- Email: [email protected]
Subtrochanteric shortening and derotational osteotomy in primary total hip arthroplasty for patients with severe hip dysplasia: 5-year follow-up
The Journal of Arthroplasty Vol. 18 No. 3 Suppl. 1 2003
Subtrochanteric Shortening and Derotational Osteotomy in Primary Total Hip Arthroplasty for Patients With Severe Hip Dysplasia 5-Year Follow-Up John L. Masonis, MD, Jig V. Patel, FRCS(Orth), Andrew Miu, FRCS, Robert B. Bourne, MD, Richard McCalden, MD, Steven J. MacDonald, MD, and Cecil H. Rorabeck, MD
Abstract: Twenty-one primary hip arthroplasties were performed in Crowe grade 3 or 4 hip dysplasia using a subtrochanteric shortening osteotomy. Average patient age was 48.2 years. Average follow-up was 5.8 years (minimum, 2 years). Femoral fixation was cemented in 10 hips and uncemented in 11 hips. All acetabular components were uncemented, with 33% requiring structural autograft. Ninety-one percent of femoral osteotomies healed without complication. Two osteotomy nonunions required revision. Two acetabular revisions were performed for malposition and polyethylene failure. Three patients experienced postoperative dislocation. One cemented femoral component was revised for loosening. However, no neurologic deficiencies were identified. Harris hip score improved from 32.5 to 73.6. Limp improved in 60% of patients and dependence on assistive walking device improved in 40% of patients. Subtrochanteric shortening osteotomy is a safe and predictable method of restoring the anatomic hip center in high developmental hip dislocation. However, the complication rate in these patients is higher than primary hip arthroplasty for osteoarthritis. Key words: hip, dysplasia, arthroplasty, shortening, osteotomy. © 2003 Elsevier Inc. All rights reserved.
Patients with untreated high developmental hip dislocations frequently develop symptoms of secondary arthritis in the fourth and fifth decades of life [1–3]. These patients present a myriad of challenges for total hip arthroplasty. The native acetabulum is hypoplastic, and its bone density is often
low because of lack of stress remodeling. The femur is small and often shows excessive neck-shaft angles and increased anteversion, which shifts the greater trochanter to a more posterior position. Soft tissues surrounding the hip joint are frequently contracted because of the chronicity of dislocation [4]. The longevity of hip arthroplasty in these patients has been improved with restoration of the anatomic hip center, which decreases the hip joint reaction force and creates an improved lever arm for the abductor musculature [5–7]. Restoration of the anatomic hip center frequently requires limb lengthening in excess of 4 cm and increases the risk of neurologic traction injury [1,8,9].
From the London Health Sciences Center, University of Western Ontario, London, Ontario, Canada. No benefits or funds were received in support of this study. Reprint requests: John L. Masonis, MD, Miller Orthopaedic Clinic, 1001 Blythe Blvd, Charlotte, NC 28203. © 2003 Elsevier Inc. All rights reserved. 0883-5403/03/1803-1016$30.00/0 doi:10.1054/arth.2003.50104
68
Subtrochanteric Shortening and Derotational Osteotomy • Masonis et al.
One option for restoring the anatomic hip center is subtrochanteric femoral shortening osteotomy. Femoral shortening osteotomy was described by Klisic and Jankovic [9] for high dislocations and adapted by Sponseller and McBeath [4] in 1993 to include simultaneous hip arthroplasty. Various techniques of shortening osteotomies have been described [9 –12]. The 2 largest series reported in the literature include a report on 25 hips treated with a step-cut shortening osteotomy based on preoperative radiographic templating [11] and a report on 25 hips treated with a transverse shortening osteotomy based on intraoperative femoral length[13]. Bruce [10] recently reported on 5 cases using a modular femoral uncemented system combined with a transverse osteotomy for femoral shortening. Major complication rates reported in the literature for total hip arthroplasty treatment of severe acetabular dysplasia using a shortening femoral osteotomy range from 12% to 41% [1,3,4,10 – 14]. We report our experience with 21 total hip arthroplasties combined with a transverse subtrochanteric shortening and derotational osteotomy with an average of 5 years of follow-up (minimum, 2 years).
69
Fig. 1. Intramedullary femoral reaming before osteotomy.
Components and Fixation
Review of a prospective database revealed 142 total hip arthroplasties performed for arthritis secondary to hip dysplasia. Chart and radiograph review was performed for 24 hips in 21 patients who underwent transverse subtrochanteric femoral shortening and rotational osteotomy at the time of primary hip arthroplasty. Twenty-one hips had greater than 2-year follow-up and were included in the study. These patients were reviewed with regards to surgical approach (posterior versus lateral), implant size, implant fixation (cemented versus uncemented), acetabular component location, use of structural graft (acetabular and femoral), femoral resection size, osteotomy healing, component loosening, dislocation, preoperative and postoperative limp, preoperative and postoperative use of assistive walking device, leg length, and Harris hip score.
All surgeries attempted restoration of the anatomic hip center. Surgical approach to the hip was lateral in 13 patients and posterior in 8 patients. Femoral component fixation was cemented (CDH stem, Zimmer, Warsaw, IN) in 10 hips. Femoral component fixation was uncemented in 11 hips. Modular femoral components (SROM, DePuy, Warsaw, IN) were used in 9 of 11 uncemented hips (Fig. 1). The remaining 2 uncemented femoral implants were nonmodular press-fit stems (MalloryHead, Biomet, Warsaw, IN; Solution, DePuy). Femoral head size was 22 mm in 20 hips and 28 mm in 1 hip. Porous-coated acetabular components with dome screws were used in all hips. Structural femoral head autograft was used in 7 (33%) of the acetabular reconstructions. The structural autograft was contoured and fixed with 2 or 3 cancellous screws in the ilium. No reinforcement rings or reconstruction cages were used.
Patient Demographics
Osteotomy
Twenty-one shortening osteotomies were performed in 19 patients. Patients included 16 women and 3 men with a mean age of 49 years (range, 21.3– 69.5) at the time of primary arthroplasty. Mean patient weight was 69.8 kg (range, 46 –114 kg). Mean follow-up was 5.8 years (range, 2.0 –11.2 years). No patients were lost to follow-up.
Transverse subtrochanteric shortening and derotational osteotomy was performed in all patients as previously described by Veal, Rorabeck, and Bourne [15]. Reaming of the femoral canal was performed before acetabular preparation (Fig. 1). In cases using modular femoral components (SROM), proximal reaming was then performed for the mod-
Materials and Methods
70 The Journal of Arthroplasty Vol. 18 No. 3 Suppl. 1 April 2003
Fig. 2. Proximal sleeve insertion and determination of proximal osteotomy level.
ular metaphyseal sleeve. After modular sleeve trial implantation, a subtrochanteric transverse osteotomy was performed 2 cm distal to the sleeve (Fig. 2). In cases using nonmodular femoral stems, initial osteotomy was performed at the inferior border of the lesser trochanter. The proximal femoral fragment was then mobilized for exposure of the acetabulum. When acetabular components were in place, a reduction of the hip was performed with the trial femoral stem seated only in the proximal fragment (Fig. 3). The amount of overlapping femoral shaft was resected for shortening (Fig. 3). This technique allows for accurate intraoperative measurement for femoral resection. The osteotomy was then reduced over the trial stem and rotational alignment was established (Fig. 4). The osteotomy site was augmented with morselized autogenous graft from the femoral head or resected femoral fragment in all cases. The osteotomy site was also augmented with cortical strut allograft and cerclage wires in 9 cases (Fig. 5). If the osteotomy site had excellent rotational stability and the stem obtained stable distal fixation, then allograft cortical strut was not used, at the discretion of the operating surgeon. Intraoperative continuous neurologic monitoring was not performed. Sciatic nerve tension was assessed intraoperatively using palpation of the nerve after hip reduction. Osteotomy healing was assessed using postoperative radiographs at 6 weeks and 3, 6, and 12 months. Criteria for union of the osteotomy in-
Fig. 3. Reduction of hip and proximal femur for measurement of femoral resection.
cluded i) callous at the osteotomy site, ii) restoration of cortical continuity between the proximal and distal fragments on the anteroposterior or lateral radiograph, and iii) no progressive migration or gapping at the osteotomy site on serial radiographs. Change in hip center was measured using pelvic radiographs. This was determined by measuring the vertical distance from the ischeal tuberosity to the center of the femoral head on both preoperative
Fig. 4. Reduction of osteotomy and hip with trial components.
Subtrochanteric Shortening and Derotational Osteotomy • Masonis et al.
71
tioned regarding the occurrence of dislocation. Limp, use of assistive walking device, and Harris hip scores were documented at each postoperative examination. Preoperative and postoperative limp was subjectively graded as none, mild, moderate, or severe. The use of assistive walking device was graded as none, cane, crutches or walker, and unable or wheelchair-bound.
Results Hip Center and Leg Length
Fig. 5. Final components with osteotomy site augmentation using cortical strut allograft and morsellized autograft.
and postoperative radiographs. Preoperative and postoperative leg lengths were measured clinically with the patient standing using spacer blocks under the shorter extremity and verifying the level of the iliac crests by palpation. Leg length discrepancy was recorded to the nearest centimeter. Component loosening was defined as either progressive symptomatic radiolucency greater than 2 mm surrounding the femoral or acetabular implant or progressive component migration on serial radiographs. Patients were examined 6 weeks, 3 months, 6 months, and yearly after hip arthroplasty. At each postoperative appointment, patients were ques-
An attempt was made to place all acetabular components at the anatomic hip center. The average hip center was translated distally 58 mm (range, 40 – 69 mm). The mean acetabular component size was 44 mm (range, 40 – 48 mm). The mean osteotomy resection was 38 mm (range, 20 – 60 mm). Average preoperative leg length discrepancy in patients with unilateral dysplasia was 42 mm (range, 35– 60 mm). Average leg lengthening at the time of surgery was 25 mm (range, 15–35 mm). Final leg length discrepancy was less than 1 cm in 9 patients, between 1 and 2 cm in 8 patients, and more than 2 cm in 4 patients. Osteotomy Nineteen of 21 (91%) osteotomies healed without complication. The average time to radiographic union was 6 months (range, 3–12 months) (Fig. 6).
Fig. 6. (A) Preoperative radiograph of bilateral high dislocations (Crowe 4). (B) Postoperative radiograph showing uncemented acetabular components and uncemented modular femoral components after subtrochanteric shortening osteotomy.
72 The Journal of Arthroplasty Vol. 18 No. 3 Suppl. 1 April 2003 Table 1. Preoperative and Postoperative Limp
Preoperative Postoperative
None
Mild
Moderate
Severe
0% 17%
0% 17%
31% 39%
69% 26%
Osteotomy nonunion occurred in 2 cases (1 uncemented and 1 cemented femoral fixation). Both nonunions were symptomatic and required revision surgery at 8 and 10 months after initial arthroplasty (see “Revision Surgery”). Limp Preoperative limp was graded as none (0%), mild (0%), moderate (31%), or severe (69%). Postoperative limp was graded as none (17%), mild (17%), moderate (39%), or severe (26%). Limp improved at least one grade in 57% of cases, remained unchanged in 35%, and became more severe in 8% (Table 1). Assistive Walking Devices Preoperative use of assistive device was none in 35%, with a cane in 43%, and with crutches or a walker in 22%. Postoperative use of assistive devices was none in 39%, with a cane in 61%, and with crutches or walker in 0% (Table 2). Hip Score The mean Harris hip score improved from 32.5 (range, 22– 45) to 73.6 (range, 42–100). The mean preoperative pain component of the Harris hip score was 17.3 (range, 10 –20). Mean postoperative pain score was 30 (range, 10 – 44). Postoperative Harris hip scores were excellent (⬎ 90) in 21%, good (80 – 89) in 21%, fair (70 –79) in 26%, and poor (⬍ 70) in 32%. Limited walking tolerance and gait abnormalities (limp) were the most frequent cause of lower Harris hip scores. Dislocation Three patients experienced postoperative dislocations. All 3 dislocations occurred in patients who had a lateral surgical approach and cemented nonmodular femoral stems at the time of arthroplasty. Dislocation became recurrent in one patient and required revision of the acetabular component to increase anteversion. The second patient sustained 2 dislocations at 7 and 15 months after the primary arthroplasty and later underwent revision at 23 months for polyethylene liner fracture. The third
patient sustained a single dislocation 7 years after the initial surgery and was managed successfully using closed reduction and bracing for 8 weeks. Revision Surgery A total of 6 revision surgeries were performed in 5 patients (24%). The reason for revision was osteotomy nonunion (2 patients), recurrent dislocation (1 patient), polyethylene fracture (1 patient), and cemented femoral component loosening at 34 months (1 patient). Nine of 10 cemented femoral components were stable at latest follow-up. Ten of 11 uncemented femoral components were stable at latest follow-up. One uncemented nonmodular femoral stem (Solution, DePuy) revealed radiographic evidence of loosening at 84 months, and revision surgery has been scheduled. No modular uncemented femoral components were revised. No primary acetabular components were revised for loosening. No infections were noted. The first patient with osteotomy nonunion underwent revision surgery at 10 months after the initial procedure. At revision, the subtrochanteric osteotomy was not united. This was revised from a cementless to cemented femoral component. The osteotomy site was treated with allograft strut augment and cerclage wires. The osteotomy healed uneventfully at 6 months with resolution of symptoms. The second osteotomy nonunion underwent revision surgery at 8 months after the initial procedure. At revision, substantial proximal osteolysis was encountered, and a proximal femoral allograft was used. This construct functioned well for 5 years. Between 5 and 8 years after revision, this patient developed femoral component loosening and varus remodeling of the proximal femur which was symptomatic. This loosening required repeat revision to a proximal femoral substituting prosthesis and constrained acetabular liner. Neurologic Injury No clinical evidence of postoperative neurologic injury was observed. The possibility of superior gluteal
Table 2. Preoperative and Postoperative Assistive Walking Device Usage
Preoperative Postoperative
None
Cane
Crutches/ Walker
Wheelchair/ Unable
35% 39%
43% 61%
22% 0%
0% 0%
Subtrochanteric Shortening and Derotational Osteotomy • Masonis et al.
nerve injury at the time of surgical exposure and resultant abductor dysfunction cannot be excluded. Electromyelographic testing was not performed.
Conclusions Total hip arthroplasty in high developmental dislocation is a challenging reconstructive procedure. Our series showed a 24% repeat surgery rate at 3 years. This number is consistent with the limited number of published reports on this topic [1,3,11–14]. Repeat surgery was directly related to the use of a shortening osteotomy in 2 patients. The transverse osteotomy union rate was identical to a recent published report using a step-cut method [11] and is technically easier to adjust intraoperatively to correct rotational abnormalities. We currently use a modular uncemented femoral component (SROM) and have had no dislocations, loosening, or revisions in these 9 patients with short- to mid-term follow-up. The average size of femoral and acetabular components was significantly smaller than standard implants. Therefore detailed preoperative templating and planning are essential. Our experience supports the use of a transverse subtrochanteric femoral osteotomy and uncemented acetabular fixation at the anatomic hip center in high developmental hip dysplasia with secondary arthritis. The subtrochanteric shortening osteotomy appears to be a safe and reliable procedure to restore the anatomic hip center and trochanteric rotation without neurologic injury. Although patients in our series obtained substantial improvement in pain and function after total hip arthroplasty, their final hip scores were lower than those for patients undergoing hip arthroplasty for degenerative osteoarthritis. Continued follow-up is required to establish the long-term results of this procedure.
References 1. Crowe JF, Mani VJ, Ranawat CS: Total hip replacement in congenital dislocation and dysplasia of the hip. J Bone Joint Surg Am 61:15, 1979
73
2. Harris WH: Etiology of arthritis of the hip. Clin Orthop 213:20, 1986 3. Symeonides PP, Pournaras J, Petsatodes G, et al: Total hip arthroplasty in neglected congenital dislocation of the hip. Clin Orthop 341: 55-61, 1997. 4. Sponseller PD, McBeath A: Subtrochanteric osteotomy for arthroplasty of the dysplastic hip. J Arthroplasty 3:151, 1988 5. Kelley SS: High hip center in revision arthroplasty. J Arthroplasty 9:503, 1994 6. Pagnano MW, Hanssen AD, Lewallen DG, Shaughnessy WJ: The effect of superior placement of the acetabular component on the rate of loosening after total hip arthroplasty: long term results of patients who have Crowe type II congenital dysplasia of the hip. J Bone Joint Surg Am 78:1004, 1996 7. Yoder SA, Brand RA, Pederson DR, O’Gorman TW: Total hip acetabular component position affects acetabular loosening rates. Clin Orthop 228:79, 1988 8. Cameron HU, Eren OT, Solomon M: Nerve injury in the prosthetic management of the dysplastic hip. Orthopedics 21:881, 1998 9. Klisic P, Jankovic L: Combined procedure of open reduction and shortening of the femur in treatment of congenital dislocation of the hips in older children. Clin Orthop 119:60, 1976 10. Bruce WJ, Rizkallah SM, Kwon YM, et al: A new technique of subtrochanteric shortening in total hip arthroplasty. J Arthroplasty 15:617, 2000 11. Sener N, Tozun I, Asik M: Femoral shortening and cementless arthroplasty in high congenital dislocation of the hip. J Arthroplasty 17:41, 2002 12. Yasgur DJ, Stuchin SA, Adler EM, et al: Subtrochanteric shortening femoral osteotomy in total hip arthroplasty for high-riding developmental dislocation of the hip. J Arthroplasty 12:880, 1997 13. Reikeraas O, Lereim P, Gabor I, et al: Femoral shortening in total hip arthroplasty for completely dislocated hips: 3-7 year results in 25 cases. Acta Orthop Scand 67:33, 1996 14. Chareancholvanich K, Becker DA, Gustilo RB: Treatment of congenital dislocated hip by arthroplasty with femoral shortening. Clin Orthop 360: 127, 1999 15. Veal GA, Rorabeck CR, Bourne RB: Subtrochanteric femoral resection in total hip arthroplasty for high riding CDH. J Orth Tech 1:33, 1993
Subtrochanteric Shortening and Derotational Osteotomy in Primary Total Hip Arthroplasty for Patients With Severe Hip Dysplasia 5-Year Follow-Up John L. Masonis, MD, Jig V. Patel, FRCS(Orth), Andrew Miu, FRCS, Robert B. Bourne, MD, Richard McCalden, MD, Steven J. MacDonald, MD, and Cecil H. Rorabeck, MD
Abstract: Twenty-one primary hip arthroplasties were performed in Crowe grade 3 or 4 hip dysplasia using a subtrochanteric shortening osteotomy. Average patient age was 48.2 years. Average follow-up was 5.8 years (minimum, 2 years). Femoral fixation was cemented in 10 hips and uncemented in 11 hips. All acetabular components were uncemented, with 33% requiring structural autograft. Ninety-one percent of femoral osteotomies healed without complication. Two osteotomy nonunions required revision. Two acetabular revisions were performed for malposition and polyethylene failure. Three patients experienced postoperative dislocation. One cemented femoral component was revised for loosening. However, no neurologic deficiencies were identified. Harris hip score improved from 32.5 to 73.6. Limp improved in 60% of patients and dependence on assistive walking device improved in 40% of patients. Subtrochanteric shortening osteotomy is a safe and predictable method of restoring the anatomic hip center in high developmental hip dislocation. However, the complication rate in these patients is higher than primary hip arthroplasty for osteoarthritis. Key words: hip, dysplasia, arthroplasty, shortening, osteotomy. © 2003 Elsevier Inc. All rights reserved.
Patients with untreated high developmental hip dislocations frequently develop symptoms of secondary arthritis in the fourth and fifth decades of life [1–3]. These patients present a myriad of challenges for total hip arthroplasty. The native acetabulum is hypoplastic, and its bone density is often
low because of lack of stress remodeling. The femur is small and often shows excessive neck-shaft angles and increased anteversion, which shifts the greater trochanter to a more posterior position. Soft tissues surrounding the hip joint are frequently contracted because of the chronicity of dislocation [4]. The longevity of hip arthroplasty in these patients has been improved with restoration of the anatomic hip center, which decreases the hip joint reaction force and creates an improved lever arm for the abductor musculature [5–7]. Restoration of the anatomic hip center frequently requires limb lengthening in excess of 4 cm and increases the risk of neurologic traction injury [1,8,9].
From the London Health Sciences Center, University of Western Ontario, London, Ontario, Canada. No benefits or funds were received in support of this study. Reprint requests: John L. Masonis, MD, Miller Orthopaedic Clinic, 1001 Blythe Blvd, Charlotte, NC 28203. © 2003 Elsevier Inc. All rights reserved. 0883-5403/03/1803-1016$30.00/0 doi:10.1054/arth.2003.50104
68
Subtrochanteric Shortening and Derotational Osteotomy • Masonis et al.
One option for restoring the anatomic hip center is subtrochanteric femoral shortening osteotomy. Femoral shortening osteotomy was described by Klisic and Jankovic [9] for high dislocations and adapted by Sponseller and McBeath [4] in 1993 to include simultaneous hip arthroplasty. Various techniques of shortening osteotomies have been described [9 –12]. The 2 largest series reported in the literature include a report on 25 hips treated with a step-cut shortening osteotomy based on preoperative radiographic templating [11] and a report on 25 hips treated with a transverse shortening osteotomy based on intraoperative femoral length[13]. Bruce [10] recently reported on 5 cases using a modular femoral uncemented system combined with a transverse osteotomy for femoral shortening. Major complication rates reported in the literature for total hip arthroplasty treatment of severe acetabular dysplasia using a shortening femoral osteotomy range from 12% to 41% [1,3,4,10 – 14]. We report our experience with 21 total hip arthroplasties combined with a transverse subtrochanteric shortening and derotational osteotomy with an average of 5 years of follow-up (minimum, 2 years).
69
Fig. 1. Intramedullary femoral reaming before osteotomy.
Components and Fixation
Review of a prospective database revealed 142 total hip arthroplasties performed for arthritis secondary to hip dysplasia. Chart and radiograph review was performed for 24 hips in 21 patients who underwent transverse subtrochanteric femoral shortening and rotational osteotomy at the time of primary hip arthroplasty. Twenty-one hips had greater than 2-year follow-up and were included in the study. These patients were reviewed with regards to surgical approach (posterior versus lateral), implant size, implant fixation (cemented versus uncemented), acetabular component location, use of structural graft (acetabular and femoral), femoral resection size, osteotomy healing, component loosening, dislocation, preoperative and postoperative limp, preoperative and postoperative use of assistive walking device, leg length, and Harris hip score.
All surgeries attempted restoration of the anatomic hip center. Surgical approach to the hip was lateral in 13 patients and posterior in 8 patients. Femoral component fixation was cemented (CDH stem, Zimmer, Warsaw, IN) in 10 hips. Femoral component fixation was uncemented in 11 hips. Modular femoral components (SROM, DePuy, Warsaw, IN) were used in 9 of 11 uncemented hips (Fig. 1). The remaining 2 uncemented femoral implants were nonmodular press-fit stems (MalloryHead, Biomet, Warsaw, IN; Solution, DePuy). Femoral head size was 22 mm in 20 hips and 28 mm in 1 hip. Porous-coated acetabular components with dome screws were used in all hips. Structural femoral head autograft was used in 7 (33%) of the acetabular reconstructions. The structural autograft was contoured and fixed with 2 or 3 cancellous screws in the ilium. No reinforcement rings or reconstruction cages were used.
Patient Demographics
Osteotomy
Twenty-one shortening osteotomies were performed in 19 patients. Patients included 16 women and 3 men with a mean age of 49 years (range, 21.3– 69.5) at the time of primary arthroplasty. Mean patient weight was 69.8 kg (range, 46 –114 kg). Mean follow-up was 5.8 years (range, 2.0 –11.2 years). No patients were lost to follow-up.
Transverse subtrochanteric shortening and derotational osteotomy was performed in all patients as previously described by Veal, Rorabeck, and Bourne [15]. Reaming of the femoral canal was performed before acetabular preparation (Fig. 1). In cases using modular femoral components (SROM), proximal reaming was then performed for the mod-
Materials and Methods
70 The Journal of Arthroplasty Vol. 18 No. 3 Suppl. 1 April 2003
Fig. 2. Proximal sleeve insertion and determination of proximal osteotomy level.
ular metaphyseal sleeve. After modular sleeve trial implantation, a subtrochanteric transverse osteotomy was performed 2 cm distal to the sleeve (Fig. 2). In cases using nonmodular femoral stems, initial osteotomy was performed at the inferior border of the lesser trochanter. The proximal femoral fragment was then mobilized for exposure of the acetabulum. When acetabular components were in place, a reduction of the hip was performed with the trial femoral stem seated only in the proximal fragment (Fig. 3). The amount of overlapping femoral shaft was resected for shortening (Fig. 3). This technique allows for accurate intraoperative measurement for femoral resection. The osteotomy was then reduced over the trial stem and rotational alignment was established (Fig. 4). The osteotomy site was augmented with morselized autogenous graft from the femoral head or resected femoral fragment in all cases. The osteotomy site was also augmented with cortical strut allograft and cerclage wires in 9 cases (Fig. 5). If the osteotomy site had excellent rotational stability and the stem obtained stable distal fixation, then allograft cortical strut was not used, at the discretion of the operating surgeon. Intraoperative continuous neurologic monitoring was not performed. Sciatic nerve tension was assessed intraoperatively using palpation of the nerve after hip reduction. Osteotomy healing was assessed using postoperative radiographs at 6 weeks and 3, 6, and 12 months. Criteria for union of the osteotomy in-
Fig. 3. Reduction of hip and proximal femur for measurement of femoral resection.
cluded i) callous at the osteotomy site, ii) restoration of cortical continuity between the proximal and distal fragments on the anteroposterior or lateral radiograph, and iii) no progressive migration or gapping at the osteotomy site on serial radiographs. Change in hip center was measured using pelvic radiographs. This was determined by measuring the vertical distance from the ischeal tuberosity to the center of the femoral head on both preoperative
Fig. 4. Reduction of osteotomy and hip with trial components.
Subtrochanteric Shortening and Derotational Osteotomy • Masonis et al.
71
tioned regarding the occurrence of dislocation. Limp, use of assistive walking device, and Harris hip scores were documented at each postoperative examination. Preoperative and postoperative limp was subjectively graded as none, mild, moderate, or severe. The use of assistive walking device was graded as none, cane, crutches or walker, and unable or wheelchair-bound.
Results Hip Center and Leg Length
Fig. 5. Final components with osteotomy site augmentation using cortical strut allograft and morsellized autograft.
and postoperative radiographs. Preoperative and postoperative leg lengths were measured clinically with the patient standing using spacer blocks under the shorter extremity and verifying the level of the iliac crests by palpation. Leg length discrepancy was recorded to the nearest centimeter. Component loosening was defined as either progressive symptomatic radiolucency greater than 2 mm surrounding the femoral or acetabular implant or progressive component migration on serial radiographs. Patients were examined 6 weeks, 3 months, 6 months, and yearly after hip arthroplasty. At each postoperative appointment, patients were ques-
An attempt was made to place all acetabular components at the anatomic hip center. The average hip center was translated distally 58 mm (range, 40 – 69 mm). The mean acetabular component size was 44 mm (range, 40 – 48 mm). The mean osteotomy resection was 38 mm (range, 20 – 60 mm). Average preoperative leg length discrepancy in patients with unilateral dysplasia was 42 mm (range, 35– 60 mm). Average leg lengthening at the time of surgery was 25 mm (range, 15–35 mm). Final leg length discrepancy was less than 1 cm in 9 patients, between 1 and 2 cm in 8 patients, and more than 2 cm in 4 patients. Osteotomy Nineteen of 21 (91%) osteotomies healed without complication. The average time to radiographic union was 6 months (range, 3–12 months) (Fig. 6).
Fig. 6. (A) Preoperative radiograph of bilateral high dislocations (Crowe 4). (B) Postoperative radiograph showing uncemented acetabular components and uncemented modular femoral components after subtrochanteric shortening osteotomy.
72 The Journal of Arthroplasty Vol. 18 No. 3 Suppl. 1 April 2003 Table 1. Preoperative and Postoperative Limp
Preoperative Postoperative
None
Mild
Moderate
Severe
0% 17%
0% 17%
31% 39%
69% 26%
Osteotomy nonunion occurred in 2 cases (1 uncemented and 1 cemented femoral fixation). Both nonunions were symptomatic and required revision surgery at 8 and 10 months after initial arthroplasty (see “Revision Surgery”). Limp Preoperative limp was graded as none (0%), mild (0%), moderate (31%), or severe (69%). Postoperative limp was graded as none (17%), mild (17%), moderate (39%), or severe (26%). Limp improved at least one grade in 57% of cases, remained unchanged in 35%, and became more severe in 8% (Table 1). Assistive Walking Devices Preoperative use of assistive device was none in 35%, with a cane in 43%, and with crutches or a walker in 22%. Postoperative use of assistive devices was none in 39%, with a cane in 61%, and with crutches or walker in 0% (Table 2). Hip Score The mean Harris hip score improved from 32.5 (range, 22– 45) to 73.6 (range, 42–100). The mean preoperative pain component of the Harris hip score was 17.3 (range, 10 –20). Mean postoperative pain score was 30 (range, 10 – 44). Postoperative Harris hip scores were excellent (⬎ 90) in 21%, good (80 – 89) in 21%, fair (70 –79) in 26%, and poor (⬍ 70) in 32%. Limited walking tolerance and gait abnormalities (limp) were the most frequent cause of lower Harris hip scores. Dislocation Three patients experienced postoperative dislocations. All 3 dislocations occurred in patients who had a lateral surgical approach and cemented nonmodular femoral stems at the time of arthroplasty. Dislocation became recurrent in one patient and required revision of the acetabular component to increase anteversion. The second patient sustained 2 dislocations at 7 and 15 months after the primary arthroplasty and later underwent revision at 23 months for polyethylene liner fracture. The third
patient sustained a single dislocation 7 years after the initial surgery and was managed successfully using closed reduction and bracing for 8 weeks. Revision Surgery A total of 6 revision surgeries were performed in 5 patients (24%). The reason for revision was osteotomy nonunion (2 patients), recurrent dislocation (1 patient), polyethylene fracture (1 patient), and cemented femoral component loosening at 34 months (1 patient). Nine of 10 cemented femoral components were stable at latest follow-up. Ten of 11 uncemented femoral components were stable at latest follow-up. One uncemented nonmodular femoral stem (Solution, DePuy) revealed radiographic evidence of loosening at 84 months, and revision surgery has been scheduled. No modular uncemented femoral components were revised. No primary acetabular components were revised for loosening. No infections were noted. The first patient with osteotomy nonunion underwent revision surgery at 10 months after the initial procedure. At revision, the subtrochanteric osteotomy was not united. This was revised from a cementless to cemented femoral component. The osteotomy site was treated with allograft strut augment and cerclage wires. The osteotomy healed uneventfully at 6 months with resolution of symptoms. The second osteotomy nonunion underwent revision surgery at 8 months after the initial procedure. At revision, substantial proximal osteolysis was encountered, and a proximal femoral allograft was used. This construct functioned well for 5 years. Between 5 and 8 years after revision, this patient developed femoral component loosening and varus remodeling of the proximal femur which was symptomatic. This loosening required repeat revision to a proximal femoral substituting prosthesis and constrained acetabular liner. Neurologic Injury No clinical evidence of postoperative neurologic injury was observed. The possibility of superior gluteal
Table 2. Preoperative and Postoperative Assistive Walking Device Usage
Preoperative Postoperative
None
Cane
Crutches/ Walker
Wheelchair/ Unable
35% 39%
43% 61%
22% 0%
0% 0%
Subtrochanteric Shortening and Derotational Osteotomy • Masonis et al.
nerve injury at the time of surgical exposure and resultant abductor dysfunction cannot be excluded. Electromyelographic testing was not performed.
Conclusions Total hip arthroplasty in high developmental dislocation is a challenging reconstructive procedure. Our series showed a 24% repeat surgery rate at 3 years. This number is consistent with the limited number of published reports on this topic [1,3,11–14]. Repeat surgery was directly related to the use of a shortening osteotomy in 2 patients. The transverse osteotomy union rate was identical to a recent published report using a step-cut method [11] and is technically easier to adjust intraoperatively to correct rotational abnormalities. We currently use a modular uncemented femoral component (SROM) and have had no dislocations, loosening, or revisions in these 9 patients with short- to mid-term follow-up. The average size of femoral and acetabular components was significantly smaller than standard implants. Therefore detailed preoperative templating and planning are essential. Our experience supports the use of a transverse subtrochanteric femoral osteotomy and uncemented acetabular fixation at the anatomic hip center in high developmental hip dysplasia with secondary arthritis. The subtrochanteric shortening osteotomy appears to be a safe and reliable procedure to restore the anatomic hip center and trochanteric rotation without neurologic injury. Although patients in our series obtained substantial improvement in pain and function after total hip arthroplasty, their final hip scores were lower than those for patients undergoing hip arthroplasty for degenerative osteoarthritis. Continued follow-up is required to establish the long-term results of this procedure.
References 1. Crowe JF, Mani VJ, Ranawat CS: Total hip replacement in congenital dislocation and dysplasia of the hip. J Bone Joint Surg Am 61:15, 1979
73
2. Harris WH: Etiology of arthritis of the hip. Clin Orthop 213:20, 1986 3. Symeonides PP, Pournaras J, Petsatodes G, et al: Total hip arthroplasty in neglected congenital dislocation of the hip. Clin Orthop 341: 55-61, 1997. 4. Sponseller PD, McBeath A: Subtrochanteric osteotomy for arthroplasty of the dysplastic hip. J Arthroplasty 3:151, 1988 5. Kelley SS: High hip center in revision arthroplasty. J Arthroplasty 9:503, 1994 6. Pagnano MW, Hanssen AD, Lewallen DG, Shaughnessy WJ: The effect of superior placement of the acetabular component on the rate of loosening after total hip arthroplasty: long term results of patients who have Crowe type II congenital dysplasia of the hip. J Bone Joint Surg Am 78:1004, 1996 7. Yoder SA, Brand RA, Pederson DR, O’Gorman TW: Total hip acetabular component position affects acetabular loosening rates. Clin Orthop 228:79, 1988 8. Cameron HU, Eren OT, Solomon M: Nerve injury in the prosthetic management of the dysplastic hip. Orthopedics 21:881, 1998 9. Klisic P, Jankovic L: Combined procedure of open reduction and shortening of the femur in treatment of congenital dislocation of the hips in older children. Clin Orthop 119:60, 1976 10. Bruce WJ, Rizkallah SM, Kwon YM, et al: A new technique of subtrochanteric shortening in total hip arthroplasty. J Arthroplasty 15:617, 2000 11. Sener N, Tozun I, Asik M: Femoral shortening and cementless arthroplasty in high congenital dislocation of the hip. J Arthroplasty 17:41, 2002 12. Yasgur DJ, Stuchin SA, Adler EM, et al: Subtrochanteric shortening femoral osteotomy in total hip arthroplasty for high-riding developmental dislocation of the hip. J Arthroplasty 12:880, 1997 13. Reikeraas O, Lereim P, Gabor I, et al: Femoral shortening in total hip arthroplasty for completely dislocated hips: 3-7 year results in 25 cases. Acta Orthop Scand 67:33, 1996 14. Chareancholvanich K, Becker DA, Gustilo RB: Treatment of congenital dislocated hip by arthroplasty with femoral shortening. Clin Orthop 360: 127, 1999 15. Veal GA, Rorabeck CR, Bourne RB: Subtrochanteric femoral resection in total hip arthroplasty for high riding CDH. J Orth Tech 1:33, 1993