- Email: [email protected]
Curved varus femoral osteotomy for minimal dysplastic hip in patients older than 45 years of age: comparison with rotational acetabular osteotomy
J Orthop Sci (2005) 10:264–269 DOI 10.1007/s00776-005-0885-9
Original article Curved varus femoral osteotomy for minimal dysplastic hip in patients older than 45 years of age: comparison with rotational acetabular osteotomy Y. Yasunaga, T. Hisatome, R. Tanaka, T. Yamasaki, and M. Ochi Department of Orthopaedic Surgery, Hiroshima University School of Medicine, 1-2-3 Kasumi, Minami-ku, Hiroshima 734-8551, Japan
Abstract Periacetabular osteotomies, such as rotational acetabular osteotomy (RAO) for osteoarthritis (OA) secondary to hip dysplasia, have produced satisfactory medium- and long-term results in early OA. However, as the degree of surgical invasion of these procedures is comparatively major and long-term postoperative rehabilitation is required, this procedure is generally not indicated for those in the fifth and sixth decades of life. Curved varus femoral osteotomy (CVO), which is a less invasive procedure, is considered a worthwhile option, although to the best of our knowledge no comparative study has been reported on these two procedures in older patients. We performed a comparative study on the outcomes of CVO for minimal dysplasia of the hip with a center-edge (CE) angle greater than 5° in 15 patients (15 joints) older than 45 years of age at a mean period of 8 years after surgery and on the outcomes of RAO in 16 patients (18 joints) performed during the same period. No significant difference was observed between the two groups regarding the Merle d’Aubigne score during follow-up. Progression of osteoarthritis was radiologically observed during follow-up in three hips in the CVO group and in one hip in the RAO group. Kaplan-Meier survivorship analysis, with radiological progression of osteoarthritis as the endpoint, predicted a survival rate of 75% at 10 years in the CVO group. In the RAO group the predicted 10-year survival rate was 94%, demonstrating no significant difference between the two groups. CVO is a less invasive surgical technique than RAO, requires a comparatively shorter period of postoperative rehabilitation, and is useful as a joint-preserving procedure in older patients with minimal dysplasia of the hip with unilateral involvement. CVO should be selected for elderly patients with a CE angle greater than 5°.
Introduction Periacetabular osteotomies for osteoarthritis (OA) secondary to hip dysplasia have theoretical advantages Offprint requests to: Y. Yasunaga Received: June 30, 2004 / Accepted: January 26, 2005
over other pelvic procedures, such as Chiari osteotomy,1 because they alter the position of the acetabulum en bloc, allowing the femoral head to be covered by cartilage.2–6 These procedures have produced satisfactory medium- and long-term results in patients with early OA secondary to dysplasia of the hip.2,7–9 However, as the degree of surgical invasion of periacetabular osteotomies is comparatively major and long-term postoperative rehabilitation is required, this procedure is generally not indicated for those in the fifth and sixth decades of life. Thus, curved varus femoral osteotomy (CVO), which is a less invasive procedure, is considered a worthwhile option, although to the best of our knowledge no comparative study has been reported on these two procedures in older patients. The purpose of this report is to compare the outcomes of CVO with those of rotational acetabular osteotomy (RAO) during the fifth and sixth decades of life.
Patients and methods Patients The indications for CVO and RAO are basically equivalent: acetabular dysplasia radiologically characterized by a center-edge (CE) angle10 of less than 20° and progressive pain interfering with daily activities. Patients are considered candidates for surgery if the hip joint demonstrated improved femoral head coverage in abduction and improved joint congruency in abduction on the anteroposterior view of a plain radiograph. Abductor strength greater than 4 (on the muscle manual test) was a precondition for surgery. However, in patients with a CE angle greater than 5° and whose age was more than 45 years, it was explained that although both procedures were indicated CVO is less surgically invasive than RAO so the patients could make an informed choice of which procedure to undergo.
Y. Yasunaga et al.: Curved varus femoral osteotomy
265
Table 1. Patient profiles Variable No. of patients No. of hips Gender (male : female) Average age at surgery (years), and range Average follow-up period, and range Charnley category (A : B)
CVO group
RAO group
15 15 2 : 13 53.3 (45–60) 7 years 8 months (4–13 years) 5 : 10
20 22 2 : 18 50.5 (46–57) 8 years 10 months (4–14 years) 6 : 14
CVO, curved varus femoral osteotomy; RAO, rotational acetabular osteotomy
The study included 15 consecutive patients (15 hips) in the early stage of OA who underwent CVO between July 1991 and June 1999, whose age at surgery was more than 45 years and who were followed up for more than 4 years (Table 1). According to the Charnley classification11 there were 5 unilateral cases (category A) and 10 bilateral cases (category B). Among those in Charnley category B, RAO was performed on the contralateral side in two patients, and in the remaining eight patients the OA on the nonoperated side was in the early stage. No patients had had previous surgery. Between October 1988 and June 1999, we performed RAO on 166 hips with prearthritis and early-stage OA secondary to dysplasia. Among these patients, there were 16 (18 hips) in the early stage of OA with a CE angle greater than 5° whose age at surgery was older than 45 years and who were followed up for more than 4 years (Table 1). According to Charnley’s classification, 6 patients had unilateral involvement (A) and 10 patients bilateral involvement (B). Among patients with bilateral involvement, RAO was performed bilaterally in two. One patient had dislocation at a high position on the nonoperated side. In the remaining seven, OA on the nonoperated side was in the early stage. No patients had had previous surgery. Osteoarthritis of the hip was classified into four radiological stages based on Japanese Orthopaedic Association (JOA) guidelines: (1) prearthritis; (2) earlystage OA; (3) advanced-stage OA; and (4) end-stage OA. Procedure The CVO procedure was performed based on the “transtrochanteric curved varus osteotomy” technique developed by Nishio and Sugioka.12,13 The varus angle was determined based on femoral head coverage and improved joint congruency in abduction in the anteroposterior view of a plain radiograph; it was set at 20°– 30°. We fixed the varus-positioned femoral head with two or three screws. Active range-of-motion exercises began on the seventh postoperative day. Partial weightbearing on two crutches was allowed from the sixth postoperative week, and full weight-bearing was initi-
ated 3 months after surgery. In one patient with a CE angle of 5° who selected varus osteotomy (VO), acetabuloplasty was also performed. The RAO procedure was performed based on the technique developed by Ninomiya and Tagawa.3 We secured the rotated acetabulum with two sapphire screws (Kyocera, Kyoto, Japan). After August 1996, poly-l-lactic screws (Gunze, Kyoto, Japan) were used for fixation instead of sapphire screws. Active range-of-motion exercises began on the seventh postoperative day in RAO patients. Partial weight-bearing on two crutches was allowed from the sixth postoperative week, and full weight bearing was initiated 5–6 months after surgery when Trendelenburg’s sign had disappeared. Clinical follow-up was based on the system of Merle d’Aubigne and Postel.14 With this system, pain, mobility, and walking ability are assessed with scores from 0 to 6. CE angle, acetabular roof obliquity (AC angle),15 and the head lateralization index (HLI)16 using a revised method developed by Ninomiya17 were measured on the anteroposterior view of radiographs obtained preoperatively, 3 months postoperatively, and once yearly thereafter. Changes in the CE and AC angles were used to describe any medial and lateral remodeling of the subchondral bone of the acetabulum caused by joint remodeling.18 Radiographic measurements were performed by two of the authors (K.T. and R.T.), and the data are expressed as group mean values. Postoperative joint congruencies in both procedures were classified into four grades:19 excellent, the curvature of the acetabulum and of the femoral head was almost identical, and joint space was adequately maintained; good, the curvatures of the acetabulum and the femoral head were not identical, but joint space was adequately maintained; fair, partial narrowing of joint space occurred; poor, joint space partially disappeared. Patient satisfaction regarding pain relief at followup was classified into four grades: excellent, very satisfied; good, satisfied; fair, slightly better than preoperative condition; poor, worse than preoperative condition.
266
Y. Yasunaga et al.: Curved varus femoral osteotomy
Table 2. Clinical score Score: mean (SD, range) Preoperative score Pain Walking Mobility Follow-up score Pain Walking Mobility
CVO group
RAO group
13.2 (1.1, 11–15) 3.3 (0.6, 2–4) 4.7 (0.5, 4–5) 5.2 (0.6, 4–6) 14.9* (3.1, 9–18) 4.7* (1.4, 2–6) 4.7# (1.0, 3–6) 5.4 (0.8, 4–6)
13.4 (1.8, 10–16) 3.4 (0.8, 2–4) 4.4 (0.8, 3–6) 5.6 (0.7, 4–6) 16.6* (1.9, 13–18) 5.4* (0.7, 4–6) 5.5** (0.7, 4–6) 5.7 (0.6, 4–6)
* P ⬍ 0.0001, ** P ⬍ 0.05 Wilcoxon rank test # P ⬍ 0.05, Mann-Whitney U-test
Statistical analysis The differences between the two group means were tested using Mann-Whitney’s U-test, Wilcoxon’s rank test, and Fisher’s exact probability test. Cumulative probabilities of the progression of radiological stage were estimated using the Kaplan-Meier product-limit method. The survivorship curves for various subgroups were compared with the log-rank test (Statview software; Abacus Concepts, Berkeley, CA, USA).
Results The operating time in the CVO group was a mean 106 min (SD 11 min, range 90–130 min), and that in the RAO group was a mean 156 min (SD 19 min, range 130– 200 min). Blood loss during and after the operative procedure in the CVO group was a mean 248 ml (SD 37 ml, range 190–320 ml), and that in the RAO group was a mean 670 ml (SD 103 ml, range 500–900 ml). Both operating time and mean blood loss were thus significantly less in the CVO group (P ⬍ 0.0001). In the CVO group, early postoperative fracture at the lesser trochanteric level was observed in one joint, and reoperation was done using an angle plate. In the RAO group, no intraoperative or early postoperative complications such as chisel perforation into the joint or postoperative movement of the rotated acetabulum were observed in any cases. The pre- and postoperative clinical scores for both groups are presented in Table 2. In the CVO group, the improvement in clinical score was mainly attributable to the scores for pain. Pain scores increased significantly from 3.3 to 4.7, whereas walking and mobility scores remained unchanged. In three patients the total clinical score decreased from 11.7 (range 11–12) to 9.7 (range 9– 10); and of these patients, two underwent total hip arthroplasty 5 and 6 years after undergoing CVO. Weakness of the abductor muscles disappeared in all
patients within 6 months postoperatively. Postoperative leg shortening ranged from 1.0 to 2.0 cm (mean 1.2 cm). Except in patients in whom the clinical score worsened, no limping due to leg shortening was observed. In the RAO group, the improved clinical score was mainly attributable to increased scores for pain reduction and increased walking ability. Pain scores increased from 3.4 to 5.4, and walking ability increased significantly from 4.4 to 5.5, whereas mobility scores remained unchanged (Table 2). Limping due to weakness of the abductor muscles had disappeared in all patients within 8 months postoperatively. The total clinical score decreased from 15 to 13 in 1 patient (1 hip). Thus far during follow-up, no patient in the RAO group has required further surgery. No significant difference in clinical scores was noted between the two groups during follow-up. Radiologically, in both the CVO and RAO groups the CE angle and head lateraligation index (HLI) showed significant improvement 3 months postoperatively, but the degree of improvement in the RAO group was greater. The improvement in the postoperative AC angle was significant in the RAO group compared with that for the CVO group, which showed no improvement (Table 3). No patient showed worsening of radiological indices. Significant changes in the CE and AC angles in both groups were seen on plain radiography when the films obtained 3 months postoperatively and at final followup were compared (Table 3, Fig. 1). Postoperative joint congruency was excellent in two hips, good in eight hips, and fair in five hips in the CVO group, and excellent in 3 hips, good in 11 hips, and fair in 4 hips in the RAO group. Poor congruency was not observed in any joint. Progression of OA was radiologically observed during follow-up in three hips (20.0%) in the CVO group and in one hip (5.6%) in the RAO group, all of which were in Charnley category B. In the CVO group, two hips progressed to the advanced stage of OA and one to the end stage; in the RAO
Y. Yasunaga et al.: Curved varus femoral osteotomy
267
Table 3. Radiological evaluation Parameter Preoperatively (SD, range) CE AC HLI 3 Months postoperatively (SD, range) CE AC HLI At follow-up (SD, range) CE AC HLI
CVO group
RAO group
7.8 (3.2, 5.0–15.0) 27 (5.0, 20–40) 0.64 (0.05, 0.56–0.72)
7.4 (2.4, 5.0–10.0) 27 (3.5, 20–33) 0.66 (0.09, 0.53–0.87)
12 (6.2, 0–30)*,## 27 (3.9, 20–33)## 0.62 (0.05, 0.50–0.68)*
37 (4.8, 30–50)**** 2.6 (5.4, 10–15)**** 0.61 (0.08, 0.45–0.80)***
17 (6.2, 10–32)**,## 33 (5.4, 23–40)**,## 0.63 (0.07, 0.50–0.71)
39 (5.0, 30–50)*** 8.2 (5.1, 0–20)**** 0.60 (0.08, 0.45–0.78)
* P ⬍ 0.05, ** P ⬍ 0.01, *** P ⬍ 0.001, **** P ⬍ 0.0001, Wilcoxon rank test ## P ⬍ 0.0001, Mann-Whitney U-test
a,b
c Fig. 1. A 52-year-old woman at the time of surgery. Her clinical score was 14 points (pain 3, mobility 6, walking 5); Charnley category B. a Early-stage osteoarthritis (OA) of the hip. Center-edge (CE) angle 5°; acetabular roof obliquity (AC) angle, 22°; head lateraligation index (HLI) 0.66. b After curved varus femoral ostetomy (CVO). CE angle 10°; AC
angle 22°; HLI 0.64. c Eight years after CVO. Clinical score 16 points (pain 5, mobility 6, walking 5); CE angle 20°; AC angle 18°; HLI 0.67. Black arrow heads show the medial margin of the subchondral bone of the acetabulum. White arrow heads show the lateral margin of the subchondral bone of the acetabulum
group, one hip progressed to the advanced stage. Among patients in whom progression occurred, postoperative joint congruency was good in one and fair in two of the three hips in the CVO group, and it was fair in one hip in the RAO group. Kaplan-Meier survivorship analysis, with radiological stage progression of OA as the endpoint, predicted a survival rate of 75% [95% confidence interval (CI) 51– 99] at 10 years in the CVO group. On the other hand, in
the RAO group the predicted survival rate was 94% (95% CI 84–100) at 10 years, demonstrating no significant difference between the two groups (log-rank test: P ⫽ 0.185;) (Fig. 2). Among bilateral cases in the CVO patient group, in eight nonoperated hips excluding two hips that underwent RAO, progressions of radiological stage were observed in two early-stage hips, which represented cases of aggravation on the operated side. Slight widening of
268
Fig. 2. Kaplan-Meier cumulative probability of osteoarthritic progression in CVO and rotational acetabular osteotomy (RAO) groups
joint space was observed in one early-stage hip. In the RAO group, slight narrowing of joint space was observed during follow-up in two of the seven hips on the nonoperated side, but they were in the early-stage of OA. In the remaining five hips, no radiological change was observed during follow-up. In the CVO group, patient satisfaction was excellent for six patients, good for four patients, fair for three patients, and poor for two patients; in the RAO group, it was excellent for seven patients, good for three patients, fair for five patients, and poor for one patient. In both groups, patients who were evaluated as having a poor result had progressive OA. There were no significant differences in poor satisfaction between the two groups (P ⫽ 0.599, Fisher’s exact probability test).
Discussion Periacetabular osteotomy procedures such as RAO are now standard for early-stage secondary OA due to hip dysplasia because they alter the position of the acetabulum en bloc, allowing coverage of the femoral head by cartilage.2–6 The results of our recent study19 indicated that RAO in elderly patients can prevent progression of OA, with a survival rate of 70% at 10 years. However, because these procedures demand great surgical skill and long-term postoperative rehabilitation, they are not usually recommended for elderly patients with relatively minimal dysplasia. If the surgical outcome is nearly equivalent, the procedure involving less operative invasion should be selected. The principle of Pauwels’ VO20 is a reduction in the combined force on the femoral head through relaxation of the iliopsoas, adductors, and abductors and extension of the lever arm. The procedure is widely accepted as a
Y. Yasunaga et al.: Curved varus femoral osteotomy
treatment for congenital hip dysplasia in children. In adults with symptomatic dysplasia, with or without secondary degenerative changes, the role of osteotomy is not as clearly defined.21–23 It is the view of Sugioka13 that the ultimate purpose of VO is to improve joint congruency in the varus position and enlarge the load-bearing joint surface, that muscle relaxation is not necessarily required, and that postoperative insufficiency of abductors and accompanying joint instability are the only negative effects. They developed the curved osteotomy technique in which the abductors at the peripheral intertrochanteric ridge are preserved. Because the osteotomy is performed only on the medial side of the abductor medium, the procedure involves almost no muscle resection, and there is much less reduction, of leg length than with subtrochanteric VO. When we compared the operating time and blood loss volume in the present study, we found that CVO was significantly less invasive than RAO and that there was no significant difference in the progression of radiological OA stage up to the eighth postoperative year in the two groups. In elderly patients with minimal hip dysplasia (CE angle more than 5° in those with unilateral involvement) CVO is thus preferred. Postoperative joint remodeling,18 which was shown as enlargement of the CE and AC angles, was also observed in the CVO group, but the degree of involvement was less than in the RAO group. This may reflect less extensive biomechanical change than that seen with RAO. Periacetabular osteotomies such as RAO are accompanied by drastic changes in joint morphology and an increase in the postoperative load stress of the articular cartilage, particularly the medial part of the acetabulum cartilage.24 Unless it is a case of prearthrosis or early stage with minimal degeneration of the articular cartilage, the joint cannot cope with biomechanical changes and OA will progress at the early postoperative stage, especially in elderly patients.25 Although the number of patients in this study was too small for us to reach strong conclusions, CVO was less surgically invasive than RAO, required a comparatively shorter period of postoperative rehabilitation, and was useful as a joint-preserving procedure in older patients with minimal dysplasia of the hip with unilateral involvement. CVO should therefore be selected for elderly patients with a CE angle greater than 5°.
References 1. Chiari K. Medial displacement osteotomy of the pelvis. Clin Orthop 1974;98:55–71. 2. De Kleuver M, Kooijman MAP, Pavlov PW, Veth RPH. Triple osteotomy of the pelvis for acetabular dysplasia. J Bone Joint Surg Br 1997;79:225–9.
Y. Yasunaga et al.: Curved varus femoral osteotomy 3. Ninomiya S, Tagawa H. Rotational acetabular osteotomy for the dysplastic hip. J Bone Joint Surg Am 1984;66:430–6. 4. Steel HH. Triple osteotomy of the innominate bone. J Bone Joint Surg Am 1973;55:343–50. 5. Trousdale RT, Ekkernkamp A, Ganz R, Wallrichs SL. Periacetabular and intertrochanteric osteotomy for the treatment of osteoarthrosis in dysplastic hips. J Bone Joint Surg Am 1995; 77:73–85. 6. Wagner H. Osteotomies for congenital hip dislocation. In: The hip. St. Louis: Mosby; 1976. p. 45–66. 7. Nakamura S, Ninomiya S, Takatori Y, Morimoto S, Ueyama T. Long-term outcome of rotational acetabular osteotomy: 145 hips followed for 10–23 years. Acta Orthop Scand 1998;69:259–65. 8. Nozawa M, Shitoto K, Matsuda K, Maezawa H, Kurosawa H. Rotational acetabular osteotomy for acetabular dysplasia: a follow-up for more than ten years. J Bone Joint Surg Br 2002;84: 59–65. 9. Yasunaga Y, Ochi M, Shimogaki K, Yamamoto S, Iwamori H. Rotational acetabular osteotomy for hip dysplasia: 61 hips followed for 8–15 years. Acta Orthop Scand 2004;75:10–5. 10. Wiberg G. Studies on dysplastic acetabula and congenital subluxation of the hip joint: with special reference to the complication of osteoarthritis. Acta Chir Scand 1939;83(suppl. 58):7–38. 11. Charnley J. Numerical grading of clinical results. In: Charnley J, editor. Low friction arthroplasty of the hip. Berlin: Springer; 1979. p. 20–4. 12. Nishio A, Sugioka Y. Transtrochanteric curved varus osteotomy. Rinshoseikeigeka 1973;8:21–9 (in Japanese). 13. Sugioka Y. Transtrochanteric curved varus osteotomy. J Musculoskeletal Syst 1993;6:739–45 (in Japanese). 14. Merle d’Aubigne R, Postel M. Functional results of hip arthroplasty with acrylic prosthesis. J Bone Joint Surg Am 1954;36:451– 75. 15. Massie WK, Howorth MB. Congenital dislocation of the hip. Part 1. Method of grading results. J Bone Joint Surg Am 1950;32:519– 31.
269 16. Hijikata H, Tagawa H, Toyoshima H, Shimode M, Takagi K, Igarashi K. Rotational acetabular osteotomy with resection of capital drop. Hip Joint 1985;11:277–82 (in Japanese). 17. Ninomiya S. Rotational acetabular osteotomy for the severely dysplastic hip in the adolescent and adult. Clin Orthop 1989;247: 127–37. 18. Yasunaga Y, Ikuta Y, Shigenobu T, Nakamura S, Yamamoto S, Nakashiro J. Rotational acetabular osteotomy for hip dysplasia; spontaneous medial enlargement of the acetabulum. Acta Orthop Scand 2001;72:8–12. 19. Yasunaga Y, Takahashi K, Ochi M, Ikuta Y, Hisatome T, Nakashiro J. Rotational acetabular osteotomy in patients fortysix years of age or older: comparison with younger patients. J Bone Joint Surg Am 2003;85:266–72. 20. Pauwels F. The place of osteotomy in the operative management of osteoarthritis of the hip. Triangle 1968;8:196–210. 21. Iwase T, Hasegawa Y, Kawamoto K, Iwasada S, Yamada K, Iwata H. Twenty years’ followup of intertrochanteric osteotomy for treatment of the dysplastic hip. Clin Orthop 1996;331:245– 55. 22. Ohsawa S, Inamori Y, Matsushita S, Norimatsu H, Ueno R. Factors influencing joint-preserving operations in the treatment of the late stages of osteoarthritis of the hip. J Bone Joint Surg Br 2000;82:369–74. 23. Pellicci PM, Hu S, Garvin KL, Salvati EA, Wilson PD Jr. Varus rotational femoral osteotomies in adults with hip dysplasia. Clin Orthop 1991;272:162–6. 24. Yasunaga Y, Ochi M, Ikuta Y, Shimogaki K, Dohi D. Rotational acetabular osteotomies: a rabbit model. Arch Orthop Trauma Surg 1997;116:74–6. 25. Yasunaga Y, Ikuta Y, Kanazawa T, Takahashi K, Hisatome T. The state of the articular cartilage at the time of surgery as an indication for rotational acetabular osteotomy. J Bone Joint Surg Br 2001;83:1001–4.
Original article Curved varus femoral osteotomy for minimal dysplastic hip in patients older than 45 years of age: comparison with rotational acetabular osteotomy Y. Yasunaga, T. Hisatome, R. Tanaka, T. Yamasaki, and M. Ochi Department of Orthopaedic Surgery, Hiroshima University School of Medicine, 1-2-3 Kasumi, Minami-ku, Hiroshima 734-8551, Japan
Abstract Periacetabular osteotomies, such as rotational acetabular osteotomy (RAO) for osteoarthritis (OA) secondary to hip dysplasia, have produced satisfactory medium- and long-term results in early OA. However, as the degree of surgical invasion of these procedures is comparatively major and long-term postoperative rehabilitation is required, this procedure is generally not indicated for those in the fifth and sixth decades of life. Curved varus femoral osteotomy (CVO), which is a less invasive procedure, is considered a worthwhile option, although to the best of our knowledge no comparative study has been reported on these two procedures in older patients. We performed a comparative study on the outcomes of CVO for minimal dysplasia of the hip with a center-edge (CE) angle greater than 5° in 15 patients (15 joints) older than 45 years of age at a mean period of 8 years after surgery and on the outcomes of RAO in 16 patients (18 joints) performed during the same period. No significant difference was observed between the two groups regarding the Merle d’Aubigne score during follow-up. Progression of osteoarthritis was radiologically observed during follow-up in three hips in the CVO group and in one hip in the RAO group. Kaplan-Meier survivorship analysis, with radiological progression of osteoarthritis as the endpoint, predicted a survival rate of 75% at 10 years in the CVO group. In the RAO group the predicted 10-year survival rate was 94%, demonstrating no significant difference between the two groups. CVO is a less invasive surgical technique than RAO, requires a comparatively shorter period of postoperative rehabilitation, and is useful as a joint-preserving procedure in older patients with minimal dysplasia of the hip with unilateral involvement. CVO should be selected for elderly patients with a CE angle greater than 5°.
Introduction Periacetabular osteotomies for osteoarthritis (OA) secondary to hip dysplasia have theoretical advantages Offprint requests to: Y. Yasunaga Received: June 30, 2004 / Accepted: January 26, 2005
over other pelvic procedures, such as Chiari osteotomy,1 because they alter the position of the acetabulum en bloc, allowing the femoral head to be covered by cartilage.2–6 These procedures have produced satisfactory medium- and long-term results in patients with early OA secondary to dysplasia of the hip.2,7–9 However, as the degree of surgical invasion of periacetabular osteotomies is comparatively major and long-term postoperative rehabilitation is required, this procedure is generally not indicated for those in the fifth and sixth decades of life. Thus, curved varus femoral osteotomy (CVO), which is a less invasive procedure, is considered a worthwhile option, although to the best of our knowledge no comparative study has been reported on these two procedures in older patients. The purpose of this report is to compare the outcomes of CVO with those of rotational acetabular osteotomy (RAO) during the fifth and sixth decades of life.
Patients and methods Patients The indications for CVO and RAO are basically equivalent: acetabular dysplasia radiologically characterized by a center-edge (CE) angle10 of less than 20° and progressive pain interfering with daily activities. Patients are considered candidates for surgery if the hip joint demonstrated improved femoral head coverage in abduction and improved joint congruency in abduction on the anteroposterior view of a plain radiograph. Abductor strength greater than 4 (on the muscle manual test) was a precondition for surgery. However, in patients with a CE angle greater than 5° and whose age was more than 45 years, it was explained that although both procedures were indicated CVO is less surgically invasive than RAO so the patients could make an informed choice of which procedure to undergo.
Y. Yasunaga et al.: Curved varus femoral osteotomy
265
Table 1. Patient profiles Variable No. of patients No. of hips Gender (male : female) Average age at surgery (years), and range Average follow-up period, and range Charnley category (A : B)
CVO group
RAO group
15 15 2 : 13 53.3 (45–60) 7 years 8 months (4–13 years) 5 : 10
20 22 2 : 18 50.5 (46–57) 8 years 10 months (4–14 years) 6 : 14
CVO, curved varus femoral osteotomy; RAO, rotational acetabular osteotomy
The study included 15 consecutive patients (15 hips) in the early stage of OA who underwent CVO between July 1991 and June 1999, whose age at surgery was more than 45 years and who were followed up for more than 4 years (Table 1). According to the Charnley classification11 there were 5 unilateral cases (category A) and 10 bilateral cases (category B). Among those in Charnley category B, RAO was performed on the contralateral side in two patients, and in the remaining eight patients the OA on the nonoperated side was in the early stage. No patients had had previous surgery. Between October 1988 and June 1999, we performed RAO on 166 hips with prearthritis and early-stage OA secondary to dysplasia. Among these patients, there were 16 (18 hips) in the early stage of OA with a CE angle greater than 5° whose age at surgery was older than 45 years and who were followed up for more than 4 years (Table 1). According to Charnley’s classification, 6 patients had unilateral involvement (A) and 10 patients bilateral involvement (B). Among patients with bilateral involvement, RAO was performed bilaterally in two. One patient had dislocation at a high position on the nonoperated side. In the remaining seven, OA on the nonoperated side was in the early stage. No patients had had previous surgery. Osteoarthritis of the hip was classified into four radiological stages based on Japanese Orthopaedic Association (JOA) guidelines: (1) prearthritis; (2) earlystage OA; (3) advanced-stage OA; and (4) end-stage OA. Procedure The CVO procedure was performed based on the “transtrochanteric curved varus osteotomy” technique developed by Nishio and Sugioka.12,13 The varus angle was determined based on femoral head coverage and improved joint congruency in abduction in the anteroposterior view of a plain radiograph; it was set at 20°– 30°. We fixed the varus-positioned femoral head with two or three screws. Active range-of-motion exercises began on the seventh postoperative day. Partial weightbearing on two crutches was allowed from the sixth postoperative week, and full weight-bearing was initi-
ated 3 months after surgery. In one patient with a CE angle of 5° who selected varus osteotomy (VO), acetabuloplasty was also performed. The RAO procedure was performed based on the technique developed by Ninomiya and Tagawa.3 We secured the rotated acetabulum with two sapphire screws (Kyocera, Kyoto, Japan). After August 1996, poly-l-lactic screws (Gunze, Kyoto, Japan) were used for fixation instead of sapphire screws. Active range-of-motion exercises began on the seventh postoperative day in RAO patients. Partial weight-bearing on two crutches was allowed from the sixth postoperative week, and full weight bearing was initiated 5–6 months after surgery when Trendelenburg’s sign had disappeared. Clinical follow-up was based on the system of Merle d’Aubigne and Postel.14 With this system, pain, mobility, and walking ability are assessed with scores from 0 to 6. CE angle, acetabular roof obliquity (AC angle),15 and the head lateralization index (HLI)16 using a revised method developed by Ninomiya17 were measured on the anteroposterior view of radiographs obtained preoperatively, 3 months postoperatively, and once yearly thereafter. Changes in the CE and AC angles were used to describe any medial and lateral remodeling of the subchondral bone of the acetabulum caused by joint remodeling.18 Radiographic measurements were performed by two of the authors (K.T. and R.T.), and the data are expressed as group mean values. Postoperative joint congruencies in both procedures were classified into four grades:19 excellent, the curvature of the acetabulum and of the femoral head was almost identical, and joint space was adequately maintained; good, the curvatures of the acetabulum and the femoral head were not identical, but joint space was adequately maintained; fair, partial narrowing of joint space occurred; poor, joint space partially disappeared. Patient satisfaction regarding pain relief at followup was classified into four grades: excellent, very satisfied; good, satisfied; fair, slightly better than preoperative condition; poor, worse than preoperative condition.
266
Y. Yasunaga et al.: Curved varus femoral osteotomy
Table 2. Clinical score Score: mean (SD, range) Preoperative score Pain Walking Mobility Follow-up score Pain Walking Mobility
CVO group
RAO group
13.2 (1.1, 11–15) 3.3 (0.6, 2–4) 4.7 (0.5, 4–5) 5.2 (0.6, 4–6) 14.9* (3.1, 9–18) 4.7* (1.4, 2–6) 4.7# (1.0, 3–6) 5.4 (0.8, 4–6)
13.4 (1.8, 10–16) 3.4 (0.8, 2–4) 4.4 (0.8, 3–6) 5.6 (0.7, 4–6) 16.6* (1.9, 13–18) 5.4* (0.7, 4–6) 5.5** (0.7, 4–6) 5.7 (0.6, 4–6)
* P ⬍ 0.0001, ** P ⬍ 0.05 Wilcoxon rank test # P ⬍ 0.05, Mann-Whitney U-test
Statistical analysis The differences between the two group means were tested using Mann-Whitney’s U-test, Wilcoxon’s rank test, and Fisher’s exact probability test. Cumulative probabilities of the progression of radiological stage were estimated using the Kaplan-Meier product-limit method. The survivorship curves for various subgroups were compared with the log-rank test (Statview software; Abacus Concepts, Berkeley, CA, USA).
Results The operating time in the CVO group was a mean 106 min (SD 11 min, range 90–130 min), and that in the RAO group was a mean 156 min (SD 19 min, range 130– 200 min). Blood loss during and after the operative procedure in the CVO group was a mean 248 ml (SD 37 ml, range 190–320 ml), and that in the RAO group was a mean 670 ml (SD 103 ml, range 500–900 ml). Both operating time and mean blood loss were thus significantly less in the CVO group (P ⬍ 0.0001). In the CVO group, early postoperative fracture at the lesser trochanteric level was observed in one joint, and reoperation was done using an angle plate. In the RAO group, no intraoperative or early postoperative complications such as chisel perforation into the joint or postoperative movement of the rotated acetabulum were observed in any cases. The pre- and postoperative clinical scores for both groups are presented in Table 2. In the CVO group, the improvement in clinical score was mainly attributable to the scores for pain. Pain scores increased significantly from 3.3 to 4.7, whereas walking and mobility scores remained unchanged. In three patients the total clinical score decreased from 11.7 (range 11–12) to 9.7 (range 9– 10); and of these patients, two underwent total hip arthroplasty 5 and 6 years after undergoing CVO. Weakness of the abductor muscles disappeared in all
patients within 6 months postoperatively. Postoperative leg shortening ranged from 1.0 to 2.0 cm (mean 1.2 cm). Except in patients in whom the clinical score worsened, no limping due to leg shortening was observed. In the RAO group, the improved clinical score was mainly attributable to increased scores for pain reduction and increased walking ability. Pain scores increased from 3.4 to 5.4, and walking ability increased significantly from 4.4 to 5.5, whereas mobility scores remained unchanged (Table 2). Limping due to weakness of the abductor muscles had disappeared in all patients within 8 months postoperatively. The total clinical score decreased from 15 to 13 in 1 patient (1 hip). Thus far during follow-up, no patient in the RAO group has required further surgery. No significant difference in clinical scores was noted between the two groups during follow-up. Radiologically, in both the CVO and RAO groups the CE angle and head lateraligation index (HLI) showed significant improvement 3 months postoperatively, but the degree of improvement in the RAO group was greater. The improvement in the postoperative AC angle was significant in the RAO group compared with that for the CVO group, which showed no improvement (Table 3). No patient showed worsening of radiological indices. Significant changes in the CE and AC angles in both groups were seen on plain radiography when the films obtained 3 months postoperatively and at final followup were compared (Table 3, Fig. 1). Postoperative joint congruency was excellent in two hips, good in eight hips, and fair in five hips in the CVO group, and excellent in 3 hips, good in 11 hips, and fair in 4 hips in the RAO group. Poor congruency was not observed in any joint. Progression of OA was radiologically observed during follow-up in three hips (20.0%) in the CVO group and in one hip (5.6%) in the RAO group, all of which were in Charnley category B. In the CVO group, two hips progressed to the advanced stage of OA and one to the end stage; in the RAO
Y. Yasunaga et al.: Curved varus femoral osteotomy
267
Table 3. Radiological evaluation Parameter Preoperatively (SD, range) CE AC HLI 3 Months postoperatively (SD, range) CE AC HLI At follow-up (SD, range) CE AC HLI
CVO group
RAO group
7.8 (3.2, 5.0–15.0) 27 (5.0, 20–40) 0.64 (0.05, 0.56–0.72)
7.4 (2.4, 5.0–10.0) 27 (3.5, 20–33) 0.66 (0.09, 0.53–0.87)
12 (6.2, 0–30)*,## 27 (3.9, 20–33)## 0.62 (0.05, 0.50–0.68)*
37 (4.8, 30–50)**** 2.6 (5.4, 10–15)**** 0.61 (0.08, 0.45–0.80)***
17 (6.2, 10–32)**,## 33 (5.4, 23–40)**,## 0.63 (0.07, 0.50–0.71)
39 (5.0, 30–50)*** 8.2 (5.1, 0–20)**** 0.60 (0.08, 0.45–0.78)
* P ⬍ 0.05, ** P ⬍ 0.01, *** P ⬍ 0.001, **** P ⬍ 0.0001, Wilcoxon rank test ## P ⬍ 0.0001, Mann-Whitney U-test
a,b
c Fig. 1. A 52-year-old woman at the time of surgery. Her clinical score was 14 points (pain 3, mobility 6, walking 5); Charnley category B. a Early-stage osteoarthritis (OA) of the hip. Center-edge (CE) angle 5°; acetabular roof obliquity (AC) angle, 22°; head lateraligation index (HLI) 0.66. b After curved varus femoral ostetomy (CVO). CE angle 10°; AC
angle 22°; HLI 0.64. c Eight years after CVO. Clinical score 16 points (pain 5, mobility 6, walking 5); CE angle 20°; AC angle 18°; HLI 0.67. Black arrow heads show the medial margin of the subchondral bone of the acetabulum. White arrow heads show the lateral margin of the subchondral bone of the acetabulum
group, one hip progressed to the advanced stage. Among patients in whom progression occurred, postoperative joint congruency was good in one and fair in two of the three hips in the CVO group, and it was fair in one hip in the RAO group. Kaplan-Meier survivorship analysis, with radiological stage progression of OA as the endpoint, predicted a survival rate of 75% [95% confidence interval (CI) 51– 99] at 10 years in the CVO group. On the other hand, in
the RAO group the predicted survival rate was 94% (95% CI 84–100) at 10 years, demonstrating no significant difference between the two groups (log-rank test: P ⫽ 0.185;) (Fig. 2). Among bilateral cases in the CVO patient group, in eight nonoperated hips excluding two hips that underwent RAO, progressions of radiological stage were observed in two early-stage hips, which represented cases of aggravation on the operated side. Slight widening of
268
Fig. 2. Kaplan-Meier cumulative probability of osteoarthritic progression in CVO and rotational acetabular osteotomy (RAO) groups
joint space was observed in one early-stage hip. In the RAO group, slight narrowing of joint space was observed during follow-up in two of the seven hips on the nonoperated side, but they were in the early-stage of OA. In the remaining five hips, no radiological change was observed during follow-up. In the CVO group, patient satisfaction was excellent for six patients, good for four patients, fair for three patients, and poor for two patients; in the RAO group, it was excellent for seven patients, good for three patients, fair for five patients, and poor for one patient. In both groups, patients who were evaluated as having a poor result had progressive OA. There were no significant differences in poor satisfaction between the two groups (P ⫽ 0.599, Fisher’s exact probability test).
Discussion Periacetabular osteotomy procedures such as RAO are now standard for early-stage secondary OA due to hip dysplasia because they alter the position of the acetabulum en bloc, allowing coverage of the femoral head by cartilage.2–6 The results of our recent study19 indicated that RAO in elderly patients can prevent progression of OA, with a survival rate of 70% at 10 years. However, because these procedures demand great surgical skill and long-term postoperative rehabilitation, they are not usually recommended for elderly patients with relatively minimal dysplasia. If the surgical outcome is nearly equivalent, the procedure involving less operative invasion should be selected. The principle of Pauwels’ VO20 is a reduction in the combined force on the femoral head through relaxation of the iliopsoas, adductors, and abductors and extension of the lever arm. The procedure is widely accepted as a
Y. Yasunaga et al.: Curved varus femoral osteotomy
treatment for congenital hip dysplasia in children. In adults with symptomatic dysplasia, with or without secondary degenerative changes, the role of osteotomy is not as clearly defined.21–23 It is the view of Sugioka13 that the ultimate purpose of VO is to improve joint congruency in the varus position and enlarge the load-bearing joint surface, that muscle relaxation is not necessarily required, and that postoperative insufficiency of abductors and accompanying joint instability are the only negative effects. They developed the curved osteotomy technique in which the abductors at the peripheral intertrochanteric ridge are preserved. Because the osteotomy is performed only on the medial side of the abductor medium, the procedure involves almost no muscle resection, and there is much less reduction, of leg length than with subtrochanteric VO. When we compared the operating time and blood loss volume in the present study, we found that CVO was significantly less invasive than RAO and that there was no significant difference in the progression of radiological OA stage up to the eighth postoperative year in the two groups. In elderly patients with minimal hip dysplasia (CE angle more than 5° in those with unilateral involvement) CVO is thus preferred. Postoperative joint remodeling,18 which was shown as enlargement of the CE and AC angles, was also observed in the CVO group, but the degree of involvement was less than in the RAO group. This may reflect less extensive biomechanical change than that seen with RAO. Periacetabular osteotomies such as RAO are accompanied by drastic changes in joint morphology and an increase in the postoperative load stress of the articular cartilage, particularly the medial part of the acetabulum cartilage.24 Unless it is a case of prearthrosis or early stage with minimal degeneration of the articular cartilage, the joint cannot cope with biomechanical changes and OA will progress at the early postoperative stage, especially in elderly patients.25 Although the number of patients in this study was too small for us to reach strong conclusions, CVO was less surgically invasive than RAO, required a comparatively shorter period of postoperative rehabilitation, and was useful as a joint-preserving procedure in older patients with minimal dysplasia of the hip with unilateral involvement. CVO should therefore be selected for elderly patients with a CE angle greater than 5°.
References 1. Chiari K. Medial displacement osteotomy of the pelvis. Clin Orthop 1974;98:55–71. 2. De Kleuver M, Kooijman MAP, Pavlov PW, Veth RPH. Triple osteotomy of the pelvis for acetabular dysplasia. J Bone Joint Surg Br 1997;79:225–9.
Y. Yasunaga et al.: Curved varus femoral osteotomy 3. Ninomiya S, Tagawa H. Rotational acetabular osteotomy for the dysplastic hip. J Bone Joint Surg Am 1984;66:430–6. 4. Steel HH. Triple osteotomy of the innominate bone. J Bone Joint Surg Am 1973;55:343–50. 5. Trousdale RT, Ekkernkamp A, Ganz R, Wallrichs SL. Periacetabular and intertrochanteric osteotomy for the treatment of osteoarthrosis in dysplastic hips. J Bone Joint Surg Am 1995; 77:73–85. 6. Wagner H. Osteotomies for congenital hip dislocation. In: The hip. St. Louis: Mosby; 1976. p. 45–66. 7. Nakamura S, Ninomiya S, Takatori Y, Morimoto S, Ueyama T. Long-term outcome of rotational acetabular osteotomy: 145 hips followed for 10–23 years. Acta Orthop Scand 1998;69:259–65. 8. Nozawa M, Shitoto K, Matsuda K, Maezawa H, Kurosawa H. Rotational acetabular osteotomy for acetabular dysplasia: a follow-up for more than ten years. J Bone Joint Surg Br 2002;84: 59–65. 9. Yasunaga Y, Ochi M, Shimogaki K, Yamamoto S, Iwamori H. Rotational acetabular osteotomy for hip dysplasia: 61 hips followed for 8–15 years. Acta Orthop Scand 2004;75:10–5. 10. Wiberg G. Studies on dysplastic acetabula and congenital subluxation of the hip joint: with special reference to the complication of osteoarthritis. Acta Chir Scand 1939;83(suppl. 58):7–38. 11. Charnley J. Numerical grading of clinical results. In: Charnley J, editor. Low friction arthroplasty of the hip. Berlin: Springer; 1979. p. 20–4. 12. Nishio A, Sugioka Y. Transtrochanteric curved varus osteotomy. Rinshoseikeigeka 1973;8:21–9 (in Japanese). 13. Sugioka Y. Transtrochanteric curved varus osteotomy. J Musculoskeletal Syst 1993;6:739–45 (in Japanese). 14. Merle d’Aubigne R, Postel M. Functional results of hip arthroplasty with acrylic prosthesis. J Bone Joint Surg Am 1954;36:451– 75. 15. Massie WK, Howorth MB. Congenital dislocation of the hip. Part 1. Method of grading results. J Bone Joint Surg Am 1950;32:519– 31.
269 16. Hijikata H, Tagawa H, Toyoshima H, Shimode M, Takagi K, Igarashi K. Rotational acetabular osteotomy with resection of capital drop. Hip Joint 1985;11:277–82 (in Japanese). 17. Ninomiya S. Rotational acetabular osteotomy for the severely dysplastic hip in the adolescent and adult. Clin Orthop 1989;247: 127–37. 18. Yasunaga Y, Ikuta Y, Shigenobu T, Nakamura S, Yamamoto S, Nakashiro J. Rotational acetabular osteotomy for hip dysplasia; spontaneous medial enlargement of the acetabulum. Acta Orthop Scand 2001;72:8–12. 19. Yasunaga Y, Takahashi K, Ochi M, Ikuta Y, Hisatome T, Nakashiro J. Rotational acetabular osteotomy in patients fortysix years of age or older: comparison with younger patients. J Bone Joint Surg Am 2003;85:266–72. 20. Pauwels F. The place of osteotomy in the operative management of osteoarthritis of the hip. Triangle 1968;8:196–210. 21. Iwase T, Hasegawa Y, Kawamoto K, Iwasada S, Yamada K, Iwata H. Twenty years’ followup of intertrochanteric osteotomy for treatment of the dysplastic hip. Clin Orthop 1996;331:245– 55. 22. Ohsawa S, Inamori Y, Matsushita S, Norimatsu H, Ueno R. Factors influencing joint-preserving operations in the treatment of the late stages of osteoarthritis of the hip. J Bone Joint Surg Br 2000;82:369–74. 23. Pellicci PM, Hu S, Garvin KL, Salvati EA, Wilson PD Jr. Varus rotational femoral osteotomies in adults with hip dysplasia. Clin Orthop 1991;272:162–6. 24. Yasunaga Y, Ochi M, Ikuta Y, Shimogaki K, Dohi D. Rotational acetabular osteotomies: a rabbit model. Arch Orthop Trauma Surg 1997;116:74–6. 25. Yasunaga Y, Ikuta Y, Kanazawa T, Takahashi K, Hisatome T. The state of the articular cartilage at the time of surgery as an indication for rotational acetabular osteotomy. J Bone Joint Surg Br 2001;83:1001–4.