Abstract
We retrospectively reviewed the five to ten-year results of an isolated proximal femoral osteotomy in two groups of patients. One group consisted of seventeen patients (eighteen hips) who had a residual deformity resulting from congenital dysplasia of the hip; there were fifteen women and two men, and the mean age was thirty-three years (range, twenty-four to fifty-eight years). The other group included sixteen patients (sixteen hips) who had idiopathic osteoarthrosis; there were three women and thirteen men, and the mean age was forty-eight years (range, thirty-eight to fifty-seven years). Both groups were evaluated clinically and radiographically, at a mean of 6.1 years (the patients who had residua of congenital dysplasia) and 6.7 years (the patients who had idiopathic osteoarthrosis).Of the seventeen patients who had residual deformity resulting from congenital dysplasia, eight (eight hips) were completely satisfied with the result of the osteotomy at least five years postoperatively; the joint space was improved in five of these patients (five hips) at the latest radiographic evaluation. Nine patients (ten hips) were not satisfied with the result of the osteotomy at the time of the most recent follow-up; four of them (four hips) had a subsequent total hip arthroplasty.Of the sixteen patients who had idiopathic osteoarthrosis, six were completely satisfied with the result of the osteotomy at least five years postoperatively. The joint space was improved in only two patients at the latest radiographic evaluation. The osteotomy did not provide lasting pain relief for seven patients, all of whom had a subsequent total hip arthroplasty.In both diagnostic groups, the most enduring clinical improvement occurred in hips that had the least evidence of osteoarthrotic changes on the preoperative radiographs. In addition, in the group that had residual deformity resulting from congenital dysplasia, the best outcomes were seen in hips with the least degree of acetabular dysplasia.
Müller stated that the appropriate indications for an isolated intertrochanteric femoral osteotomy include malunion of a fracture of the proximal part of the femur, pseudarthrosis following a fracture of the femoral neck, and deformity of the proximal part of the femur as a result of hip disease in childhood. He proposed that osteoarthrosis of the hip was a less appropriate indication for this procedure. Other authors have reported that the results of intertrochanteric osteotomy for osteoarthrosis can be unpredictable unless the patient meets certain clinical and radiographic criteria9,12. Morscher and Feinsein stated that the ideal candidate for intertrochanteric osteotomy is a patient who is less than fifty years old, has a sedentary job, has a mechanical rather than an inflammatory etiology of the osteoarthrosis, has an improvement in the congruency of the femoral head in relation to the acetabulum on appropriate radiographs made with the hip in the position of simulated correction, and has an arc of hip flexion of at least 80 degrees.
The role of an isolated intertrochanteric osteotomy in patients who have residual deformity resulting from congenital dysplasia of the hip is even more controversial. Most authors have stated that precise indications are still being developed7-9,12, but recent reviews have suggested that a pelvic osteotomy should be done as the primary procedure, to reorient the acetabulum, with or without a femoral osteotomy6,15.
In this retrospective review, we sought to determine the radiographic features that were associated with a favorable or unfavorable clinical result of an isolated femoral osteotomy in two distinct diagnostic groups: patients who had residual deformity as a result of congenital dysplasia of the hip and those who had idiopathic osteoarthrosis of the hip. We analyzed the preoperative and follow-up radiographs to determine the severity of the osteoarthrosis as well as the degree of the dysplasia. We then tried to determine whether there was a relationship between the preoperative radiographic findings and the clinical outcome.
*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, Brigham and Women's Hospital, Ambulatory Building, 75 Francis Street, Boston, Massachusetts 02115.
Data on the Patients
All of the patients were referred to the senior one of us (R. Poss) because of increasing pain in the hip and a decreasing ability to walk and to perform their normal work. Because of their young age, the level of activity that they desired, or their personal preference, we considered them to be candidates for an operative procedure other than a total hip arthroplasty. On the basis of the findings on preoperative radiographs, the patients were divided into two diagnostic groups. Seventeen patients (eighteen hips) had residual deformity as a result of congenital dysplasia. This group included fifteen women and two men, and the mean age was thirty-three years (range, twenty-four to fifty-eight years). Sixteen patients (sixteen hips) had idiopathic osteoarthrosis. This group included three women and thirteen men, and the mean age was forty-eight years (range, thirty-eight to fifty-seven years).
The diagnosis of congenital dysplasia was made according to the criteria of Weinstein. The acetabular depth index2 was abnormal (less than 250) in all of the hips in this group. Idiopathic osteoarthrosis was defined as osteoarthrosis of the hip without known pre-existing anatomical deformity, documented injury, or previous clinical or radiographic evidence of hip disease. Five of the patients with residua of congenital dysplasia had involvement of both hips; none of the patients with idiopathic osteoarthrosis had bilateral involvement.
Evaluation of the Patients
With use of previously published criteria4, we evaluated the patients' level of pain, walking ability, and capability to perform their normal work. We also assessed gait and the abductor mechanics, limb-length discrepancy, and range of motion of the hip. Preoperatively and postoperatively, anteroposterior radiographs of the pelvis and frog-leg lateral radiographs were made at one radiographic center, with the patient supine and thus non-weight-bearing. For the present study, only the anteroposterior radiographs of the pelvis were evaluated. Fluoroscopic examination was used preoperatively to determine the most congruent position of the hip joint6,10,11.
The patients were examined clinically and radiographically at three months, six months, one year, and then every two years. These evaluations were carried out by us, as well as independently by physiotherapists in our joint registry, with use of the clinical criteria recommended by the Société Internationale de Chirurgie Orthopédique et de Traumatologie (SICOT), The American Academy of Orthopaedic Surgeons, and The Hip Society4. The mean duration of follow-up was 6.1 years (range, three to nine years) for the group with residua of congenital dysplasia and 6.7 years (range, one to ten years) for the group with idiopathic osteoarthrosis.
Operative Technique
An anterolateral approach was used, with intraoperative radiographic control and blade-plate fixation. The details of the operative technique have been described previously11. An anterior arthrotomy was possible with this method but was used in only three patients, who had residual deformity resulting from congenital dysplasia; it was used because a labral tear was suspected. The position in which the distal part of the femur was placed in the sagittal plane varied. In the group with residua of congenital dysplasia, one pure varus osteotomy was done, ten varus osteotomies were done in combination with extension of the distal portion of the femur, and seven valgus osteotomies were done in combination with extension of the distal portion of the femur. In the group with idiopathic osteoarthrosis, one pure valgus osteotomy was done, eight varus osteotomies were done in combination with extension of the distal part of the femur, and seven valgus osteotomies were done in combination with extension of the distal portion of the femur. Early protected weight-bearing was allowed postoperatively.
Radiographic Review
We analyzed the anteroposterior radiographs of the pelvis that had been made preoperatively and at six months, one to two years, and at least five years postoperatively. We attempted to control for the variability inherent in a retrospective radiographic review. One observer, who was blinded to the clinical result, reviewed all of the radiographs. The same blinded examiner made repeat measurements in a different sequence at a second sitting.
For both groups of patients, the degree of osteoarthrotic changes was assessed preoperatively according to the criteria of Kellgren and Lawrence. In the group with idiopathic osteoarthrosis, no hip was rated as grade 0 (a normal joint space); the changes were rated as grade 1 (narrowing of the joint space) in two hips, as grade 2 (sclerosis and slight formation of osteophytes) in eight hips, as grade 3 (formation of secondary cysts and substantial formation of osteophytes) in four hips, and as grade 4 (virtually no joint space) in two hips. In the group with residua of congenital dysplasia, no hip was rated as grade 0; the changes were rated as grade 1 in three hips, as grade 2 in four hips, as grade 3 in ten hips, and as grade 4 in one hip. Thus, seven hips that had residual deformity from congenital dysplasia had grade-1 or 2 changes, compared with ten hips with idiopathic osteoarthrosis. Eleven hips in the group with residua of congenital dysplasia had grade-3 or 4 changes, compared with only six hips with idiopathic osteoarthrosis.
For the group with residua of congenital dysplasia, we also evaluated the degree of preoperative acetabular dysplasia by measuring the distance from the teardrop to the center of the femoral head3, the acetabular depth index2, the center-edge angle of Wiberg, the acetabular angle of Sharp, the width of the weight-bearing surface1, and the status of the Shenton line. We considered twelve hips to have grade-I dysplasia, characterized by a center-edge angle of greater than 15 degrees, a Sharp angle of less than 50 degrees, an intact or only slightly interrupted Shenton line, and a normal or only slightly increased distance from the teardrop to the center of the femoral head. We considered six hips to have grade-II dysplasia, characterized by a center-edge angle of 15 degrees or less, a Sharp angle of 50 degrees or more, an interrupted Shenton line, and an increased distance from the teardrop to the center of the femoral head.
Group with Residual Deformity
Resulting from Congenital Dysplasia
Clinical Outcome
All of the eighteen hips (seventeen patients) in the group with residua of congenital dysplasia had caused pain preoperatively. By one year after the osteotomy, the pain had decreased in all of the hips and ten hips were pain-free. Only nine hips (nine patients) remained either pain-free (four hips) or mildly painful (five hips) at least five years postoperatively. The remaining nine hips (eight patients) caused moderate pain (five hips) or severe pain (four hips) (Table I). A total hip arthroplasty was done on the four hips (four patients) that caused severe pain, including one hip in the patient who had had a bilateral osteotomy.
Preoperatively, all of the patients had a mild or severe decrease in the ability to walk and the capability to perform their normal work. At least five years postoperatively, nine patients (nine hips) had no or only a mild impairment of walking ability and of the capability to perform their normal work (Table I). All of the patients had a limp preoperatively, and long-term improvement of the limp with improved abductor mechanics was noted in only three patients (three hips). The patient who had had a bilateral osteotomy showed no improvement in any of these categories.
At least five years postoperatively, eight patients (eight hips) were completely satisfied with the result of the osteotomy. The level of satisfaction decreased with time for the remaining nine patients (ten hips), including one patient who had mild pain but residual stiffness of the hip and the four patients (four hips) who had a subsequent total hip arthroplasty.
Radiographic Findings
Seven hips (seven patients) had early or slight (grade-1 or 2) osteoarthrotic changes on the preoperative radiographs. The superior joint space was at least two millimeters thick on all but one of the preoperative radiographs. At one year postoperatively, the joint space was increased in thirteen hips (thirteen patients). However, this improvement was maintained in only five hips (five patients) at the time of the latest follow-up. Only one hip (one patient) was seen to have fewer periacetabular cysts on the latest radiograph. The Shenton line was restored to normal in seven hips (seven patients) for which a varus osteotomy had been done, and the center-edge angle was improved in three hips (three patients) for which a varus osteotomy had been done. No long-term improvements were noted with regard to the acetabular depth index, the Sharp angle, or the width of the weight-bearing surface.
Relationship between Clinical Outcome and Radiographic Findings
All of the seven hips (seven patients) that had had evidence of grade-1 or 2 osteoarthrotic changes on the preoperative radiographs caused no or only mild pain at least five years postoperatively. None of the seven hips needed a subsequent arthroplasty. In contrast, only two of the eleven hips (ten patients) that had had grade-3 or 4 changes preoperatively remained clinically improved at the time of the latest follow-up. A subsequent arthroplasty was done on four of the eleven hips, including one of the hips in the patient who had had a bilateral osteotomy.
Of the twelve patients (twelve hips) who had grade-I dysplasia preoperatively, seven remained satisfied with the result of the osteotomy at the time of the most recent follow-up and only one had a subsequent arthroplasty. Of the six patients (six hips) who had grade-II dysplasia, only one remained satisfied with the result of the osteotomy at the time of the latest follow-up and three patients (three hips) were subsequently managed with an arthroplasty. (The patient who had had a bilateral osteotomy had an arthroplasty on one hip.)
Group with Idiopathic Osteoarthrosis
Clinical Outcome
All of the sixteen hips in the group with idiopathic osteoarthrosis caused pain preoperatively. By one year postoperatively, the pain had decreased in all of the hips and eleven were pain-free. Nine hips caused no or mild pain, and seven hips caused moderate or severe pain at the time of the latest follow-up (Table I). A total hip arthroplasty was done on all seven of these latter hips.
Preoperatively, all of the patients had a decrease, ranging from mild to severe, in the ability to walk and in the capability to perform their normal work. At least five years postoperatively, eight patients had improvement in both of these categories (Table I). All of the patients had a limp preoperatively, and long-term improvement of the limp with improved abductor mechanics was noted in nine patients.
More than five years postoperatively, six patients reported that they were completely satisfied with the result of the osteotomy. For ten patients, including the seven patients who chose to have a subsequent total hip arthroplasty, the level of satisfaction decreased with time.
Radiographic Findings
Ten of the sixteen hips had evidence of early or slight (grade-1 or 2) osteoarthrotic changes on the preoperative radiographs. Two hips had virtually no joint space, and fourteen had at least two millimeters of superior joint space. At one year postoperatively, the joint space was increased in eleven hips. However, this improvement was maintained in only two hips at the latest follow-up examination. None of the hips had improvement with regard to the number of periacetabular cysts or the sphericity of the femoral head at the time of the most recent follow-up.
Relationship between Clinical Outcome and Radiographic Findings
Of the ten hips that had had evidence of grade-1 or 2 osteoarthrotic changes on the preoperative radiographs, six remained clinically improved (with the patient satisfied with the result) at the time of the latest follow-up, two had some clinical improvement but the patient was not satisfied, and two were not clinically improved and an arthroplasty was done. In contrast, none of the six hips that had had grade-3 or 4 changes preoperatively were clinically improved at the most recent follow-up examination; an arthroplasty was done on five of these six hips.
The purposes of this retrospective study were to assess the results of an isolated femoral intertrochanteric osteotomy in young adults who had symptomatic hip disease and to evaluate the outcome on the basis of the preoperative diagnosis of either residual deformity resulting from congenital dysplasia or idiopathic osteoarthrosis. We believed that five to ten years of follow-up would allow sufficiently important trends to be observed so that we could either confirm or change our indications for the use of this procedure for these two types of hip disease. We also sought to identify additional key factors in the radiographic appearance of the hip that could aid in the selection of patients.
The number of patients in this study was too small for us to reach strong conclusions; only trends could be identified. Although a larger cohort could have been analyzed had we pooled the two groups, we believe that it is important to study these different types of disease separately because the demographic and functional presentations of the groups were so different. The patients who had residua of congenital dysplasia were predominantly women, and they were a mean of fifteen years younger, compared with the patients who had idiopathic osteoarthrosis, when they were first seen. Almost one-third of the patients who had residua of congenital dysplasia had involvement of both hips, and they also tended to have more advanced (grade-3 or 4) osteoarthrotic changes preoperatively, indicating that they were seen later in the course of the disease than were the patients who had idiopathic osteoarthrosis. Gait disturbance, limb-length discrepancy, and a limited preoperative range of motion were also more common in the group with residua of congenital dysplasia.
In both groups, the results were more favorable when an isolated femoral osteotomy had been done at an earlier stage of osteoarthrotic involvement of the hip. Of the seventeen patients who had early osteoarthrotic changes in the hip, most (twelve) believed that they had obtained long-term benefit with marked relief of pain after at least five years of follow-up. These results agree with those of previous studies, particularly with regard to the group with residua of congenital dysplasia. Scher and Jakim reported on combined femoral and Chiari osteotomies for sequelae of congenital dysplasia of the hip and concluded that the best results occurred in patients who had mild osteoarthrotic changes and less severe dysplasia. Pellicci et al. also concluded that better results occurred when a femoral osteotomy had been done in patients with congenital dysplasia who had early osteoarthrotic changes. Trousdale et al. recently reported the results of the Bernese periacetabular osteotomy for forty-two patients who had congenital dysplasia and osteoarthrosis. At an average of four years, they found that patients who had had more advanced osteoarthrotic changes fared more poorly and were more likely to have had a total hip arthroplasty.
For patients who have residua of congenital dysplasia, it appears that the degree of acetabular dysplasia must be considered when assessing the chances of enduring clinical improvement after an isolated proximal femoral osteotomy. The clinical outcome at the time of the most recent follow-up was less satisfying for patients who had had a greater degree of dysplasia on the preoperative radiographs. Radiographic signs of subluxation, including an interruption of the Shenton line or an increased distance from the teardrop to the center of the femoral head, were associated with a poor outcome in five of the six hips that had had grade-II acetabular dysplasia. We currently believe that an isolated femoral intertrochanteric osteotomy should be done only in patients who have, at most, early osteoarthrotic changes and slight dysplasia; specifically, we believe that the osteoarthrotic changes should be no worse than grade 1 or 2 and, for patients who have congenital dysplasia, the center-edge angle should be greater than 15 degrees or the Sharp angle should be less than 50 degrees. Hips with more severe dysplasia (those with a center-edge angle of 15 degrees or less or a Sharp angle of 50 degrees or more, or both) and any degree of subluxation should not be treated with a femoral osteotomy alone. Other studies have supported the addition of a pelvic osteotomy to correct the acetabular deformity of a moderately dysplastic hip6,15.
An isolated femoral osteotomy that is done on a hip with more advanced osteoarthrotic changes or a greater degree of dysplasia is likely to achieve only short-term benefits. Our evaluation of the results at five to ten years postoperatively revealed deterioration of the outcome in patients who had had more advanced osteoarthrotic changes preoperatively. Widespread early benefit was noted: at one year, the hip was less painful after twenty-two of the thirty-four osteotomies, the ability to walk was improved after twenty-eight of the osteotomies, and the capability to perform work was improved after thirty-two. These early encouraging results deteriorated with time, particularly in patients who had had more advanced osteoarthrotic changes preoperatively. Such patients must be made aware of the limited benefit that can be expected from a femoral osteotomy and that they may be better served by a total hip arthroplasty.
Even to patients who are good candidates for an isolated intertrochanteric osteotomy, we would emphasize that our results at five to ten years must be considered preliminary. In our view, a successful alternative to total hip arthroplasty must predictably achieve improved clinical function for at least ten years to justify its continued use.
Bombelli, R., and Aronson, J.: Biomechanical classification of osteoarthritis of the hip. In The Intertrochanteric Osteotomy, pp. 69-134. Edited by J. Schatzker. New York, Springer, 1984.
Cooperman, D. R.; Wallensten, R.; and |and |Stulberg, S. D.: Acetabular dysplasia in the adult. Clin. Orthop.,175: 79-85, 1983.17579
1983
[PubMed]
Harris, W. H.: Etiology of osteoarthritis of the hip. Clin. Orthop.,213: 20-33, 1986.21320
1986
[PubMed]
Johnston, R. C.; Fitzgerald, R. H., Jr.; Harris, W. H.; Poss, R.; Müller, M. E.; and |and |Sledge, C. B.: Clinical and radiographic evaluation of total hip replacement. A standard system of terminology for reporting results. J. Bone and Joint Surg.,72-A: 161-168, Feb. 1990.72-A161
1990
Kellgren, J. H., and |and |Lawrence, J. S.: Radiological assessment of osteoarthrosis. Ann. Rheumat. Dis.,16: 494-502, 1957.16494
1957
[PubMed][CrossRef]
Millis, M. B.; Murphy, S. B.; and |and |Poss, R.: Instructional course lecture. Osteotomies about the hip for the prevention and treatment of osteoarthrosis. J. Bone and Joint Surg.,77-A: 626-647, April 1995.77-A626
1995
Morscher, E., and Feinsein, R: Results of intertrochanteric osteotomy in the treatment of osteoarthritis of the hip. In The Intertrochanteric Osteotomy, pp. 169-177. Edited by J. Schatzker. New York, Springer, 1984.
Müller, M. E.: Intertrochanteric osteotomy: indication, preoperative planning, technique. In The Intertrochanteric Osteotomy, pp. 25-66. Edited by J. Schatzker. New York, Springer, 1984.
Pellicci, P. M.; Hu, S.; Garvin, K. L.; Salvati, E. A.; and |and |Wilson, P. D., Jr.: Varus rotational femoral osteotomies in adults with hip dysplasia. Clin. Orthop.,272: 162-166, 1991.272162
1991
[PubMed]
Poss, R.: Current concepts review. The role of osteotomy in the treatment of osteoarthritis of the hip. J. Bone and Joint Surg.,66-A: 144-151, Jan. 1984.66-A144
1984
Poss, R.: Intertrochanteric osteotomy in osteoarthritis of the hip. In Instructional Course Lectures, The American Academy of Orthopaedic Surgeons. Vol. 35, pp. 129-143. St. Louis, C. V. Mosby, 1986.
Reigstad, A., and |and |Grønmark, T.: Osteoarthritis of the hip treated by intertrochanteric osteotomy. A long-term follow-up. J. Bone and Joint Surg.,66-A: 1-6, Jan. 1984.66-A1
1984
Scher, M. A., and |and |Jakim, I.: Combined intertrochanteric and Chiari pelvic osteotomies for hip dysplasia. J. Bone and Joint Surg.,73-B(4): 626-631, 1991.73-B(4)626
1991
Sharp, I. K.: Acetabular dysplasia. The acetabular angle. J. Bone and Joint Surg.,43-B(2): 268-272, 1961.43-B(2)268
1961
Trousdale, R. T.; Ekkernkamp, A.; Ganz, R.; and |and |Wallrichs, S. L.: Periacetabular and intertrochanteric osteotomy for the treatment of osteoarthrosis in dysplastic hips. J. Bone and Joint Surg.,77-A: 73-85, Jan. 1995.77-A73
1995
Weinstein, S. L.: Natural history of congenital hip dislocation (CDH) and hip dysplasia. Clin. Orthop.,225: 62-76, 1987.22562
1987
[PubMed]
Wiberg, G.: Studies on dysplastic acetabula and congenital subluxation of the hip joint. With special reference to the complication of osteo-arthritis. Acta Chir. Scandinavica,Supplementum 58: 1939.Supplementum 58
1939