Surgical containment of the femoral head in Legg-Calvé-Perthes disease (LCPD) is currently the most common treatment. It has been recently demonstrated that surgical containment produce better long-term results than nonsurgical methods of treatment with respect to sphericity of the femoral head using specific criteria.1 Associated with improved sphericity will be a decreased risk for degenerative osteoarthritis in adulthood, the major goal of treatment in LCPD.2,3 Surgical options include: (1) pelvic osteotomies; (2) proximal femoral varus osteotomies; and (3) combination of both. This portion of the symposium is dedicated to the Salter or innominate osteotomy, the most common pelvic osteotomy performed in LCPD.
Pelvic osteotomies are classified into 3 categories: (1) acetabular rotational osteotomies; (2) medial displacement or Chiari osteotomy; and (3) shelf procedures. The most commonly performed procedures in LCPD are the acetabular rotational osteotomies, particularly the Salter osteotomy. Other rotational osteotomies include the double osteotomy (Sutherland) and triple osteotomies (Steel, Tonnis). The Chiari medial displacement osteotomy and the shelf procedures are usually used for salvage of a deformed femoral head. Any pelvic osteotomy can be combined with a proximal femoral osteotomy (varus 110 to 115 degrees), particularly if the femoral head that can not be contained by a pelvic or proximal femoral varus osteotomy alone. The shelf procedure can also be combined with an acetabular rotational osteotomy, such as the Salter osteotomy, to gain further lateral coverage of the femoral head.4
The indications for acetabular rotational osteotomies in LCPD, including the Salter osteotomy, are the same as for any form of containment treatment.5,6 These include: (1) 6 to 10 years of age at clinical onset (perhaps 5 y in girls); (2) greater than one-half capital femoral epiphysis (CFE) involvement [Catterall groups III and IV,7 Salter-Thompson group B,8 and lateral pillar groups B, B/C border, and C]9; and (3) loss of femoral head containment (subluxation) on the anteroposterior radiograph. The general prerequisites include: (1) a good to full range of hip motion preoperatively, especially abduction; (2) no residual irritability; and (3) a minimally deformed femoral head.
The Salter or innominate osteotomy in LCPD is performed as described for developmental dislocation of the hip.10 It is designed to gain anterior and lateral coverage of the femoral head by acetabular rotation.11,12 The anterior and anterolateral region of the hip are the areas of greatest stress concentration. Containment is never absolute, as the femoral head is larger than the acetabulum. The improved coverage is gained at the expense of posterior coverage of the femoral head. Although difficult to visualize radiographically, this osteotomy also displaces the acetabulum 1 to 1.5 cm medially, thereby decreasing the biomechanical compression forces across the hip joint.11,12 It also displaces the femoral head distally by a similar amount improving the commonly associated lower extremity length discrepancy.13 This osteotomy depends on rotation occurring at the pubic symphysis. Postoperatively, it is desirable to maintain hip motion to minimize stiffness and enhance remodeling. The use of crutches and partial weight bearing in reliable children is recommended. Once complete healing has occurred the internal fixation devices, usually 2 or 3 threaded Steinmann pins, can be removed. The child is then allowed to return to normal activities, including sports, as tolerated.
The Salter osteotomy is a relatively easy procedure to perform but requires a precise understanding of its various steps to achieve the desired results. As a consequence, appropriate training and experience should be obtained before attempting this osteotomy or any other acetabular rotational osteotomy. Some of the key technical points are as follows:
- The oblique skin incision must be centered half way between the anterior superior iliac spine and the greater trochanter. There is a tendency to perform the incision more proximally toward the iliac crest. Unfortunately, the hip joint is more distal. A proximal incision will make the exposure of the hip capsule and sciatic notch more difficult.
- Recess the iliopsoas at the musculotendinous junction at the pelvic brim. This will enhance the opening of the osteotomy site.
- The osteotomy is made from the greater sciatic notch to just above the anterior inferior iliac spine. The graft is obtained from the anterior ilium. Retractors designed by Rang can aide in the exposure of the sciatic notch and passage of the wire.14
- Opening the osteotomy is enhanced by the use of 2 large Lewin clamps to control the ilium and the acetabular segment. The foot on the involved side is placed on the opposite thigh by positioning the involved hip in a flexed, externally rotated position. Extending the hip results in the opening of the osteotomy. The clamps are used to maintain 1.0 to 1.5 cm of anterior displacement of the acetabular segment and laterally tilting the iliac segment. It is important to align all 3 medial cortices (ilium, bone graft, and acetabular segment).
- A large iliac crest bone graft (35 degree apical angle) is inserted. Two or 3 large threaded Steinmann pins are used to secure the 3 segments. The size is determined by the width of the cancellous bone of the ilium. The diameter is large and usually varies between 3.2 and 4.0 mm.
- After wound closure, a small folded towel is placed under the lumbar spine to restore normal or physiologic lordosis. The final radiograph will now demonstrate the typical changes of the osteotomy process—increased lateral coverage of the CFE, apparent closure or narrowing of the obturator foramen, increased prominence of the ischial spine, and distal displacement of the femoral head of approximately 1.0 to 1.5 cm.
The results of treatment in LCPD have been difficult to compare due to a lack of standardization in the preoperative classification of the extent of CFE involvement (Catterall, Salter-Thompson, or lateral pillar classifications), the age when treatment was performed, the postoperative radiographic analysis of the shape of the femoral head at the completion of the disease process (Stulberg et al or Mose Circle Criteria), and the lack of matched control groups.15 Currently, the Stulberg et al3 classification is the most widely used and is the best predictor of degenerative osteoarthritis in adulthood. In this classification, class I is a spherical femoral head that is equal in size to the opposite, uninvolved hip; class II, a spherical femoral head with coxa magna and coxa breva; class III, a nonspherical (ovoid) femoral head with a congruent acetabulum; class IV, a flat femoral head with abnormalities of the femoral neck and acetabulum; and class V, a flat femoral head and normal acetabulum. Thus, class I and II are spherical femoral heads, class III, nonspherical but oval and congruent hips, and class IV and V are incongruent hips. Class I, II, and III are usually considered satisfactory results and class IV and V unsatisfactory results (Figs. 1A–1H). Whether class III hips are truly a satisfactory result is a debatable point as these patients have an increased risk for degenerative osteoarthritis as an older adult. The Mose2 circle criteria can also be used but is less popular because it is very restrictive. The sphericity of the femoral head in the anteroposterior and lateral radiograph is measured using a transparent template with concentric circles at 2 mm intervals. If the sphericity is equal in both projections the hip is rated “good.” Variance of up to 2 mm is rated “fair,” whereas a variance of 3 mm or more is rated “poor.” The good and fair results are considered satisfactory, whereas poor ratings are unsatisfactory.
Salter Osteotomy Alone
In 1980, Salter16 reported the results of his “innominate” osteotomy in 110 hips treated over a 15 years period compared with 38 hips treated years earlier by a weight-relieving sling (noncontainment treatment). All children in both groups were 6 years of age or older at clinical onset, had extensive CFE involvement (Catterall group III or IV), and mild subluxation before the head became significantly deformed. The radiographic results were assessed by the Mose circle criteria. The radiographic results in the 110 hips were 77% good and 17% fair or 94% satisfactory results and only 6% poor or unsatisfactory results. In the 38 hips treated by noncontainment, the results were 37% good and 29% fair or 66% satisfactory results and 34% poor or unsatisfactory. Salter and Brown17 expanded this series to 159 patients in 1988. Using the same criteria 79% were rated good, 16% fair (96% satisfactory) and only 4% poor. Clinically, 93% of the patients were asymptomatic.
In 1982, Ingman et al18 compared the results of 38 Salter osteotomies with 33 cases treated by recumbency in hip spica casts. They used their own clinical criteria and the Mose circle criteria radiographically to assess their results. They reported satisfactory results (good and fair) in 75% of the patients treated by Salter osteotomy compared with 82% by conservative treatment. However, when some of these patients plus additional patients were reevaluated in 1991 by Paterson et al19 the incidence of satisfactory results increased to 96%. They felt this improvement was due to femoral head remodeling during subsequent growth.
Other studies regarding the Salter osteotomy have shown a similar incidence of satisfactory clinical and radiographic results.16,18,20–29 Variations in the original technique have also been performed to try to improve on the initial results. Canale et al30 and Cottler and Donahue31 used a trapezoidal rather than a triangular shape and reported results similar to the Salter osteotomy. Yoon et al32 described a different iliac osteotomy to produce similar but more stable orientation of the osteotomy in LCPD. Sanchez Mesa and Yamhure33 developed a percutaneous technique without the need of a bone graft.
The recent multicenter study group report on LCPD reported on 438 patients with 451 involved hips.1 Three hundred forty-five hips in 337 patients were followed until skeletal maturity. The inclusion criteria for treatment were the same as described by Salter.5 There were 68 hips treated by a Salter osteotomy and 52 hips with a proximal femoral varus osteotomy. The results between these 2 surgical treatment groups were similar and significantly improved over no treatment (19 hips), range of motion (77 hips), and brace treatment (129 hips) using the Stulberg et al classification. In the Salter osteotomy group, there were 39 hips (57%) and in the proximal femoral osteotomy group, there were 34 hips (65%) with Stulberg et al class I and II results. There were 22 hips (32%) and 13 hips (25%) with Stulberg et al class III hips, respectively. Thus, 61 of 68 hips (90%) with a Salter osteotomy had a satisfactory result, as did 47 of 52 hips (90%) with a proximal femoral varus osteotomy. The remaining hips in both groups had class IV or V results.
Other studies have compared Salter osteotomy and proximal varus osteotomy.34,35 Moberg et al34 in 1997 compared Salter osteotomy and proximal femoral varus osteotomies and found similar results with respect to femoral head sphericity. However, this study did show increased femoral head coverage by the center edge angle after Salter osteotomy.34 Similar findings were reported by Kitakoji et al35 in 2005.
Few studies have followed their patients until skeletal maturity.1,36 In addition to the multicenter study,1 Ishida et al36 in 2004 reported on 32 patients (37 hips) with LCPD treated with a Salter osteotomy and followed clinically and radiographically until skeletal maturity. All hips were classified using the Salter-Thompson and the Catterall classification. They attempted to divide their patients into those younger than 7 years of age and those 7 years of age and older at presentation making comparisons with other studies more difficult. The mean age was 7.7 years (range, 4 to 12.7 y). This is a different comparison due to inclusion of patients 6 years of age or younger who have an inherently good prognosis. Overall, age was not a significant factor with respect to results according to the Mose circle criteria and the Stulberg et al ratings. There were 16 hips (43%) rated Mose good, no fair, and 21 poor results (57%). There were 11 Stulberg et al class I hips, 5 class II hips, 13 class III hips, 8 class IV hips, and no class V hips. Thus, there were 29 hips (78%) with satisfactory results and 8 hips (22%) with unsatisfactory results using this criteria.
The use of a combined Salter and proximal femoral varus osteotomies in LCPD has been performed more recently.37–41 They are generally used for patients with an older age at clinical onset (≥9 y of age), laterally extruded (subluxated), and deformed femoral heads, and in those patients in which either osteotomy alone would not provide adequate containment. The results of these studies are more standardized and due to their more recent publication allow appropriate comparative analysis. Olney and Asher37 in 1985 studied 9 patients at a mean follow-up of 4.2 years (range, 1.7 to 7.7 y). They used the clinical criteria of Ratliff42 and the radiographic criteria of Mose2 and Lloyd-Roberts et al43 in evaluating their results. The Ratliff criteria includes pain, limited range of motion, and limp. A patient is rated good if none of these findings are present, fair if 1 was present, and poor if 2 or 3 were present. Postoperatively, there were 7 good (78%) and 2 fair (22%) clinical results. The 2 patients with fair results had a persistent limp due to abductor muscle weakness from greater trochanteric overgrowth. No patient had significant pain or a functional loss of hip motion. By the Mose circle criteria, there were 5 satisfactory (3 good and 2 fair) and 4 poor or unsatisfactory results. The criteria adapted from Lloyd-Roberts et al43 improved the radiographic results to 8% or 89% satisfactory results (4 good and 4 fair) and only 1 poor or unsatisfactory results. A good result implies the femoral head is round, congruous, and has minimal loss of epiphyseal height. A fair result indicates the femoral head is not completely spherical but congruous, had no more than one fifth of the femoral head uncovered, and a mild loss of epiphyseal height. A poor result had obvious flattening of the femoral head with loss of congruity, greater than one fifth lateral extrusion, and secondary acetabular changes. Similar results were found by Crutcher and Staheli38 in 1992. They studied 14 patients using the same preoperative and postoperative clinical and radiographic criteria with the exception they also included the Stulberg et al3 rating in their postoperative radiographic evaluation. The latter demonstrated 7 class II hips, 6 class III hips (93% satisfactory), and only 1 class IV hip at a mean follow-up of 8 years (range, 5 to 10.5 y). This indicated a significant salvage rate of hips with a potentially poor prognosis. They observed that 11 of 14 hips had documentable improvement in femoral head sphericity with growth due to remodeling. Sarassa et al39 in 2008 reported on 10 older patients treated with combined osteotomies. Preoperatively there were 4 Catterall III hips and 6 Catterall IV hips. At a mean follow-up of almost 4 years, there was 1 Stulberg et al class I hip, 5 class II hips, 3 class III hips, and only 1 class IV hip. Most recently, Javid and Wedge40 in 2009 published their results on combined osteotomies in 20 older patients with LCPD. All had the clinical onset of 8 years of age or older including 6 patients who were 10 years of age or older. Surgery was performed at 9 years of age or older. All had lateral subluxation of the femoral head and were followed until skeletal maturity. They used the lateral pillar classification preoperatively and the Stulberg et al rating at maturity. There were 11 lateral pillar B, 7 lateral pillar B/C, and only 2 lateral pillar C hips preoperatively. At skeletal maturity, there were 6 Stulberg et al class II hips, 9 class III hips, and 5 class IV hips. There were no class I or V hips. The lateral pillar classification was not particularly predictive of results. They felt that the results of the combined osteotomies at skeletal maturity were improved compared with natural history of untreated hips. Vukasinovic et al41 in 2000 evaluated patients treated with a combined Salter osteotomy and a proximal femoral shortening osteotomy. They found similar results, including a better center-edge angle, in these patients. These results were similar to those of Moberg et al34 and Kitakoji et al.35
The results of the Salter osteotomy have been demonstrated to alter the natural history of untreated LCPD. The previous published studies indicate improved sphericity of the femoral head. This may be related to the continuous containment of the femoral head compared with short-term containment for individuals treated with a proximal femoral varus osteotomy that will correct with growth. The persistent containment enhances remodeling of the femoral head, which continues until skeletal maturity. Improved femoral head blood supply has demonstrated by Shim et al44 in dogs after a Salter osteotomy may shorten the healing process. Grzegorzewski et al45 observed more normalization of the acetabulum after surgical containment than with nonsurgical treatment. They concluded that this contributes considerably to the remodeling of the femoral head. With the improved sphericity of the femoral head at maturity, there should be a decreased risk for degenerative osteoarthritis in adulthood. The effects of the osteotomy with the exception of the improved acetabular coverage of the femoral head resolved with time.34,35,46,47 Saleh et al46 demonstrated that the changes in the obturator foramen resolve within 6 months. Loder47 found only a slight change in obstetric pelvimetry in female patients after pelvic osteotomies as mean follow-up of 8.5±5.2 years. The advantages of the Salter osteotomy and other acetabular rotational osteotomies compared with a proximal femoral varus osteotomy include: (1) better anterior and lateral coverage of the femoral head; (2) it does not result in further shortening of the femoral neck (coxa breva); (3) it increases the length of the involved extremity by approximately 1 cm rather than further shortening; (4) it will improve the Trendelenberg gait; (5) easier metal removal; and (6) no risk for pathologic fracture through 7 holes in the proximal femur.5,6 However, the Salter osteotomy require extensive experience to be performed satisfactorily. These are technically demanding procedures. Combined procedures can even be more challenging.
1. Herring JA, Kim HT, Browne R. Legg-Calvé-Perthes disease
. Part II: prospective multicenter study of the effect of treatment on outcome J Bone Joint Surg Am.. 2004;86:2121–2134
2. Mose K. Methods of measuring in Legg-Calvé-Perthes disease
with special regard to the prognosis Clin Orthop Relat Res.. 1980;150:103–109
3. Stulberg SD, Cooperman DR, Wallensten R. The natural history of Legg-Calvé-Perthes disease
J Bone Joint Surg Am.. 1981;63:1095–1108
4. Kuwajima SS, Crawford AH, Ishida A, et al. Comparison between Salter's innominate osteotomy
and augmented acetabuloplasty in the treatment of patients with severe Legg-Calvé-Perthes disease
: analysis of 90 hips with special reference to roentgenographic sphericity and coverage of the femoral head J Pediatr Orthop B.. 2002;11:15–28
5. Salter RB. Current concept review: the present status of surgical treatment for Legg-Perthes disease J Bone Joint Surg Am.. 1984;66:961–966
6. Thompson GH, Salter RB. Legg-Calvé-Perthes disease
: current concepts and controversies Orthop Clin North Am.. 1987;18:617–635
7. Catterall A. The natural history of perthes disease J Bone Joint Surg Br.. 1971;53:37–53
8. Salter RB, Thompson GH. Legg-Calvé-Perthes disease
: the prognostic significance of the subchondral fracture and a two-group classification of the femoral head involvement J Bone Joint Surg Am.. 1984;66:479–499
9. Herring JA, Neustadt JB, Williams JJ, et al. The lateral pillar classification
of Legg-Calvé-Perthes disease
J Pediatr Orthop.. 1992;12:143–150
10. Salter RB. Innominate osteotomy
in the treatment of congenital dislocation and subluxation of the hip J Bone Joint Surg Br.. 1961;43:518–539
11. Rab GT. Biomechanical aspects of Salter osteotomy
Clin Orthop Relat Res.. 1978;132:82–87
12. Rab GT. Containment of the hip: a theoretical comparison of osteotomies Clin Orthop Relat Res.. 1981;154:191–196
13. Grzegorzewski A, Synder M, Kozlowski P, et al. Leg length discrepancy in Legg-Calvé-Perthes disease
J Pediatr Orthop.. 2005;25:206–209
14. Thomas S, Theologis T, Wainwright AM. Range retractors for sciatic notch exposure during pelvic osteotomy J Bone Joint Surg Br.. 2005;87:408–409
15. Herring JA. A current concepts review: t: he treatment of Legg-Calvé-Perthes disease
. A critical review of the literature J Bone Joint Surg. Am.. 1994;76:448–458
16. Salter RB. Legg-Perthes disease: the scientific basis for the methods of treatment and their indications Clin Orthop Relat Res.. 1980;150:8–11
17. Salter RB, Brown LW. Legg-Perthes disease: the long-term results of innominate osteotomy
for children with a poor prognosis J Bone Joint Surg Br.. 1988;70:335
18. Ingman AM, Paterson DC, Sutherland AD. A comparison between innominate osteotomy
and hip spica in the treatment of Legg-Calvé-Perthes disease
Clin Orthop Relat Res.. 1982;163:141–147
19. Paterson DC, Leitch JM, Foster BK. Results of innominate osteotomy
in the treatment of Legg-Calvé-Perthes disease
Clin Orthop.. 1991;266:96–103
20. Salter RB. Legg–Perthes disease. Part V. Treatment by innominate osteotomy
. Instr Course Lect.
21. Barer M. Role of innominate osteotomy
in the treatment of children with Legg-Perthes-disease Clin Orthop Relat Res.. 1978;135:82–89
22. Dekker M, Van Rens THJG, Stooff TJJH. Salter's pelvic osteotomy in the treatment of perthes disease J Bone Joint Surg Br.. 1981;63:782
23. Stevens PM, Williams P, Menlaus M. Innominate osteotomy
for perthes disease J Pediatr Orthop.. 1981;1:47–54
24. Maxted MJ, Jackson RK. Innominate osteotomy
in perthes disease: a radiological survey of results J Bone Joint Surg Br.. 1985;67:399–401
25. Robinson JH Jr, Putter H, Sigmond MB, et al. Innominate osteotomy
in perthes disease J Pediatr Orthop.. 1988;8:426–435
26. Sponseller PD, Desai SS, Millis MB. Comparison of femoral and innominate osteotomies for the treatment of Legg-Calvé-Perthes disease
J Bone Joint Surg Am.. 1988;70:1131–1139
27. Denton JR. Experience with Legg-Calvé-Perthes Disease
(LCPD): 1968 to 1974 at the New York Orthopaedic Hospital Clin Orthop Relat Res.. 1980;150:36–42
28. Jasiewicz B, Zarzycka M, Zarzycki D, et al. Salter osteotomy
in perthes disease: late radiological results Ortop Traumatol Rehabil.. 2004;6:733–739
29. Kacki W, Zarzycka M, Zarzycki D, et al. Comparison of radiological results of conservative and operative treatment by Salter osteotomy
in severe cases of perthes disease Ortop Traumatol Rehabil.. 2004;6:740–747
30. Canale ST, D'Anca AF, Cotler JM, et al. Innominate osteotomy
in Legg-Calvé-Perthes disease
J Bone Joint Surg Am.. 1972;54:25–40
31. Cottler JE, Donahue J. Innominate osteotomy
in the treatment Legg-Calvé-Perthes disease
Clin Orthop Relat Res.. 1980;150:95–102
32. Yoon TR, Rowe SM, Chung JY, et al. A new innominate osteotomy
in Perthes' disease J Pediatr Orthop.. 2003;23:363–367
33. Sanchez Mesa PA, Yamhure FH. Percutaneous innominate osteotomy
without the use of bone graft for femoral head coverage in children 2 to 8 years of age J Pediatr Orthop.. 2010;19:256–263
34. Moberg A, Hansson G, Kaniklides C. Results after femoral and innominate osteotomy
in Legg-Calvé-Perthes disease
Clin Orthop Relat Res.. 1997;334:257–264
35. Kitakoji T, Hattori T, Kitoh H, et al. Which is a better method for Perthes' disease: femoral varus or Salter osteotomy
? Clin Orthop Relat Res.. 2005;430:163–170
36. Ishida A, Kuwajima SS, Laredo Filho J, et al. Salter innominate osteotomy
in the treatment of severe Legg-Calvé-Perthes disease
: clinical and radiographic results in 32 patients (37 hips) at skeletal maturity J Pediatr Orthop.. 2004;24:257–264
37. Olney BW, Asher MA. Combined innominate and femoral osteotomy for the treatment of severe Legg-Calvé-Perthes disease
J Pediatr Orthop.. 1985;5:645–651
38. Crutcher JP, Staheli LT. Combined osteotomy as a salvage procedure for severe Legg-Calvé-Perthes disease
J Pediatr Orthop.. 1992;12:151–156
39. Sarassa CA, Herrera AM, Carvajal J, et al. Early clinical and radiological outcomes after double osteotomy in patients with late presentation Legg-Calvé-Perthes disease
J Child Orthop.. 2008;2:425–429
40. Javid M, Wedge JH. Radiographic results of combined Salter innominate and femoral osteotomy in Legg-Calvé-Perthes disease
in older children J Child Orthop.. 2009;3:229–234
41. Vukasinovic Z, Slavkovic S, Milickovic S, et al. Combined salter innominate osteotomy
with femoral shortening versus other methods of treatment for Legg-Calvé-Perthes disease
J Pediatr Orthop B.. 2000;9:28–33
42. Ratliff AHC. Pseudocoxalgia J Bone Joint Surg.. 1956;38-B:498–512
43. Lloyd-Roberts GC, Catterall A, Salamon PB. A controlled study of the indications for and the results for femoral osteotomy in perthes disease J Bone Joint Surg Br.. 1976;58:31–36
44. Shim SS, Day B, Leung G. Circulatory and vascular changes in the hip following innominate osteotomy
: an experimental study Clin Orthop Relat Res.. 1981;160:258–267
45. Grzegorzewski A, Synder M, Kozlowski P, et al. The role of the acetabulum in Perthes disease J Pediatr Orthop.. 2006;26:316–321
46. Saleh JM, O'Sullivan ME, O'Brien TM. Pelvic remodeling after Salter osteotomy
J Pediatr Orthop.. 1995;15:342–345
47. Loder RT. The long-term effect of pelvic osteotomy on birth canal size Arch Orthop Trauma Surg.. 2002;122:29–34