for a minimum of two years had pain relief at the time of the latest follow-up (Table I). Likewise, functional improvement, as evidenced by a decrease in limping and in use of walking aids and an increase in walking distance, was significant (p < 0.001) at one year postoperatively compared with preoperatively. However, function deteriorated over time, with a significant increase, at the time of follow-up compared with the status at one year, in limping and in the use of walking supports and a decrease in walking distance, for the entire group of patients. In parallel with this deterioration, disability related to the Parkinson disease increased and definite progression of the neurological deficit was noted. At the time of the latest follow-up, forty-three (57 percent) of seventy-five hips were in patients who had progression of the Parkinson disease to functional stage IV or V (Table II).
Radiographic results: Only forty-three patients could be evaluated radiographically. At the time of the latest follow-up, more than half of the patients were severely debilitated and many of them elected not to have radiographs made because of the lack of symptoms and, therefore, the lack of therapeutic consequences. Excluding the revised components, no radiographically loose components were seen.
From this review, we extracted certain conclusions that may be of value and may offer guidance to practicing physicians dealing with severe hip disease associated with Parkinson disease.
1. The prevalence of complications of total hip replacement in patients who have Parkinson disease is relatively high and is greater in patients in whom the neurological disability is more advanced.
2. The prevalence of local complications, particularly dislocation (Figs. 2-A, 2-B, and 2-C), is somewhat increased compared with the prevalence in patients who do not have Parkinson disease. Attention to technical detail when managing these patients cannot be over-emphasized. Adductor tenotomy may be helpful, particularly in patients who have more severe contractures.
3. A higher mortality rate for patients with Parkinson disease can be expected in the first six months after total hip replacement.
4. Finally, and most importantly, progression of Parkinson disease is the rule. Of the 856 patients with Parkinson disease in the study by Hoehn and Yahr6, about 25 percent had severe disability within less than five years after the onset of the disease. At five to nine years this prevalence increased to 66 percent, and at ten to fourteen years more than 80 percent of the patients were severely disabled. Our study confirmed this progression of the neurological condition.
5. On the basis of this information, it appears that total hip arthroplasty with cement should be the procedure of choice in these patients.
Stroke
No information regarding total hip arthroplasty in patients who have had a stroke is available from the literature, to our knowledge. Total hip arthroplasty might be occasionally indicated in the involved limb of a hemiplegic person. Adductor tenotomy and postoperative use of a hip guide brace might be advisable. However, we have had no experience with total hip arthroplasty in patients who have had a stroke.