Aug 28, 2003 - THE JOURNAL OF BONE AND JOINT SURGERY. Children's orthopaedics. Valgus femoral osteotomy for hinge abduction in Perthes' disease.
Children’s orthopaedics
Valgus femoral osteotomy for hinge abduction in Perthes’ disease DECISION-MAKING AND OUTCOMES
W. J. Yoo, I. H. Choi, C. Y. Chung, T. J. Cho, H. Y. Kim From Seoul National University Children’s Hospital, Seoul, Korea
W. J. Yoo, MD, Instructor I. H. Choi, MD, Professor and Chairman C. Y. Chung, MD, Associate Professor T.-J. Cho, MD, Associate Professor Department of Orthopaedic Surgery, Seoul National University Children’s Hospital, 28 Yongdon-dong, Chongno-gu, Seoul 110-744, Korea. H. Y. Kim, MD, Associate Professor Department of Orthopaedic Surgery, Eulji University Hospital, Mok-dong 24, Jung-gu, Taejeon 301-726, Korea. Correspondence should be sent to Professor I. H. Choi. ©2004 British Editorial Society of Bone and Joint Surgery doi:10.1302/0301-620X.86B5. 13897 $2.00 J Bone Joint Surg [Br] 2004;86-B:726-30. Received 1 October 2002; Accepted after revision 28 August 2003 726
We studied, clinically and radiologically, the growth and remodelling of 21 hips after valgus femoral osteotomy with both rotational and sagittal correction for hinge abduction in 21 patients (mean age, 9.7 years) with Perthes’ disease. The exact type of osteotomy performed was based on the pre-operative clinical and radiological assessment and the results of intra-operative dynamic arthrography. The mean IOWA hip score was 66 (34 to 76) before surgery and 92 (80 to 100) at a mean follow-up of 7.1 years (3.0 to 15.0). Radiological measurements revealed favourable remodelling of the femoral head and improved hip joint mechanics. Valgus osteotomy, with both rotational and sagittal correction, can improve symptoms, function and remodelling of the hip in patients with Perthes’ disease.
In children with severe Perthes’ disease abnormal painful hinge movement may take place during both the early and late stages of the disease. This phenomenon is thought to be due to impingement of the protruded anterolateral osteocartilaginous segment of the deformed femoral head against the acetabular rim.1-8 Valgus femoral osteotomy has been advocated in this condition, as it moves the ‘hump’ on the femoral head away from the acetabular margin, thus alleviating the hinge movement and improving joint congruity. In the literature various types of valgus osteotomy have been recommended for hinge abduction, including valgus-extension, valgusexternal rotation and valgus-flexion-internal rotation osteotomy.5,7,9-12 These differing opinions reflect a failure to appreciate the three-dimensional patho-anatomy of the hinge segment which may differ between patients. We have found that three-dimensional assessment of the spatial features of the ‘hump’ and the associated pattern of impingement is essential in order to determine whether rotational and/or sagittal correction should be part of the valgus osteotomy. In this study we report a clinical and radiological review of our experience with valgus femoral osteotomy combined with rotational and/or sagittal correction in the treatment of hinge abduction.
Patients and Methods Between 1984 and 1999, we undertook 21 consecutive proximal femoral valgus osteoto-
mies combined with rotational and/or sagittal correction in 21 patients (16 boys and five girls) with Perthes’ disease (Table I). The mean age at surgery was 9.7 years (6.1 to 15.3); 17 were older than eight years. The mean followup after surgery was 7.1 years (3.0 to 15.0). Of the 21 patients 14 were skeletally mature, and seven were near to skeletal maturity at the latest follow-up. All hips were classified as Catterall3 group III (ten hips) or IV (11 hips). They were in the re-ossification (18 hips) and residual (three hips) stage of the disease. Previous treatment consisted of containment bracing with an Atlanta brace in eight hips, hip abduction cast with or without soft tissue release in five, femoral derotational osteotomy in one and femoral varus-derotational osteotomy in two. No previous treatment has been undertaken in the remaining five hips. The presenting symptoms in all patients were pain and limitation of movement of the hip. Imaging techniques. When a diagnosis of impingement was suspected, plain anteroposterior (AP) radiographs with the hip in neutral, abduction, adduction and frog-leg lateral position in addition to a false profile view were taken. Computed tomography (CT) was carried out on some hips (3-dimensional CT in seven hips and 2-dimensional CT in five hips).13 The intra-operative range of movement of the hip was recorded and dynamic hip arthrography undertaken in all patients in order to determine the position of maximal congruity THE JOURNAL OF BONE AND JOINT SURGERY
VALGUS FEMORAL OSTEOTOMY FOR HINGE ABDUCTION IN PERTHES’ DISEASE
727
Table I. Details of the 21 patients who underwent femoral valgus osteotomy
Gender
Age at operation (years) Catterall group
1 2 3 4 5
M M M M M
10.3 8.5 6.1 6.9 8.3
6 7 8 9 10 11 12 13
M F M M M M F F
14 15 16 17† 18 19 20 21
M M M M F M M F
Case
Femoral osteotomy* (degrees) V
R
S
Secondary surgery
III III IV IV IV
15 15 20 20 17
10 20 10 20 0
0 0 0 0 10
Abductor fasciotomy
6.1 10.3 9.4 9.5 11.1 8.3 9.1 12.4
IV III III III III IV III III
20 20 15 25 25 15 18 20
0 5 5 0 -25 -10 20 -18
10 0 0 10 10 0 -10 10
7.4 9.1 8.8 15.3 10.3 15.0 9.5 11.2
IV IV IV III IV IV III IV
25 15 30 30 20 10 20 30
-20 -12 -25 -30 -10 0 0 10
0 0 0 30 0 10 0 25
Femoral varus osteotomy, greater trochanteric transfer
Shelf acetabuloplasty Contralateral distal femoral epiphysiodesis
Excision of myositis ossificans, anterior release, medial release
Follow-up period (years) 5.5 7.5 15.0 8.2 11.1 6.9 4.4 8.6 9.9 8.2 9.4 7.7 5.7 6.1 5.2 4.1 5.5 9.3 3.0 3.8 3.0
* V, valgus; R, rotation (positive values for external rotation, and negative values for internal rotation); S, sagittal component (positive values for flexion, and negative values for extension). † the femoral capital hump was also excised.
of the femoral head within the acetabulum. Lateral impingement was studied on the AP arthrograms by moving the leg in adduction and abduction combined with internal and external rotation. Anterior impingement was studied on true lateral arthrograms by moving the hip into flexion and extension. The congruent position of the superior portion of the head was also confirmed from the craniocaudal projection of the image with the hip in extension. This systematic examination enabled us to assess the position of maximal congruence or ‘best fit’ (Fig. 1). Operative techniques. Valgus osteotomy was undertaken at the intertrochanteric level, and fixed with either a blade plate or multiple threaded pins. The femoral head and neck was effectively redirected into the most congruent position by valgus osteotomy in one hip, valgus-internal rotation in five hips, valgus-external rotation in six, valgus and flexion in four, valgus-internal rotation and flexion in three, valgusexternal rotation and flexion in one, and valgus-external rotation and extension in one. A number of soft-tissue procedures were also carried out at the same time; 16 adductor tenotomies, ten psoas tenotomies, and ten abductor fasciotomies. We undertook abductor fasciotomy at the time of the osteotomy when the abductors were felt to be too tight. Post-operatively, the hips were immobilised in a hip spica for three to six weeks. In one patient, a secondary shelf acetabuloplasty was carried out 15 months after femoral osteotomy (case 8). Methods of evaluation. Growth and remodelling of the hip after valgus osteotomy was assessed clinically and radiologically. The clinical parameters were pain, limp, moveVOL. 86-B, No. 5, JULY 2004
ment of the hip, and post-operative complications. The post-operative IOWA hip score was compared with the preoperative score. Radiological measurements were made of the femoral neck-shaft angle, the acetabular cover, the width of the medial and superior joint spaces, leg-length discrepancy, the sphericity index of Mose,14 and the hips were classified using the method of Stulberg, Cooperman and Wallensten.15 The acetabular cover was measured using the method described by Stulberg et al.15 The width of the medial joint space was measured as the horizontal distance between the most medial part of the femoral head and lateral margin of tear drop. The width of the superior joint space was measured as the vertical distance between the summit of the head and the acetabulum. Leg-length discrepancy was measured by orthoroentgenograms or CT scanograms. For statistical comparisons between the values before surgery and those after surgery we used a non-parametric method, the Wilcoxon signed-rank test. Statistical significance was set at p = 0.05.
Results Clinical. The clinical results are summarised in Table II. Before surgery, all patients complained of pain or discomfort. At the latest follow-up, 18 patients had no pain, and led active lives including taking part in some sport. The remaining three had occasional aching in the groin or thigh after vigorous exercise, which did not limit the activities of daily living. The range of movement increased in all hips in all directions except adduction. At the final evaluation, 17 patients were able to walk without a noticeable limp. Four
728
W. J. YOO, I. H. CHOI, C. Y. CHUNG, T. J. CHO, H. Y. KIM
Fig. 1a
Fig. 1b
Fig. 1c
Fig. 1d
Fig. 1e
Fig. 1f
Anterior impingement in a 15-year-old boy who presented with a painful limp and an out-toeing gait (case 17). Pre-operative hip movement was 40˚ flexion, 5˚ internal rotation, 30˚ abduction, and 45˚ external rotation. As the hip was flexed beyond 40˚, the thigh abducted and externally rotated. Figure 1a – An AP radiograph of the right hip showing an almost spherical femoral head. Figure 1b – The true lateral arthrogram and c) the operative photograph of the hip in flexion (40˚) showing impingement of the anterior ‘hump’ of the femoral head (arrow heads) against the acetabular rim. The asterisk denotes the labrum. Figure 1d – The true lateral arthrogram and e) the operative photograph of the hip on extension show that the ‘hump’ (arrow heads) has moved away from the acetabular margin. The arrow denotes the femoral head and the asterisk the labrum. Figure 1f – An AP radiograph 4 years after valgus-internal rotation-flexion osteotomy and excision of the ‘hump’. The range of movement of the hip was 115˚ flexion, 25˚ internal rotation, 40˚ abduction, and 45˚ external rotation. THE JOURNAL OF BONE AND JOINT SURGERY
VALGUS FEMORAL OSTEOTOMY FOR HINGE ABDUCTION IN PERTHES’ DISEASE
729
Table II. Clinical results following valgus osteotomy for hinge movement in 21 children with Perthes’ disease
IOWA hip score Hip range of abduction movement (˚) Internal rotation Leg-length inequality (mm)
Before surgery
Latest follow-up
p value*
66 (34 to 76) 28 (10 to 40) 14 (-10 to 45) 8.0 (0 to 45)
92 (80 to 100) 51 (30 to 60) 26 (0 to 50) 6.6 (0 to 30)
0.00059 0.00012 0.00066 >0.05
* Wilcoxon signed rank test
Table III. Radiological results following valgus osteotomy for hinge movement in 21 children with Perthes’ disease
Neck-shaft angle (˚) Acetabular coverage (%) Medial joint space (mm) Superior joint space (mm)
Before surgery
Latest follow-up
p value*
127 (100 to 140) 69 (47 to 87) 10.6 (7 to 16) 5.4 (3 to 8)
132 (123 to 144) 73 (58 to 85) 8.8 (6 to 12) 6.7 (3 to 11)
0.023 0.042 0.0070 0.029
* Wilcoxon signed rank test
patients still had a mild limp; one walked with a flexed hip gait due to myositis ossificans, two with a short-leg gait and one a long-leg gait. One patient with a short-leg gait (case 10) was treated by contralateral epiphysiodesis of the distal femur. Complications. An abduction contracture developed postoperatively in one hip (case 1), which was treated by abductor fasciotomy. Diffuse myositis ossificans occurred in one hip (case 13) after valgus-internal rotation-flexion osteotomy. One patient developed overlengthening (18 mm) of the leg two years after operation (case 5), which required varus femoral osteotomy with distal transfer of the greater trochanter. Radiological. The radiological results are summarised in Table III. Sustained remodelling of the femoral neck after valgus osteotomy was observed. The mean neck-shaft angle was 127˚ before surgery, 144˚ immediately after, 135˚ two years after, and 132˚ at the latest follow-up. Acetabular cover increased slightly. The mean width of the medial joint space of the affected hip decreased from 10.6 mm before surgery to 9.6 mm immediately after, and 8.8 mm at the latest follow-up, while the contralateral side increased from 7.5 mm before surgery to 8.5 mm at the latest follow-up. The mean width of the superior joint space of the affected side increased from 5.4 mm before surgery to 6.2 mm immediately after, and 6.7 mm at the latest follow-up, while it decreased on the contralateral side from 5.3 mm before surgery to 5.1 mm at the latest follow-up. The mean Mose sphericity index was 4.7 mm (1 to 13) at the latest followup. The final Stulberg classification of the radiographs was class II in three (14.3%), class III in 12 (57.1%), and class IV in six hips (28.6%).
Discussion The rationale for valgus femoral osteotomy in the treatment of hinge abduction in Perthes’ disease is to alleviate abnormal hinge movement and to re-align the leg with the VOL. 86-B, No. 5, JULY 2004
hip in the position of best fit in the neutral weight-bearing position. In order to achieve this the nature of the abnormal hinge movement must be carefully assessed. Little attention has been paid to the spatial features of the ‘hump’ on the femoral head which causes hinge abduction. We found two patients (cases 13, 17) in whom the main ‘hump’ on the femoral head was located relatively anteriorly. These patients had typical physical findings before surgery, namely a conspicuous out-toeing gait, and pain on passive flexion of the hip. As flexion was increased from the extended position, the thigh of the affected leg abducted and externally rotated (Fig. 1). We think that these two patients are probably similar to those described as having “anterior impingement” by Snow et al.8 We also saw one patient (case 21) who had a marked in-toe gait. She had limited hip extension when walking and complained of pain on passive extension of the hip. These different clinical and radiological signs suggest that ‘hinge abduction’ may be a manifestation of an abnormal hinge movement which varies according to whether the impingement is lateral or anterior. It is, therefore, important to consider the threedimensional nature of the hinge movement when planning surgical treatment. Valgus-extension osteotomy, popularised by Bombelli2 and Quain and Catterall,7 increases articular congruity by rotating the inferomedial part of the head into the weightbearing surface moving the femoral head ‘hump’ away from the acetabular margin. The extension component of the osteotomy is designed to reduce compressive forces on the anterolateral part of the head and to relocate its anterior quadrant in the acetabulum. We would agree that a valgus-extension osteotomy is an appropriate procedure, when hinge abduction is combined with a flexion deformity of the hip.7 When the ‘hump’ on the femoral head is located anteriorly and is too large to be contained with the acetabulum, however, valgus-extension osteotomy may not be indicated because the extension component can cause an-
730
W. J. YOO, I. H. CHOI, C. Y. CHUNG, T. J. CHO, H. Y. KIM
terior impingement from the ‘hump’. In our study, those hips that showed maximal congruity on the dynamic arthrogram with the extended leg in abduction and internal rotation were treated by valgus-external rotation osteotomy as has been suggested by Raney et al.10 By contrast, those hips that were most congruent in a position of adduction and external rotation with the leg extended were treated by a valgus and flexion-internal rotation osteotomy as suggested by Kim and Wenger.11,12 Based on a three-dimensional CT they claimed that some hips may demonstrate “functional coxa vara” and “functional retrotorsion” in late Perthes’ disease. There is also controversy as to whether an acetabular procedure such as an innominate osteotomy, shelf acetabuloplasty or Chiari osteotomy, should be undertaken to cover the extruded femoral head at the same time as the valgus osteotomy or as a second stage procedure. Kim and Wenger12 recommended a combined acetabular procedure (triple innominate osteotomy) with femoral valgus and flexion-internal rotation osteotomy to prevent “potential incongruity” which might develop after femoral osteotomy alone. They assessed incongruity between the femoral head and acetabulum using plain radiographs and three-dimensional CT images. Bankes et al9 reported favourable remodelling of the femoral head ten years after valgus-extension osteotomy for hinge abduction, without any simultaeneous acetabular procedure. They reported that the prominent lateral ridge on the femoral head disappeared in eight of nine patients. We also observed that the impinging hump reduced in size with time after surgery. These findings suggest that Chiari osteotomy or other pelvic procedures should be delayed in the younger patient following valgus osteotomy, unless the hip is markedly uncovered and unstable as lateral acetabular growth often improves remarkably once the pressure due to hinging is relieved. In our study, a secondary shelf acetabuloplasty was carried out 15 months after valgus osteotomy in one patient. Of the remaining 20 patients, 15 are still under observation and may require a secondary acetabular procedure in the future for unsatisfactory acetabular cover (less than 80%) according to Sundt.16 Our results of valgus femoral osteotomy with rotational and sagittal correction are comparable to those reported by others.7 The most conspicuous post-operative finding was the relief of pain, which disappeared in all except three patients. We believe that valgus osteotomy is a more effective treatment for pain than pelvic procedures such as Chiari osteotomy or a shelf procedure because it produces intact cartilage to cartilage contact between the femoral
head and the acetabulum and eliminates hinge abduction. Follow-up radiological studies showed sustained remodelling of the femoral head and acetabulum with time after the osteotomy. The neck-shaft angle decreased gradually. There was no increase in lateral subluxation and the increased superior joint space was maintained. However, these radiological findings do not preclude the possibility of the development of degenerative arthritis in the future. In summary, hinge abduction appears to be a complex manifestation of the hinge movement which occurs in the continuum between lateral and anterior impingement. It is essential to assess carefully the nature of the abnormal hinge movement in order to determine the optimal congruent position of the hip. Valgus osteotomy combined with appropriate rotational and sagittal components results in the sustained improvement of symptoms and function, and beneficially influences remodeling of the hip in a subgroup of children with hinge abduction caused by Perthes’ disease. 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.
References 1. Grossbard GD. Hip pain during adolescence after Perthes’ disease. J Bone Joint Surg [Br] 1981;63-B:572-4. 2. Bombelli R. Structure and function in normal and abnormal hips: how to rescue mechanically jeopardized hips. Third ed. New York: Springer-Verlag 1993:123-44. 3. Catterall A. The natural history of Perthes’ disease. J Bone Joint Surg [Br] 1971;53B:37-53. 4. Catterall A. Legg-Calve-Perthes syndrome. Clin Orthop 1981;158:41-52. 5. Catterall A. Adolescent hip pain after Perthes’ disease. Clin Orthop 1986;209:65-9. 6. Reinker KA. Early diagnosis and treatment of hinge abduction in Legg-Perthes disease. J Pediatr Orthop 1996;16:3-9. 7. Quain S, Catterall A. Hinge abduction of the hip: diagnosis and treatment. J Bone Joint Surg [Br] 1986;68-B:61-4. 8. Snow SW, Keret D, Scarangella S, Bowen JR. Anterior impingement of the femoral head: a late phenomenon of Legg-Calve-Perthes’ disease. J Pediatr Orthop 1993;13:286-9. 9. Bankes MJ, Catterall A, Hashemi-Nejad A. Valgus extension osteotomy for ‘hinge abduction’ in Perthes’ disease: results at maturity and factors influencing the radiological outcome. J Bone Joint Surg [Br[ 2000;82-B:548-54. 10. Raney EM, Grogan DP, Hurley ME, Ogden MJ. The role of proximal femoral valgus osteotomy in Legg-Calve-Perthes disease. Orthop 2002;25:513-7. 11. Kim HT, Wenger DR. “Functional retroversion” of the femoral head in Legg-CalvePerthes disease and epiphyseal dysplasia: analysis of head-neck deformity and its effect on limb position using three-dimensional computed tomography. J Pediatr Orthop 1997;17:240-6. 12. Kim HT, Wenger DR. Surgical correction of “functional retroversion” and “functional coxa vara” in late Legg-Calve-Perthes disease and epiphyseal dysplasia: correction of deformity defined by new imaging modalities. J Pediatr Orthop 1997;17:247-54. 13. Lee DY, Choi IH, Lee CK, Cho TJ. Assessment of complex hip deformity using threedimensional CT image. J Pediatr Orthop 1991;11:13-19. 14. Mose K. Methods of measuring in Legg-Calve-Perthes disease with special regard to the prognosis. Clin Orthop 1980;150:103-9. 15. Stulberg SD, Cooperman DR, Wallensten R. The natural history of Legg-CalvePerthes disease. J Bone Joint Surg [Am] 1981;63-A:1095-108. 16. Sundt H. Further investigation respecting malum coxae Calvé-Legg-Perthes’ disease with special regard to the prognosis and treatment. Acta Chir Scand 1949:Suppl 148.
THE JOURNAL OF BONE AND JOINT SURGERY
