Abstract
The results were evaluated for twenty-eight adolescents and adults (thirty-one hips) who had had a Salter innominate osteotomy because of acetabular dysplasia and pain in the hip. The mean age at the time of the index operation was twenty-two years, and the mean duration of radiographic follow-up was seventy-one months.Radiographs were available for twenty-five patients (twenty-eight hips) at the most recent follow-up evaluation. The radiographic evaluation included determination of the acetabular angle, the center-edge angle, the coverage of the femoral head, the height of the joint space, and the Shenton line. Compared with the preoperative measurements, the acetabular angle had decreased by a mean of 10.0 degrees, the center-edge angle had increased by a mean of 13.2 degrees, and the coverage of the femoral head had increased by a mean of 15 per cent (p < 0.001 for all three values). The height of the joint space had decreased but, with the numbers available, this change was not significant.The Harris hip scores, determined for twenty patients (twenty-one hips), improved from a mean of 71.2 points preoperatively to a mean of 88.3 points at the latest follow-up evaluation (p < 0.001). A questionnaire, completed by twenty-seven patients (thirty hips), revealed that twenty-one patients (78 per cent) were satisfied with the result of the operation; twenty-two patients (81 per cent) said that they would recommend this procedure to others who had a similar condition. The pain score (with 1 point indicating mild pain and 5 points indicating severe pain) improved from a mean of 3.7 points preoperatively to a mean of 2.5 points at the latest follow-up evaluation (p < 0.001).There were six postoperative complications. One patient had a non-union; one, an infection; one, heterotopic bone; and three, numbness. Two of these patients subsequently had a total hip arthroplasty at ninety-eight and 150 months, and one patient had a Chiari osteotomy at twenty-two months.Our findings demonstrate that the Salter innominate osteotomy provides notable clinical improvement as well as improvements in the radiographic measurements of the hip in patients who have acetabular dysplasia.
The relationship between acetabular dysplasia and degenerative osteoarthrosis of the hip has been studied extensively yet it is not completely understood. Various authors have reported a 1 to 48 per cent prevalence of dysplasia of the hip in patients who have degenerative osteoarthrosis1,7,11,14,21. However, the extent to which dysplasia results in premature degenerative osteoarthrosis of the hip is difficult to determine.
Wiberg studied the radiographs of all patients with hip disease who had been admitted to the Vanföreanstalten between 1913 and 1925 and to the Serafimer Hospital between 1908 and 1922. Forty-four patients had maldevelopment of the acetabulum and congenital subluxation of the hip without osteoarthrosis. Wiberg was able to re-examine seventeen patients from this group. In fourteen of them, osteoarthrosis was first detected before the end of the fifth decade of life.
Cooperman et al., in an effort to determine the relationship between dysplasia and degenerative osteoarthrosis of the hip, studied the radiographs of all patients with radiographic evidence of hip disease who had been managed at the Norrbacka Institute between 1935 and 1960. They found that thirty-two hips in twenty patients met several criteria: a center-edge angle of 20 degrees or less, an intact Shenton line, no history of a serious disorder of the hip in childhood or of systemic disease that might have affected the hip, and no radiographic evidence of osteoarthrosis when the patient was first examined. Four (13 per cent) of these thirty-two hips had evidence of severe osteoarthrosis (at least a 75 per cent loss of the normal joint space) by the end of the fifth decade of life.
It has been demonstrated that subluxation of a dysplastic hip leads to an increased unit load13 that may be too great for the bone and cartilage of the hip joint to adapt to, resulting in degenerative osteoarthrosis. It therefore seems reasonable that restoration of the anatomy of the hip joint to a normal configuration decreases the unit load and helps to prevent osteoarthrosis. The use of pelvic osteotomies for this purpose in the treatment of dysplasia of the hip in adolescents and adults has been reported. Various procedures have been described, including innominate osteotomy10,18,26, double innominate osteotomy23, triple innominate osteotomy5,20, Chiari osteotomy2,9,15,16, shelf osteotomy12,22,25, and several different types of periacetabular osteotomies4,6,24,29.
Salter was the first investigator to apply the principle of innominate osteotomy to the treatment of dysplastic hips in children, to our knowledge17. He and colleagues18,26 subsequently demonstrated that, with the proper indications, an innominate osteotomy for a dysplastic hip can also be performed in adolescents and adults, with few complications and good relief of pain. The indications for the procedure in this older age-group included a joint-space height that was more than 50 per cent of normal, a range of motion of the hip joint that was at least 60 per cent of normal, reasonable radiographic congruity of the hip joint, and subluxation from the true acetabulum18.
The purpose of the current study was to document our results with the Salter innominate osteotomy for the treatment of acetabular dysplasia in adolescents and adults.
*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. Funds were received in total or partial support of the research or clinical study presented in this article. The funding source was the Department of Orthopaedics, Cleveland Clinic Foundation.
†Department of Orthopaedics, Cleveland Clinic Foundation, 9500 Euclid Avenue, Suite A51, Cleveland, Ohio 44195.
Thirty-seven hips in thirty-three patients had a Salter innominate osteotomy for acetabular dysplasia, performed by the senior one of us (A. R. G.) between 1979 and 1991. Three patients (three hips) could not be contacted for follow-up or requested not to be included in the study. One patient (one hip) died of unrelated causes nine years after the operation. One patient (one hip) who had the procedure did not meet the criteria of Salter et al. as described previously18. One patient, who had had staged innominate osteotomies for bilateral dysplasia, had a Chiari osteotomy twenty-two months after the second operation because of continued pain. The hip that had this operation was not included in the study because of inadequate follow-up; the result was considered a failure. Thirty-one hips in twenty-eight patients were included in the study.
The mean age of the twenty-eight patients at the time of the index operation was twenty-two years (range, twelve to forty-six years). Twenty-three patients were female and five were male. The operation was done on twenty right and eleven left hips.
The mean age of the four patients who had a bilateral procedure was fifteen and one-half years (range, twelve to eighteen years) at the time of the first operation. All four patients were female, and the mean interval between the operations was nine months (range, six to twelve months).
The operative indications included acetabular dysplasia (an acetabular angle of at least 40 degrees and a center-edge angle of no more than 20 degrees); a painful, reasonably congruous hip joint; and an age of less than sixty years. Contraindications to the operation, as described by Salter et al.18, included subluxation of the femoral head from a so-called false acetabulum, incongruity of the hip joint, a joint-space height that was less than one-half of normal for an individual of that age, and a range of motion that was less than 60 per cent of normal.
The procedure was performed as described by Salter et al.18,26. The adductor and iliopsoas muscles are released in all patients. The innominate bone is osteotomized with a Gigli saw from the sciatic notch to the anterior inferior iliac spine. The osteotomy site is hinged and held closed posteriorly, and it is opened anteriorly by external rotation of the femur. A triangular block of iliac bone is inserted into the site of the osteotomy and is held with three threaded Steinmann pins. The hip is moved through a range of motion to ensure that the pins do not traverse the hip joint. The amount of acetabular coverage obtained as well as the position of the pins are confirmed with fluoroscopy.
Postoperatively, assisted and active range-of-motion exercises for the hip are begun as soon as tolerated. Weight-bearing is limited to 30 per cent of full weight-bearing for three months and is increased as radiographic evidence of union at the site of the osteotomy is seen. The Steinmann pins are removed when there is radiographic evidence of union at the site of the graft.
Eight patients had an additional procedure on the hip, performed concomitantly with the Salter procedure. One patient had a core decompression (for avascular necrosis of the femoral head), one had advancement of the greater trochanter, one had a Wagner sliding subtrochanteric osteotomy (for limb-lengthening), two had a proximal femoral varus osteotomy, and three had a shelf procedure.
Thirteen patients had had a total of fifteen previous procedures on the involved hip. Seven had had a closed reduction and application of a cast; two, an open reduction and application of a cast; one, a shelf procedure; two, a proximal femoral osteotomy; one, an open reduction and internal fixation for a pelvic fracture; one, an earlier innominate osteotomy (when the patient was two years old); and one, bone-grafting for a cyst in the proximal part of the femur.
Radiographs of the pelvis, made preoperatively, early postoperatively (at a mean of 3.6 months), and at the most recent follow-up visit (at least two years after the operation), were reviewed. Anteroposterior radiographs were used to determine the acetabular angle19, the center-edge angle27, and the coverage of the femoral head (Fig. 1). The height of the joint space was measured in millimeters, at two locations. The first measurement—the weight-bearing joint space—was performed at the superior aspect of the dome of the acetabulum, as determined by a vertical line through the center of the femoral head, or at the lateralmost margin of the acetabular dome if the center-edge angle was negative. The second measurement—the minimum joint space—was performed at the point of the greatest narrowing of the joint space perpendicular to the articular surface (Fig. 2). The Shenton line was also evaluated for evidence of subluxation.
In addition to the radiographic examination, an interim history was obtained and a physical examination of the hip was done at the most recent follow-up visit. The findings of the clinical evaluation were quantified with use of the Cleveland Clinic hip score, a modified version of the Harris hip score. The Cleveland Clinic hip score is very similar to the Harris hip score, with function of the hip rated on a 100-point scale. Points are assigned for pain, function, absence of deformity, and range of motion, with 91 to 100 points considered an excellent result; 81 to 90 points, a good result; 71 to 80 points, a fair result; and 70 points or less, a poor result. However, there are several differences between the two rating systems. The Cleveland Clinic system assigns 4 points for use of one crutch, whereas the Harris system assigns only 3 points. The Cleveland Clinic system assigns 4 points for sitting in any chair for one hour and 2 points for sitting in a high chair for one-half hour, whereas the Harris system assigns 5 and 3 points, respectively. Range of motion is calculated with use of a table with the Cleveland Clinic system, whereas a formula is used with the original Harris system. If a hip score had not been obtained preoperatively, the range of motion was determined from the patient's medical record and preoperative estimates of pain and function were obtained from the patient at the time of follow-up.
The patients were asked to complete a questionnaire inquiring if they were satisfied with the result of the procedure, if they would recommend it to another person who had a similar condition, and what medications they were currently taking. The patient assigned pain in the hip a score of 1 to 5 points, with 1 point indicating no pain and 5 points, severe pain. If a patient could not return for a follow-up visit, the history and the responses to the questionnaire were obtained in a telephone interview, and an attempt was made to obtain pelvic radiographs from the patient's local physician.
Intraobserver variation was evaluated by comparing the measurements of the acetabular and center-edge angles made on ten preoperative radiographs by one of us (J. J. McC.) with those made on the same radiographs by the same one of us two weeks later. Interobserver variation was evaluated by comparing measurements of the acetabular and center-edge angles made on ten preoperative radiographs by one of us (J. J. McC.) with those made on the same radiographs by other orthopaedic surgeons.
The data were analyzed by the Department of Biostatistics and Epidemiology at our institution. Repeated-measures analysis of variance was used to determine if there had been a significant change in the radiographic measurements of the hip joint over time (p < 0.05). For measurements that changed significantly over time, pairwise comparisons were done (the early postoperative measurements were compared with the preoperative measurements; the measurements made at the latest follow-up evaluation, with the early postoperative measurements; and the measurements made at the latest follow-up evaluation, with the preoperative measurements), with use of either the paired t test or the Wilcoxon signed-rank test28 as appropriate. The Bonferroni adjustment for multiple comparisons was used (p < 0.017). All tests were two-tailed.
Radiographic Studies
Radiographs were available for twenty-five patients (89 per cent; twenty-eight hips) at a mean of seventy-one months (range, twenty-four to 163 months) postoperatively. The radiographs showed that the acetabular angle had decreased by a mean of 10.0 degrees (range, 2 to 20 degrees), the center-edge angle had increased by a mean of 13.2 degrees (range, -26.0 to 42.0 degrees), and the coverage of the femoral head had increased by a mean of 15 per cent (range, -25 to 43 per cent) as compared with the preoperative measurements (p < 0.001 for all three values) (Table I).
The weight-bearing and minimum joint spaces had decreased by a mean of 0.9 millimeter (range, -7.0 to 2.0 millimeters; p = 0.041) and a mean of 0.2 millimeter (range, -9.0 to 4.0 millimeters; p = 0.67), respectively, between the preoperative and early postoperative measurements. These measurements continued to decrease postoperatively, by a mean of 1.2 millimeters (range, -3.0 to 2.0 millimeters; p = 0.31) and 0.7 millimeter (range, -5.0 to 1.0 millimeter; p = 0.051) for the weight-bearing and minimum joint spaces, respectively, at the most recent follow-up evaluation.
The Shenton line was assessed preoperatively in twenty-two patients (79 per cent; twenty-four hips) and was found to be intact in six hips (25 per cent). At the latest follow-up evaluation, the Shenton line was assessed in twenty-five patients (twenty-eight hips) and was intact in eleven hips (39 per cent).
Hip Scores
The Harris hip score was determined for twenty patients (71 per cent; twenty-one hips) at a mean of eighty-seven months (range, twenty-eight to 163 months) postoperatively. The mean hip score was 71.2 points (range, 34 to 99 points) preoperatively and 88.3 points (range, 58 to 100 points) at the latest follow-up evaluation. The mean difference between these scores was 17.1 points (range, -6 to 56 points; p < 0.001). Of the twenty-one hips, thirteen (62 per cent) had an excellent result; three (14 per cent), a good result; four (19 per cent), a fair result; and one (5 per cent), a poor result. An additional patient had had a failure of the operation at the latest follow-up examination.
Questionnaire Responses
The questionnaire was completed by twenty-seven patients (96 per cent; thirty hips) at a mean of eighty-three months (range, twenty-eight to 164 months) postoperatively. Twenty-one patients (78 per cent) reported that they were satisfied with the result of the operation, and twenty-two (81 per cent) said that they would recommend this procedure to others who had a similar condition. Four patients (15 per cent) were taking medication for pain in the hip at the time of the latest follow-up evaluation. The mean pain score decreased from 3.7 points (range, 1 to 5 points) preoperatively to 2.5 points (range, 1 to 5 points) early postoperatively (p < 0.001); this score remained the same at the latest follow-up evaluation.
Complications
Complications occurred postoperatively in six patients (six hips), three of whom needed an additional procedure. One patient had a non-union, which necessitated internal fixation; one, an infection, which was treated with irrigation and débridement; and one, formation of heterotopic bone, which was treated with débridement. Three patients reported numbness: two, in the ipsilateral thigh and one, in the medial aspect of the ipsilateral knee. The numbness decreased with time and did not interfere with the patients' daily activities.
Two patients had a total hip arthroplasty, at ninety-eight and 150 months after the innominate osteotomy, because of continued pain and radiographic evidence of degenerative osteoarthrosis. As mentioned, a third patient, who had had bilateral dysplasia of the hip and had been managed with staged innominate osteotomies, had a Chiari osteotomy twenty-two months after the second operation because of continued pain. This hip, which was not included in the study because of inadequate follow-up, was considered a failure. A fourth patient had been diagnosed with avascular necrosis of the ipsilateral femoral head preoperatively, and a core decompression had been done at the time of the index operation. Nine years after the original procedure, exploration and débridement was performed for removal of loose bodies in the hip joint. Seven patients reported pain, which was believed to be related to the presence of the Steinmann pins and which decreased after the pins were removed.
Intraobserver and Interobserver Variation in the Radiographic Measurements
The correlation for intraobserver reliability was 0.95 for both the acetabular angle and the center-edge angle. The correlation for interobserver reliability was 0.99 for the acetabular angle and 0.85 for the center-edge angle.
Several types of osteotomies have been described for the treatment of acetabular dysplasia and subluxation in adolescents and young adults. All of these procedures are based on the same hypothesis: that the increased forces acting on the weight-bearing surface of the subluxated femoral head lead to increased stresses and early degenerative changes. An early anatomical restoration that provides a greater weight-bearing surface to the femoral head may delay or prevent this process.
Sutherland and Greenfield reported the results of a double innominate osteotomy in twenty-five patients, primarily adolescents, of whom twelve had had a diagnosis of congenital dislocation of the hip. Postoperatively, the acetabular angle decreased by a mean of 19.5 degrees and the center-edge angle increased by a mean of 27 degrees. All but two of the twenty-five hips were considered to be stable at a mean of twenty months postoperatively.
Steel described the use of a triple innominate osteotomy in forty-five patients (fifty-two hips) whose ages ranged from seven to seventeen years; twenty-three patients had a diagnosis of congenital dislocation of the hip. A satisfactory result was defined as a "painless and stable and reasonably mobile" hip. According to these criteria, forty of the fifty-two hips and nineteen of the twenty-three hips that had congenital dysplasia were considered to have a satisfactory result at a mean of two years or more after the operation.
More recently, Faciszewski et al. reported the results for fifty-six hips in forty-four patients who had had a triple innominate osteotomy for the treatment of acetabular dysplasia. The mean age of the patients was twenty-eight years. Fifty-three hips (95 per cent) had decreased pain and improved function, and forty-two patients (95 per cent) said that they would recommend the procedure. The center-edge angle had increased by a mean of 33 degrees and the acetabular angle had decreased by a mean of 15 degrees at a mean of seven years postoperatively.
The potentially beneficial results of an innominate osteotomy in young adults were first reported by Salter et al.18. Husby et al. also reported improvement in the radiographic findings and satisfactory relief of pain after an innominate osteotomy in young adults.
Our findings are similar to those of both Salter et al.18 and Husby et al. with respect to improved function, decreased pain, and improvement according to the radiographic criteria (Table II). However, we did not see a resolution of the preoperative osteoarthrotic changes that Salter et al. described. In fact, we found that the joint space narrowed slightly between the early postoperative and latest follow-up evaluations, although, with the numbers available, this change was not significant. The weight-bearing joint space decreased significantly (p = 0.041) between the times of the preoperative and early postoperative radiographs. This decrease may be due in part to the displacement of the acetabulum caused by the operation rather than to osteoarthrotic changes. The relatively short period of time between the operation and when the early postoperative radiographs were made (a mean of 3.6 months) and the decreased level of activity during this period support this hypothesis. The minimum joint space did not decrease significantly during this same time-period, and there was no change in the height of the joint space in the contralateral hip.
Salter et al.18 reported that the Shenton line was intact postoperatively in more than 85 per cent of their patients, whereas this was true in only eleven (39 per cent) of our patients.
Our study is unique, to our knowledge, in that we reported on both the coverage of the femoral head and the hip scores, both of which improved considerably.
Adam, A., and |and |Spence, A. J.: Intertrochanteric osteotomy for osteoarthritis of the hip. J. Bone and Joint Surg.,40-B(2): 219-226, 1958.40-B(2)219
1958
Colton, C. L.: Chiari osteotomy for acetabular dysplasia in young subjects. J. Bone and Joint Surg.,54-B(4): 578-589, 1972.54-B(4)578
1972
Cooperman, D. R.; Wallensten, R.; and |and |Stulberg, S. D.: Acetabular dysplasia in the adult. Clin. Orthop.,175: 79-85, 1983.17579
1983
[PubMed]
Eppright, R. H.: Dial osteotomy of the acetabulum in the treatment of dysplasia of the hip. In Proceedings of The American Academy of Orthopaedic Surgeons. J. Bone and Joint Surg.,58-A: 726, July 1976.58-A726
1976
Faciszewski, T.; Coleman, S. S.; and |and |Biddulph, G.: Triple innominate osteotomy for acetabular dysplasia. J. Pediat. Orthop.,13: 426-430, 1993.13426
1993
[CrossRef]
Ganz, R.; Klaue, K.; Vinh, T. S.; and |and |Mast, J. W.: A new periacetabular osteotomy for the treatment of hip dysplasias. Technique and preliminary results. Clin. Orthop.,232: 26-36, 1988.23226
1988
[PubMed]
Gudmundsson, G.: Intertrochanteric displacement osteotomy for painful osteoarthritis of the hip. Acta Orthop. Scandinavica,41: 91-109, 1970.4191
1970
[CrossRef]
Harris, W. H.: 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 and Joint Surg.,51-A: 737-755, June 1969.51-A737
1969
Høgh, J., and |and |Macnicol, M. F.: The Chiari pelvic osteotomy. A long-term review of clinical and radiographic results. J. Bone and Joint Surg.,69-B(3): 365-373, 1987.69-B(3)365
1987
Husby, O. S.; Lunde, O. D.; and |and |Sudmann, E.: Salters bekkenosteotomi. Et behandlingstilbud for unge pasienter med begynnende degenerativg hoftelidelse. Tidsskr. Norske Laegeforen,112: 881-883, 1992.112881
1992
Lloyd-Roberts, G. C.: Osteoarthritis of the hip. J. Bone and Joint Surg.,37-B(1): 8-47, 1955.37-B(1)8
1955
Love, B. R.; Stevens, P. M.; and |and |Williams, P. F.: A long-term review of shelf arthroplasty. J. Bone and Joint Surg.,62-B(3): 321-325, 1980.62-B(3)321
1980
Pauwels, F.: Biomechanics of the Normal and Diseased Hip. Theoretical Foundation, Technique and Results of Treatment. New York, Springer, 1976.
Ranawat, C. S.; Atkinson, R. E.; Salvati, E. A.; and |and |Wilson, P. D., Jr.: Conventional total hip arthroplasty for degenerative joint disease in patients between the ages of forty and sixty years. J. Bone and Joint Surg.,66-A: 745-752, June 1984.66-A745
1984
Reynolds, D. A.: Chiari innominate osteotomy in adults. Technique, indications and contra-indications. J. Bone and Joint Surg.,68-B(1): 45-54, 1986.68-B(1)45
1986
Rush, J.: Chiari osteotomy in the adult: a long-term follow-up study. Australian and New Zealand J. Surg.,61: 761-764, 1991.61761
1991
Salter, R. B., and Thompson, G. H.: The role of innominate osteotomy in young adults. In The Hip. Proceedings of the Seventh Open Scientific Meeting of The Hip Society, pp. 278-312. St. Louis, C. V. Mosby, 1979.
Salter, R. B.; Hansson, G.; and |and |Thompson, G. H.: Innominate osteotomy in the management of residual congenital subluxation of the hip in young adults. Clin. Orthop.,182: 53-68, 1984.18253
1984
[PubMed]
Sharp, I. K.: Acetabular dysplasia. The acetabular angle. J. Bone and Joint Surg.,43-B(2): 268-272, 1961.43-B(2)268
1961
Steel, H. H.: Triple osteotomy of the innominate bone. J. Bone and Joint Surg.,55-A: 343-350, March 1973.55-A343
1973
Stulberg, S. D., and Harris, W. H.: Acetabular dysplasia and development of osteoarthritis of the hip. In The Hip. Proceedings of the Second Open Meeting of The Hip Society, pp. 82-93. St. Louis, C. V. Mosby, 1974.
Summers, B. N.; Turner, A.; and |and |Wynn-Jones, C. H.: The shelf operation in the management of late presentation of congenital hip dysplasia. J. Bone and Joint Surg.,70-B(1): 63-68, 1988.70-B(1)63
1988
Sutherland, D. H., and |and |Greenfield, R.: Double innominate osteotomy. J. Bone and Joint Surg.,59-A: 1082-1091, Dec. 1977.59-A1082
1977
Wagner, H.: Experiences with spherical acetabular osteotomy for the correction of the dysplastic acetabulum. Prog. Orthop. Surg.,2: 131-145, 1978.2131
1978
Wainwright, D.: The shelf operation for hip dysplasia in adolescence. J. Bone and Joint Surg.,58-B(2): 159-163, 1976.58-B(2)159
1976
Wedge, J. H., and |and |Salter, R. B.: Immominate osteotomy: its role in the arrest of secondary degenerative arthritis of the hip in the adult. Clin. Orthop.,98: 214-224, 1974.98214
1974
[PubMed][CrossRef]
Wiberg, G.: Studies on dysplastic acetabula and congenital subluxation of the hip joint. With special reference to the complication of osteoarthritis. Acta Chir. Scandinavica,Supplementum 58: 1939.Supplementum 58
1939
Wilcoxon, F.: Individual comparisons by ranking methods. Biomet. Bull.,1: 80-83, 1945.180
1945
[CrossRef]
Yano, H.; Sano, S.; Nagata, Y.; Tabuchi, K.; Okinaga, S.; Seki, H.; and |and |Suyama, T.: Modified rotational acetabular osteotomy (RAO) for advanced osteoarthritis of the hip joint in the middle-aged person. First report. Arch. Orthop. and Trauma Surg.,109: 121-125, 1990.109121
1990
[CrossRef]