Long-Term Results of Cementless Total Hip Arthroplasty With Subtrochanteric Shortening Osteotomy in Crowe Type IV Developmental Dysplasia

The Journal of Arthroplasty xxx (2016) 1e9

Contents lists available at ScienceDirect

The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org

Original Article

Long-Term Results of Cementless Total Hip Arthroplasty With Subtrochanteric Shortening Osteotomy in Crowe Type IV Developmental Dysplasia Duan Wang, MD, Ling-Li Li, MD, Hao-Yang Wang, MD, Fu-Xing Pei, MD, Zong-Ke Zhou, MD, PhD * Department of Orthopedics, West China Hospital, West China School of Medicine, Sichuan University, Chengdu, People's Republic of China

a r t i c l e i n f o

a b s t r a c t

Article history: Received 7 August 2016 Received in revised form 27 October 2016 Accepted 8 November 2016 Available online xxx

Background: When surgeons reconstruct hips with a high dislocation related to severe developmental dysplasia of the hip (DDH) in total hip arthroplasty (THA), archiving long-term stable implant fixation and improving patient function and satisfaction remain challenging. The purpose of this study was to evaluate the 10-year outcomes of transverse subtrochanteric shortening osteotomy in cementless, modular THA in Crowe type IV-Hartofilakidis type III DDH. Methods: We reviewed 62 patients (76 hips) who underwent cementless THA with transverse subtrochanteric shortening osteotomy from 2002-2010. There were 49 women and 13 men with a mean age of 38.8 years, all of whom had Crowe type IV DDH. Mean follow-up period was 10 years. The acetabular cup was implanted in placement of the anatomical hip center in all hips. Results: The mean Harris Hip Score significantly improved from 38.8 points to 86.1 points. Similarly, modified Merle d'Aubigne and Postel Hip Score, Hip dysfunction and Osteoarthritis Outcome Score, and SF-12 also significantly improved. The mean limb length discrepancy was reduced from 4.3 cm to 1.0 cm. At mean follow-up of 10 years, there were 3 cases of postoperative dislocation, 2 cases of transient nerve palsy, 1 case of nonunion, and 4 cases of intraoperative fracture. Revision surgery was performed in 2 patients due to isolated loosening of acetabular component and femoral stem, respectively. Conclusion: Our data demonstrated that the cementless, modular THA combined with transverse subtrochanteric shortening osteotomy was an effective and reliable technique with high rates of successful fixation of the implants and satisfactory clinical outcomes. © 2016 Elsevier Inc. All rights reserved.

Keywords: total hip arthroplasty shortening osteotomy developmental dysplasia of the hip limb length discrepancy long-term survivorship

Total hip arthroplasty (THA) for the treatment of Crowe type IV-Hartofilakidis type III developmental dysplasia of the hip (DDH) is a technically demanding procedure due to triangular-shaped and hypoplastic acetabulum filled with fibrous tissue and fat [1,2]; femoral deformities with straight and narrow medullary canal [3,4]; soft tissue abnormalities including hypertrophic capsule and shortened abductor muscle [5,6]; and biomechanical alterations Funding: This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. No author associated with this paper has disclosed any potential or pertinent conflicts which may be perceived to have impending conflict with this work. For full disclosure statements refer to http://dx.doi.org/10.1016/j.arth.2016.11.005. Duan Wang, Ling-Li Li, and Hao-Yang Wang contributed equally to this work. * Reprint requests: Zong-Ke Zhou, MD, PhD, Department of Orthopaedics, West China Hospital, Sichuan University, 37# Wuhou Guoxue Road, Chengdu 610041, People' Republic of China. http://dx.doi.org/10.1016/j.arth.2016.11.005 0883-5403/© 2016 Elsevier Inc. All rights reserved.

[7]. In Crowe type IV dysplasia, the true acetabulum is the best region to place a cup with sufficient bone stock and good biomechanics. Compared with nonanatomic hip center, the placement of acetabular components in an anatomic position promotes longterm durability with lower loosening and aseptic revision rates [8]. In addition, the reduction of the femoral head into the true acetabulum has been reported to yield the durable results in patients with Crowe type IV DDH [9-11]. However, restoration of the anatomical hip rotation center can generally lead to a hip that is difficult to reduce due to soft tissue contracture, and a limb that is excessively lengthened due to excessive soft tissue tension, abductor impairment, and nerve palsy [6,12]. Therefore, femoral shortening is a useful or sometimes necessary technique to facilitate reduction without stretching the sciatic nerve, equalize limb lengths, and overcome the contractures [10,13,14]. Various subtrochanteric osteotomy techniques with different cutting shape, such as transverse, oblique chevron, or

2

D. Wang et al. / The Journal of Arthroplasty xxx (2016) 1e9

Z-shaped, have been previous described with good clinical results [11,15-18]. The risk of fracture and nonunion remains the major concern in the procedure. Compared with a monoblock stem, a modular stem could be desirable to fit and fill in the straight and small femoral canal of a severely dysplasia hip and to stabilize the metaphyseal and diaphyseal fragments with good torsional stability at the osteotomy site [9]. And, a modular stem is also beneficial as these patients may have excessive anteversion and thus the ability to place the proximal segment separate from the diaphyseal piece [14,19]. We utilized a modular stem with simultaneous transverse subtrochanteric shortening osteotomy for Crowe IV DDH in our center. Several recent series have documented short-term or mid-term results of a modest number of cementless THAs with use of a modular stem for Crowe IV DDH [9-11,17,20-22]. However, some prior reports have included cemented and cementless acetabular and femoral components [4,9-11], while others have included hips that were treated with no femoral osteotomy [21,22] and that were treated with mixed groups of different anatomic levels and cutting shapes [17,20]. In addition, reliable long-term clinical and radiographic outcomes have not been previously described with large sample size. The purpose of this study was to evaluate 10-year functional and radiographic results of cementless and modular THAs combined with transverse subtrochanteric shortening osteotomy in a group of consecutive hips with Crowe type IV-Hartofilakidis type III DDH. Patients and Methods The retrospective study protocol was approved by the Institutional Review Board, and informed consent was obtained from all patients. From September 2002 to December 2010, a total of 65 consecutive patients (81 hips) with Crowe IV-Hartofilakidis type III developmental dysplasia were treated with cementless THA at our institution. One patient (2 hips) was lost to follow-up after surgery and could not been contacted via E-mail and telephone. Two patients (3 hips) were dead due to causes unrelated to the THA at 3 years after surgery. The left 76 hips in 62 patients (48 with unilateral and 14 with bilateral THA) who were treated with THA combined with simultaneous shortening subtrochanteric osteotomy were eligible in this study. Preoperative clinical evaluations, surgical data, and postoperative clinical and radiographic examinations were available. These data in total joint registry were collected for all patients. Clinical Data The indications for THA were severe pain unresponsive to nonsurgical management, resulting in stiffness, limping, and low quality of life. The eligible patients included 13 men and 49 women with a mean age of 46.5 years (range 19-73) at the time of surgery. The average weight was 61.9 kg (range 34-92). The average height was 155.6 cm (range 143-178), and the average body mass index was 25.4 kg/m2 (range 21-29.8). In our cases, the mean operative time was 122 min (range 89-149), and the average blood loss was 1.05 L (range 0.7-1.6). At clinical evaluation, clinical follow-up was conducted at 3 weeks, 6 weeks, 12 weeks, and 6 months after surgery and annually thereafter until the last follow-up. Clinical evaluation protocol was utilized for all patients with the use of Harris Hip Score, the Merle d'Aubigne and Postel Hip Score (including pain, range of motion, and gait function), Hip dysfunction and Osteoarthritis Outcome Score (covering pain, symptoms, daily living, sports, and quality of life), and SF-12 scale. The preoperative or postoperative limb length discrepancy (LLD) and the severity of limp were also recorded. The

distance between the anterior superior iliac spine and the medial malleolus represented the LLD, suggesting the length discrepancy of lower extremities. In addition, abduction strength was evaluated by the Trendelenburg test. All complications were reviewed. Surgical Technique For preoperative templating, an acetabular template was placed at the true anatomic hip center and a femoral template was placed for the femoral component. When the templating indicated that the maximum amount of lengthening was >3-4 cm, the length of subtrochanteric shortening bone was planned to equalize the leg lengths and lengthen the limb by <3-4 cm. In addition, 3-dimensional computed tomography (CT) scans were utilized on a routine basis to evaluate the acetabular bone defects. All procedures were performed via a posterolateral approach with the patient in the lateral decubitus position. After total capsulotomy, we resected femoral head and removed the osteophytes and fibrous scar tissue to recognize the true acetabulum before reaming. The acetabulum was gradually reamed with hemispherical reamers to reach the medial wall of the true acetabulum, with bleeding cancellous bone. The porous-coated acetabular component (DePuy, Pinnacle, Warsaw, IN) was inserted in the anatomic acetabular position with the use of press-fit technique and fixed with dome screws in all hips. If the acetabular bone was deficient, bulk bone autografts with the resected femoral head were utilized to provide adequate coverage of acetabular cup. The median diameter of the acetabular cup was 42 mm (range 40-46). The median femoral head size was 28 mm in 34 hips, 32 mm in 30 hips, and 36 mm in 12 hips. Ceramic-on-ceramic wear bearing material was used in 60 hips, and metal-on-poly in 16 hips. After insertion of acetabular cup, attention was then redirected to the femoral preparation. The sequentially proximal femur was reamed with straight reamers to archive appropriate stem size. A transverse subtrochanteric osteotomy was carried out, usually at 1-2 cm beneath the lesser trochanter, to shorten the femur by the planned amount and decrease the risk of excessively stretching sciatic nerve. A mean amount of femoral bone resected was 2.68 cm (range 2.0-4.6). If hip reduction with a femoral trial stem was impossible, additional bone was gradually resected at the osteotomy site to archive satisfactory hip reduction and avoid excessive tension on the sciatic nerve. The 2 osteotomy fragments were held with autogenous cortical bone plate affixed with 2 cables or wires to prevent intraoperative splitting of the femur and enhance biologic healing. In addition, autogenous cancellous bone cut from the resected femoral bone fragment was used to fill any gap at the osteotomy site. After osteotomy had been completed, the distal part of the femur was prepared for implantation of a cementless modular femoral stem (S-ROM, DePuy). When the rotational alignment of the 2 fragments was adjusted to allow approximately 15 -20 of anteversion of trial stem, the femoral components were inserted into the femur in all cases. All patients were encouraged to conduct early mobilization and limb exercise on a bed immediately after surgery. The mean hospital stay was 7.5 days (range 6-14). The patients walked with partial weight bearing for approximately 2 weeks, and then gradually progressive full weight bearing was allowed at 4-6 weeks after surgery depending on the stability of femoral stem and positively osseous healing at osteotomy site. Radiographic Analysis Anteroposterior and lateral radiographs of the hip, full-length view of the lower extremities, and 3-dimensional computed tomography of the hip were taken and reviewed at each follow-up

D. Wang et al. / The Journal of Arthroplasty xxx (2016) 1e9

time point. The cup inclination was measured as previously described by Mu et al [1]. The amount of limb lengthening was measured on radiographs by subtracting the distance of femoral shorting from the distance between the tip of the greater trochanter preoperatively and postoperatively [9]. Bone union at the osteotomy site was assessed on postoperative radiographs by criteria for bone union as described by Masonis et al [23]. The serial radiographs were also evaluated for the evidence of component migration, heterotopic ossification, osteolysis, subsidence, and linear polyethylene wear. The 7 zones around the femoral component were described by Gruen et al [24]. The acetabular component loosening was defined as progressive radiolucent lines of >2 mm around the inserted cup, or migration, or a change in the position of the cup [25]. The femoral component loosening was evaluated by radiographic analysis as described by Engh et al [26]. The osteointegration of the femoral stem was classified as bone ingrown, fibrous stable, or loose according to the classification described by Engh et al [26]. Subsidence of the femoral component was defined by the method described by Heekin et al [27]. Statistical Analysis Two-sided, paired Student t-test was used to analyze preoperative and postoperative continuous variables, with P < .05 considered to be significant. The data are presented as mean values with ranges. Kaplan-Meier was utilized in the analysis of survivorship and conducted with the following study end points for survival: (1) revision for any reason with any component and (2) revision for radiographic loosening for any component. Statistical analysis was performed with the use of SPSS Statistics software version 19.0 (IBM, Armonk, NY). Results Clinical Results The mean duration of follow-up was 10 years (range 6.6-13.2). The mean Harris Hip Score improved significantly from 38.8 points (range 21-59) preoperatively to 86.1 points (range 76-98) postoperatively. The mean modified Merle d'Aubigne and Postel Hip Score was 6.7 points (range 3-10) preoperatively and 15.9 points (range 12-17) postoperatively, and the difference revealed significance. The Hip dysfunction and Osteoarthritis Outcome Score and SF-12 score improved postoperatively compared with these preoperatively, and this difference was significant (Table 1). Preoperative and postoperative LLD was 4.3 cm (range 2.1-6.5) and 1.0 cm (range 0.6-1.7), respectively (Table 1). The postoperative LLD in the unrevised hips was <1 cm in 37 cases and between 1 and 2 cm in 24 cases. No cases with >2 cm of LLD was identified by the final follow-up. Two of these patients complained of shorter ipsilateral limb length than the nonoperative side for unilateral THA, but the physical measurement indicated that no operative limb was shorter than the contralateral side. At the 5-month follow-up, 2 patients reported no such shorter LLD. In addition, the mean leg lengthening was 2.2 cm (range 1.2-3.8) postoperatively. Moreover, postoperative limp was moderate in 2 cases, slight in 15, and none in 45. Two patients (2 hips) had suffered from severe pain in the nonoperative limb due to hip osteoarthritis, which may explain moderate limp.

3

Table 1 Clinical Results. Indicator Harris Hip Score Mean in points (range) Rating (no. of hips) Excellent (90-100 points) Good (80-89 points) Fair (70-79 points) Poor (<70) Modified MAP Mean in points (range) Pain Walking ROM Limp (no. of hips) Severe Moderate Slight None Limb length discrepancy Mean in cm (range) <1 (no. of hips) 1-2 (no. of hips) 2-3 (no. of hips) 3-4 (no. of hips) 4-5 (no. of hips) >5 (no. of hips) Trendelenburg sign (no. of hips) Yes No SF-12 PCS MCS HOOS Symptoms Pain Daily living Sports and recreational activities Quality of life

Preoperative

Postoperative

P Valuea

38.8 (21-59)

86.1 (76-98)

<.01

0 0 0 76

15 56 3 0

6.7 2.3 2.3 2.1

(3-10) (1-4) (1-3) (1-3)

15.9 (12-17) 5.4 (4-6) 5.3 (4-6) 5.2 (4-6)

<.01 <.01 <.01 <.01

41 25 10 0

0 2 16 58

4.3 (2.1-6.5) 0 0 10 19 26 21

1.0 (0.6-1.7) 45 30 0 0 0 0

74 2

1 76

10.9 (6-17) 13.9 (9-20)

22.4 (19-25) 25.5 (22-29)

<.01 <.01

8.8 (4-14) 16.3 (6-24) 29.2 (19-37) 5.5 (3-8) 4.8 (1-8)

16.5 36.5 61.1 12.5 13.6

<.01 <.01 <.01 <.01 <.01

(14-19) (34-39) (56-66) (10-15) (12-15)

<.01

Values are expressed as mean with range. MAP, Merle d'Aubigne and Postel; PCS, physical component summary; MCS, mental component summary; ROM, range of motion; HOOS, Hip dysfunction and Osteoarthritis Outcome Score; SF-12, 12-item short-form health survey questionnaire. a P < .05 is significant.

and stable fibrous ingrowth in 11 hips at the final follow-up according to the Engh classification. One hip had progressive radiolucent lines around the proximal femoral stem (zone 1 and 7), which resulted in progressive stem subsidence and then loosening of the femoral component. Stem revision with fully porous-coated stem was conducted 3.4 years after the index arthroplasty, and the new femoral component was stable. At the final follow-up, all acetabular components remained in situ and stable radiographically in unrevised hips and required no revision at the time of the last follow-up (Fig. 1). One patient had acetabular and greater trochanter fracture resulting in screw breakage and loosening of the acetabular cup at 7 years, and the acetabular component was revised. Asymptomatic heterotopic ossification was seen on followup radiographs in 3 hips (1 in class I and 2 in class II with Brooker classification system) [28]. In addition, 2 hips had small focal osteolysis in the greater trochanter on follow-up radiographs, but no surgery was required.

Radiographic Outcomes

Complications

Solid bone union in 75 hips was archived at the osteotomy site without any complication, with the mean time of union being 6 months (range 4-8). Stable bony ingrowth was identified in 64 hips,

Intraoperative femoral fracture occurred in 4 hips (1 in distal and 3 in proximal femur) without displaced cracks during bone preparation. All fractures were treated with cerclage cable and

4

D. Wang et al. / The Journal of Arthroplasty xxx (2016) 1e9

Fig. 1. Radiographs of a 45-year-old woman with unilateral Crowe type IV developmental dysplasia of the hip. (A) Preoperative anteroposterior view. (B) Postoperative radiographic image. Hip was reconstructed at the level of the anatomic hip center by total hip arthroplasty combined with transverse osteotomy, and cerclage wires were placed at the osteotomy site to protect the femur from intraoperative fracture. Reconstruction of the acetabulum with screws was performed. (C) Postoperative radiographic image at 7-month follow-up. (D) At 4-year follow-up, bone union was detected at the osteotomy site, and no radiolucent lines around the femoral stem were found. (E) At 11-year follow-up, no subsidence or loosening was identified, and the femoral and acetabular components were considered to be stable.

healed without further sequelae. Three patients (3 hips) experienced postoperative dislocation. In these patients, dislocation occurred once after surgery and was treated with closed reduction, and confinement to bed for 3 weeks; the dislocation did not recur. Distal femur perforation was demonstrated in one hip due to the narrow and irregular femoral canal, which was successfully treated with a longer stem. Two patients (2 hips) had transient sciatic nerve palsy with weak dorsiflexion of ankle and numbness of foot. The 2 hips had leg lengthening of 3.8 and 3.7 cm, respectively. The 2 patients with nerve palsy were treated with medication and recovered within 8 months without any sensory or motor deficit. Nonunion of femoral osteotomy site was seen in the previous revised hip due to loosening of the femoral component and was treated simultaneously with bone grafting at the nonunion site. In this case, the osteotomy site archived solid union uneventfully. No evidence of deep venous thrombosis, infection, and polyethylene wear was identified during follow-up evaluation.

nonunion, nerve injury, and thigh pain) and early loosening of stem components [1,9-11,13,14,18,19,29-39] (Table 2). However, most prior reports of subtrochanteric shortening osteotomy included mixed groups of different types of components or different osteotomy techniques in a small number of patients [12,21-23]. To our knowledge, this is the largest reported series in which the clinical and radiological results of exclusively THA combined with transverse subtrochanteric femoral osteotomy for the treatment of Crowe type IV DDH were evaluated, including (1) intraoperative and postoperative short-term complications; (2) clinical outcomes; and (3) survivorship of acetabular and femoral prostheses. The rate of intraoperative femoral fracture during insertion of the femoral stem has been reported to vary between 5.2% and 26.8% for cementless THA in Crowe type IV DDH [1,40,41]. In our series, restoration of the anatomical hip center was achieved in all

Survivorship Analysis At 10 years after surgery, Kaplan-Meier survivorship with revision for any reason as end point was 97% (95% confidence interval 84-99). Kaplan-Meier survivorship with revision for radiographic loosening for any component was 97% (95% confidence interval 8499) at 10 years (Fig. 2). Discussion THA is challenging in such patients with Crowe type IVHartofilakidis type III DDH. It is difficult to achieve and maintain the reduction of the femoral head into the true acetabulum due to severe soft tissue contracture, abnormality in the neurovascular structures, and leg length discrepancies. Subtrochanteric femoral shortening osteotomy is an effective approach, which makes restoration of hip rotation center easier and ensures femoral deformity correction but may be associated with complications (eg,

Fig. 2. Kaplan-Meier survivorship curve with radiographic loosening for any component as the end points.

Table 2 Overview of Relevant Literature in the Treatment of Crowe IV Dysplasia Combined With Subtrochanteric Femoral Shortening Osteotomy. Study

Patient DDH Subtrochanteric Mean Age (Hips) Type Osteotomy (y), Range

Shang et al [35]

15 (17) IV

Transverse

Ollivier et al [34]

24 (28) IV

Transverse

Mu et al 58 (71) IV [1]

Mean Follow- Stem Type Up (y), Range

30.9 (22-58) 2.75 (0.5-5)

48 (30-72)

Cementless (S-ROM)

10 (0.8-14.5) Cementless

Acetabular Type

Postoperative Preoperative LLD Function Score

Cementless <1.5 cm

Cementless 6 (0-20) mm

Postoperative Function Score

HHS: 34.0 ± 6.5; pain: 15.7 ± 6.5; function: 15.0 ± 3.8; deformity: 1.4 ± 0.7; ROM: 1.4 ± 0.7 HHS: 43 (24-89)

HHS: 85.0 ± 7.3; pain: 39.0 ± 5.6; function: 37.2 ± 9.8; deformity: 3.2 ± 0.5; ROM: 3.5 ± 0.6 HHS: 87 (56-93)

HHS: 82.9 (61-98)

38.5 (16-56) 5.88 (3.3-8.2)

Cementless Cementless 1.01 (0-3) cm HHS: 35.6 (28-51) (Zweymüller)

25 (25) IV

Transverse

30.5 (22-39) 10

Cementless HA-coated stem

Zhu et al 20 (21) IV [19]

Transverse

36 (21-77) 3.33 (2-5)

Nonmodular Cementless 1.2 ± 0.4 cm cementless conical stem

HHS: 52.4 ± 6.8

HHS: 90.5 ± 15.1

Hua et al 21 (24) IV [30]

Transverse

55 (36-71)

3.5 (1.5-9)

Cemented or Cementless 0.5 (0-3) cm uncemented

HHS: 47.5 ± 8.7

HHS: 88.5 ± 3.1

9 (12) IV Oinuma et al [13] Sun et al 30 (36) IV [37]

Transverse

61.5 (46-73)

3.7 (1.5-6.3)

Cementless (S-ROM)

Cementless <1 cm

MAP: 9.2 (7-13)

MAP: 17 (16-18)

Transverse

46.8 (38-65)

4 (0.58-7)

Cementless (S-ROM)

Cementless 0.9 (0-2.7)

HHS: 41.7 ± 3.7

HHS: 89.1 ± 2.9

Sofu et al 74 (87) IV [36]

Transverse

46.8 (20-75)

4.8 (2-7)

Cementless 2.7-6 cm Porouscoated femoral stem

HHS: 41.8 (35-51)

HHS: 86.2 (57-94)

Oe et al [33]

Transverse

64.9 (35-80)

5.2 (3-10)

Transverse

25 (20-35)

All polyethylene acetabular components Cementless (S-ROM)

Imam et al [31]

Li et al [32]

26 (34) IV

8 (8)

IV

3 (1-5)

Cementless 4.02 (3-5) cm HHS: 38.4; OHS: 17 HHS: 88.6; OHS: 38.6

HS32 N narrow stem (Cemented) Cementless

<1 cm: 16; 1- JOA: 50.2 (16-74) 2 cm: 10; >2 cm: 0

JOA: 84.6 (62-97)

0.4 (0.6-1) cm HHS: 53.9 ± 6.6

HHS: 87.6 ± 0.3

0%

7.10%

2.80%

0%

4.70%

0%

18%

0%

0%

4.70% Fem AL 4.7% 0%

0%

0%

0%

0%

0%

0%

0%

14%

0% (bony ingrowth 17; fibrous ingrowth 18) 0%

2.90%

0%

0%

0%

Survival (%)

SNI 2

100%

89% for free of revision for AL; 82% for free of revision for any reason Intraop FF 20; PW 91.40% for free of revision for 4; dislocation 1; FNI 6; SNI 2; cup any reason breakage 1 100% Heterotopic ossification 5; dislocation 2; DVT 3; PE 1; superficial infection 3 SNI 3; delayed 95.2 for free of union 1 revision for subsidence or head motion Trochanteric 100% fracture 3; SNI 1

Intraop FF 5; Acet reoperation 1; AL 3.5%; dislocation 3 fem AL 10.7%

9.80% Acet AL 1.4% 0%

Complication

Severe limp 4; dislocation 1

100% for free of revision for any reason 100% for free of revision for any reason 87.21% for free of revision for any reason

Severe limp 3; periprosthetic fracture 2; SNI 1 Dislocation 6; Acet neurological AL 4.59%; complication 2; fem AL pseudoarthrosis 2; 4.59% heterotopic ossification 1 0% Radiolucency 1 100% for free of revision for any reason 0%

DVT 2; SNI 1

D. Wang et al. / The Journal of Arthroplasty xxx (2016) 1e9

Transverse/no osteotomy

Nonunion Revision AL Rate Rate Rate

100% for free of revision for any reason

(continued on next page)

5

Study

6

Table 2 (continued ) Patient DDH Subtrochanteric Mean Age (Hips) Type Osteotomy (y), Range Transverse

41.6 ± 8.19

Zhong et al [39] Takao et al [9]

28 (36) IV

Transverse

45.2 (36-56)

25 (33) IV

Step cut

Kawai et al [10] Charity et al [11] Akiyama et al [29] Yalcin et al [38]

12 (19) IV

Transverse

15 (18) IV

Transverse

11 (15) IV

Duan et al (This study)

60 (45-76)

(3-8)

Porouscoated femoral components

4.4 (2.3-7.83) Cementless (S-ROM) 8 (5-11)

64.8 (57-73) 3.17 (0.5-8)

Acetabular Type

Postoperative Preoperative LLD Function Score

Postoperative Function Score

Nonunion Revision AL Rate Rate Rate

Cementless NA

HHS: 36.2 ± 9.8

HHS: 90.8 ± 2.5

0%

Cementless 6 mm

HHS: 39

HHS: 87

0%

Cementless S-ROM or cemented acetabular cup

1.9 (0-4.1) cm MAP: 9 (3-14)

MAP: 16 (13-18)

0%

Cemented

3.4 (1.5-4.5) cm

JOA: 45.1 (27-58)

JOA: 79.6 (72-90)

0%

Pain: 2.4 (1-4); function: 2.3 (1-4); walk: 3.4 (1-6) MAP: 8.1 ± 2.5

5.56% Pain: 5.2 (3-6); function: 4.4 (3-6); walk: 5.2 (4- 6) MAP: 15.1 ± 1.3 20%

Cemented

Transverse

9.5 (4.33-14) Cemented Exeter femoral stem 58.9 (42-77) 6.25 (2.75-9.7) Cemented

Cementless NA or cemented Cemented 0.33 (0-1.4) cm

31 (44) IV

Transverse

43.2 (22-63) 5.17 (2-8)

Cementless 2.5-3.5 cm: 6; HHS: 36.2 (31-41) <2 cm: 25

IV

Transverse

46.5 (19-73)

51 (33-75)

Cementless

10 (6.6-13.2) Cementless (S-ROM)

Cementless 1.0 (0.6-1.7)

HHS: 81.2 (65-93)

HHS: 38.8 (21-59); HHS: 86.1 (76-98); MAP: 6.7 (3-10) MAP: 15.9 (12-17)

4.80% Fem AL 14.3%

0%

3.03%

0%

22.22%

0%

11.40%

6.80%

1.32%

2.62%

0%

Complication

Survival (%)

Trendelenburg sign 7; thigh pain 2; radiolucency 2; intraop FF 5; dislocation 2; SNI 1 Periprosthetic fracture 2

95.2% for free of revision for any reason

100% for free of revision for any reason 97% for free of Trendelenburg Fem sign 2; intraop FF stem revision; AL 97% for free of 3.03% 8; dislocation 2; stem aseptic radiolucency 1 loosening 0% Heterotopic 100% for free ossification 2 of revision for any reason Trendelenburg 77.8% for free Acet sign 12 of revision for AL any reason 16.7% 0% Dislocation 2 100% for free of revision for any reason 93.2% for free Trendelenburg Acet of revision for sign 16; AL any reason dislocation 2; 6.8% superficial infection 2 97% for free of Dislocation 3; Acet intraop FF 4; SNI revision for AL 1.32%; 2; Trendelenburg any reason fem AL sign 1 1.32%

intraop AF, intraoperative acetabular fracture; intraop FF, intraoperative femoral fracture; acet AL, acetabulum aseptic loosening; fem AL, femoral component aseptic loosening; PW, polyethylene wear; FNI, femoral nerve injury; SNI, sciatic nerve injury; OHS, Oxford Hip Score; DVT, deep vein thrombosis; PE, pulmonary embolism; MAP, Merle d'Aubigne and Postel; JOA, Japanese Orthopaedic Association Hip Score; HHS, Harris Hip Score; ROM, range of motion; HA, hydroxyapatite coating; NA, not applicable.

D. Wang et al. / The Journal of Arthroplasty xxx (2016) 1e9

Baz et al 15 (21) IV [18]

Mean Follow- Stem Type Up (y), Range

D. Wang et al. / The Journal of Arthroplasty xxx (2016) 1e9

hips. We have identified 4 hips (5.3%) with perioperative femoral fractures because of tight press-fit of the prosthesis into the straight and small medullary canal to achieve rigidly rotational stability, and thereby hoop strains in medial and anteromedial femoral cortex. All fractures were stabilized with a plate and cable system to prevent their propagation and healed without loosening, subsidence, and nonunion. Ollivier et al [34] retrospectively reviewed 28 consecutive cementless THAs in Crowe IV DDH, and intraoperative fracture occurred in 5 of 28 hips (17.8%). However, these fractures were managed properly with no subsequent sequelae or new problems. Although successful management of fractures did not affect the clinical outcome and survival of the implant, a cerclage band was prophylactically placed before implant insertion. In addition, these femoral fracture signs, such as sudden subsidence of the implant during broaching and rotational instability of the stem at final insertion, should be cautious. We selected S-ROM stem (DePuy) for these reconstructions, which was widely used with more satisfactory outcomes and showed relatively low intraoperative fracture rate ranging from 1.4%-10.7% [42,43]. The nonunion at the sites of subtrochanteric femoral osteotomies is a common complication. The previous studies have reported nonunion rates ranging from 2.8%-7.1% at the osteotomy sites in Crowe IV hips [1,19,34]. Various osteotomy techniques, including transverse [1,13,19,30,31,34], v-shaped [44], and step-cut [5,20], have been reported. Muratli et al [15] performed a biomechanical study comparing 4 different subtrochanteric osteotomy groups, and demonstrated that there was no single inherent feature increasing the stability of the osteotomy designs. A comprehensive meta-analysis by Li et al [45] suggested that no significant difference between transverse and modified method regarding complications and survivorship was identified. And, transverse osteotomy may be recommended because of technical simplicity in adjusting the anteversion angle and minimal damage of the periosteum at the osteotomy site [45,46]. In addition, transverse osteotomy has several advantages compared with other modified methods, including preservation of the proximal femoral metaphysis, simultaneous shortening and correction of rotational abnormalities, which may decrease the risk of torsional instability for the stem and provide more normal proximal femoral anatomy. Therefore, transverse osteotomy was utilized in our study, and the nonunion rate of 1.3% in our series was parallel with the previous reports [1,19,34]. The mean time to union at osteotomy site was 6 months that was similar to the previous studies [1,21,34]. Surgeons should be aware of several risk factors related to bone nonunion, and take correspondent measures to prevent this complication. First, the interface and canal diameters were congruent between proximal and distal segments, which may affect the union. Bone resection should be assisted by assistance with axial resistance at the knee to achieve a better anastomosis interface, and the 2 osteotomy interfaces should be cut smoother to maintain the interfaces intact as much as possible. Second, when preparing osteotomy sites and prophylactical cerclage wiring, we removed circumferential periosteum and thereby damaged the osteoblastic activity of the periosteum. Therefore, the osteotomy site should be held with autologous cancellous and then stabilized with autogenous cortical bone plate affixed with cables or wires. In our study, we consider dislocation as a common postoperative complication. The press-fit between the femoral stem and distal femoral fragment may compromise closed reduction for treatment of dislocation, thereby resulting in bone nonunion and stem loosening [14,19]. Previous studies have reported dislocation rates of 3.8%-10.7% in patients with Crowe type IV DDH [19,34]. In our series, the dislocation rate was 3.9%, which was comparable to previous studies. All dislocations were treated successfully with

7

closed reduction and prophylactically fixed with plaster bandage to prevent recurrent dislocation. No stem loosening and nonunion was seen in the patients who had a dislocation (Fig. 3). Although femoral shortening osteotomy was performed, the anatomic placement of acetabular components may also lead to excessive stretching and nerve injury in severe DDH [1,19,31,34]. There was no evidence regarding maximum amount as indications to perform a femoral osteotomy without a neurological complication. However, the study by Eggli et al [47] demonstrated that the incidence of nerve damage was not associated with the amount of limb lengthening, which was most commonly caused by direct or indirect mechanical trauma. And, several risk factors, including postoperative dislocation, operative time, a posterior surgical approach, and retractor-induced injury, may also increase the odds for nerve palsy. Nerve injury rates ranging from 5%-11.3% were reported in other series [1,48]. The transient nerve injuries in 2 hips occurred in our series because of the relatively large amount of leg lengthening and less experience and time-consumption at the beginning. All nerve injuries were healed without any further complication by the latest follow-up evaluation. We believe that the leg lengthening of >3.5 cm may increase the risk of incidence of nerve injury. Transverse subtrochanteric shortening osteotomy with appropriate amount was valuable to avoid nerve injury for the patients with severe dysplasia of the hip. Moreover, lumbar scoliosis and inclination of the pelvis and severe limp should be taken into consideration to prevent undercorrection of limb length. Significant decrease in the limb length discrepancies was identified in our series. Sixteen patients (21.1%) had slight limping postoperatively at the latest follow-up, which may be attributed to long-term abductor muscle weakness and gait abnormality. The long-term functional outcomes improved significantly in 74 hips of our study group, which was parallel with the previous report in the literature [10,14,19,34]. The improved hip function and quality of life resulted from postoperative pain relief, decreased limb length, and especially the restoration of femoral offset, which could reduce forces of joint reaction and simultaneously restore lever arm of abductor muscles [19]. The placement of acetabular components in an anatomic position promoted long-term stability of acetabular and femoral components. All acetabular cups were inserted in the true acetabulum in our series, which provided adequate coverage and restored normal rotation center of hip and abductor muscle function [6]. Watts et al [8] reported that the placement of acetabular cups in true acetabulum decreased the risk of acetabular loosening and revision compared with the placement at the nonanatomic hip center at a mean of 36 years in 60 dysplasia hips. In our study, all acetabular components were implanted in true acetabulum, and 1 acetabular cup loosening was identified due to acetabular fracture and thereby screw breakage. Hasegawa et al [49] reported 1 femoral stem aseptic loosening in 18 patients (20 hips) with Crowe type IV DDH. Takao et al [9] also observed 1 femoral stem revision due to progressive stem subsidence and loosening in their series of cementless THA combined with subtrochanteric transverse osteotomy for Crowe type IV DDH. Similarly, we observe 1 femoral stem loosening at the latest follow-up, which may be due to short femoral neck, poor bone quality, and thereby insufficient rotatory stability. As for modular femoral components, some risk factors, such as fretting, corrosion, and mechanical failure, may affect the stability of the stem and result in osteolysis and loosening. In our consecutive series, only 2 hips had osteolysis in greater trochanter without further loosening. During the stem revision surgery, no gross metallic debris or blackening of periprosthetic tissue was identified. There are several advantages to using modular femoral components rather than nonmodular femoral components. Its anteversion angle is adjustable to restore the normal anteversion.

8

D. Wang et al. / The Journal of Arthroplasty xxx (2016) 1e9

Fig. 3. Radiographs of a 50-year-old woman with bilateral Crowe type IV developmental dysplasia of the hip. (A-C) Preoperative anteroposterior view. (D-F) Postoperative radiographic image at 1-week follow-up. Total hip arthroplasty was performed combined with transverse osteotomy in the right hip. Fixation was conducted with plate and screw. (G-H) Postoperative anteroposterior view showed dislocation at 2-week follow-up, which was treated successfully through closed relocation. No dislocation recurred. (I-K) Postoperative anteroposterior view showed total hip arthroplasty combined with transverse subtrochanteric shortening osteotomy in the left hip. (L-P) At 5-year follow-up, osseous healing was completed at the osteotomy site, and no radiolucent lines around the femoral and acetabular components were identified. (Q-U) At 10-year follow-up, the postoperative Harris Hip Score was 96. No loosening of components was identified, and the femoral stem was judged to be stable with bone ingrowth.

Moreover, the small-diameter straight femoral stem is applicable for the small and straight medullary cavity in patients with Crowe type IV DDH. In addition, it supplies different pairs of sleeve and femoral handle for treatment of different types of metaphysic femur mismatch. We note some limitations in this study. First, the study is a retrospective evaluation of prospectively followed patients without control group. Second, the number of patients in the study group was relatively small. However, our study group is a large series of patients with such uncommon diagnosis. Third, there was no sufficient data to analyze the results of THA combined with subtrochanteric shortening osteotomy between unilateral and bilateral cases. Fourth, all operations were performed by 5 senior surgeons, which may affect the validity of our findings. However, in this series, all femoral and acetabular implants and the surgical techniques of subtrochanteric femoral osteotomy and fixation were identical for all THAs. Conclusion Cementless, modular THA with placement of the cup at the anatomic hip center combined with transverse subtrochanteric femoral shortening osteotomy for treatment of Crowe type IV DDH led to satisfactory outcomes, including evident improvement in clinical function, high rate of stable fixation of cementless implant, restoration of more normal limb lengths, and low incidence of complications.

References 1. Mu W, Yang D, Xu B, et al. Midterm outcome of cementless total hip arthroplasty in Crowe IV-Hartofilakidis type III developmental dysplasia of the hip. J Arthroplasty 2016;31:668. 2. Sugano N, Noble PC, Kamaric E, et al. The morphology of the femur in developmental dysplasia of the hip. J Bone Joint Surg Br 1998;80:711. 3. Ito H, Tanino H, Yamanaka Y, et al. Hybrid total hip arthroplasty using specifically-designed stems for patients with developmental dysplasia of the hip. A minimum five-year follow-up study. Int Orthop 2011;35:1289. 4. Bao N, Meng J, Zhou L, et al. Lesser trochanteric osteotomy in total hip arthroplasty for treating CROWE type IV developmental dysplasia of hip. Int Orthop 2013;37:385. 5. Makita H, Inaba Y, Hirakawa K, et al. Results on total hip arthroplasties with femoral shortening for Crowe's group IV dislocated hips. J Arthroplasty 2007;22:32. 6. Bernasek TL, Haidukewych GJ, Gustke KA, et al. Total hip arthroplasty requiring subtrochanteric osteotomy for developmental hip dysplasia: 5- to 14-year results. J Arthroplasty 2007;22:145. 7. Anwar MM, Sugano N, Masuhara K, et al. Total hip arthroplasty in the neglected congenital dislocation of the hip. A five- to 14-year follow-up study. Clin Orthop Relat Res 1993:127. 8. Watts CD, Abdel MP, Hanssen AD, et al. Anatomic hip center decreases aseptic loosening rates after total hip arthroplasty with cement in patients with Crowe type-II dysplasia: a concise follow-up report at a mean of thirty-six years. J Bone Joint Surg Am 2016;98:910. 9. Takao M, Ohzono K, Nishii T, et al. Cementless modular total hip arthroplasty with subtrochanteric shortening osteotomy for hips with developmental dysplasia. J Bone Joint Surg Am 2011;93:548. 10. Kawai T, Tanaka C, Ikenaga M, et al. Cemented total hip arthroplasty with transverse subtrochanteric shortening osteotomy for Crowe group IV dislocated hip. J Arthroplasty 2011;26:229. 11. Charity JA, Tsiridis E, Sheeraz A, et al. Treatment of Crowe IV high hip dysplasia with total hip replacement using the Exeter stem and shortening derotational subtrochanteric osteotomy. J Bone Joint Surg Br 2011;93:34.

D. Wang et al. / The Journal of Arthroplasty xxx (2016) 1e9 12. Lai KA, Shen WJ, Huang LW, et al. Cementless total hip arthroplasty and limblength equalization in patients with unilateral Crowe type-IV hip dislocation. J Bone Joint Surg Am 2005;87:339. 13. Oinuma K, Tamaki T, Miura Y, et al. Total hip arthroplasty with subtrochanteric shortening osteotomy for Crowe grade 4 dysplasia using the direct anterior approach. J Arthroplasty 2014;29:626. 14. Krych AJ, Howard JL, Trousdale RT, et al. Total hip arthroplasty with shortening subtrochanteric osteotomy in Crowe type-IV developmental dysplasia. J Bone Joint Surg Am 2009;91:2213. 15. Muratli KS, Karatosun V, Uzun B, et al. Subtrochanteric shortening in total hip arthroplasty: biomechanical comparison of four techniques. J Arthroplasty 2014;29:836. 16. Kilicoglu OI, Turker M, Akgul T, et al. Cementless total hip arthroplasty with modified oblique femoral shortening osteotomy in Crowe type IV congenital hip dislocation. J Arthroplasty 2013;28:117. 17. Dallari D, Pignatti G, Stagni C, et al. Total hip arthroplasty with shortening osteotomy in congenital major hip dislocation sequelae. Orthopedics 2011;34:e328. 18. Baz AB, Senol V, Akalin S, et al. Treatment of high hip dislocation with a cementless stem combined with a shortening osteotomy. Arch Orthop Trauma Surg 2012;132:1481. 19. Zhu J, Shen C, Chen X, et al. Total hip arthroplasty with a non-modular conical stem and transverse subtrochanteric osteotomy in treatment of high dislocated hips. J Arthroplasty 2015;30:611. 20. Erdemli B, Yilmaz C, Atalar H, et al. Total hip arthroplasty in developmental high dislocation of the hip. J Arthroplasty 2005;20:1021. 21. Kawai T, Tanaka C, Kanoe H. Total hip arthroplasty for Crowe IV hip without subtrochanteric shortening osteotomyda long term follow up study. BMC Musculoskelet Disord 2014;15:72. 22. Imbuldeniya AM, Walter WL, Zicat BA, et al. Cementless total hip replacement without femoral osteotomy in patients with severe developmental dysplasia of the hip: minimum 15-year clinical and radiological results. Bone Joint J 2014;96-B:1449. 23. Masonis JL, Patel JV, Miu A, et al. Subtrochanteric shortening and derotational osteotomy in primary total hip arthroplasty for patients with severe hip dysplasia: 5-year follow-up. J Arthroplasty 2003;18:68. 24. Gruen TA, McNeice GM, Amstutz HC. “Modes of failure” of cemented stem-type femoral components: a radiographic analysis of loosening. Clin Orthop Relat Res 1979:17. 25. DeLee JG, Charnley J. Radiological demarcation of cemented sockets in total hip replacement. Clin Orthop Relat Res 1976:20. 26. Engh CA, Massin P, Suthers KE. Roentgenographic assessment of the biologic fixation of porous-surfaced femoral components. Clin Orthop Relat Res 1990:107. 27. Heekin RD, Engh CA, Herzwurm PJ. Fractures through cystic lesions of the greater trochanter. A cause of late pain after cementless total hip arthroplasty. J Arthroplasty 1996;11:757. 28. Brooker AF, Bowerman JW, Robinson RA, et al. Ectopic ossification following total hip replacement. Incidence and a method of classification. J Bone Joint Surg Am 1973;55:1629. 29. Akiyama H, Kawanabe K, Yamamoto K, et al. Cemented total hip arthroplasty with subtrochanteric femoral shortening transverse osteotomy for severely dislocated hips: outcome with a 3- to 10-year follow-up period. J Orthop Sci 2011;16:270. 30. Hua WB, Yang SH, Xu WH, et al. Total hip arthroplasty with subtrochanteric femoral shortening osteotomy for high hip dislocation. Orthop Surg 2015;7:112.

9

31. Imam MA, Fathalla I, Holton J, et al. Cementless total hip replacement for the management of severe developmental dysplasia of the hip in the middle eastern population: a prospective analysis. Front Surg 2016;3:31. 32. Li W, Zhang W, Bai G, et al. Total hip arthroplasty for treatment of Crowe type IV congenital dysplasia of hip with dislocation in adults. Zhongguo Xiu Fu Chong Jian Wai Ke Za Zhi 2013;27:1153. 33. Oe K, Iida H, Nakamura T, et al. Subtrochanteric shortening osteotomy combined with cemented total hip arthroplasty for Crowe group IV hips. Arch Orthop Trauma Surg 2013;133:1763. 34. Ollivier M, Abdel MP, Krych AJ, et al. Long-term results of total hip arthroplasty with shortening subtrochanteric osteotomy in Crowe IV developmental dysplasia. J Arthroplasty 2016;31:1756. 35. Shang DC, Zhong SC, Zhang XZ, et al. Surgical treatment for osteoarthritis secondary to Crowe type IV developmental dysplasia of hip in adults. Zhongguo Gu Shang 2016;29:125. 36. Sofu H, Ahin VS, Gursu S, et al. Cementless total hip arthroplasty in patients with Crowe type-4 developmental dysplasia. Hip Int 2013;23:472. 37. Sun QC, Wang XH, Song BS, et al. Total hip arthroplasty for Crowe type IV developmental dysplasia of the hip with S-ROM prosthesis. Zhongguo Gu Shang 2013;26:153. 38. Yalcin N, Kilicarslan K, Karatas F, et al. Cementless total hip arthroplasty with subtrochanteric transverse shortening osteotomy for severely dysplastic or dislocated hips. Hip Int 2010;20:87. 39. Zhong C, Cai XZ, Yan SG, et al. S-ROM modular arthroplasty combined with transverse subtrochanteric shortening for Crowe type IV congenital dislocation of hip. Chin Med J (Engl) 2011;124:3891. 40. Eskelinen A, Helenius I, Remes V, et al. Cementless total hip arthroplasty in patients with high congenital hip dislocation. J Bone Joint Surg Am 2006;88:80. 41. Kim YH, Seo HS, Kim JS. Outcomes after THA in patients with high hip dislocation after childhood sepsis. Clin Orthop Relat Res 2009;467:2371. 42. Biant LC, Bruce WJ, Assini JB, et al. Primary total hip arthroplasty in severe developmental dysplasia of the hip. Ten-year results using a cementless modular stem. J Arthroplasty 2009;24:27. 43. Tamegai H, Otani T, Fujii H, et al. A modified S-ROM stem in primary total hip arthroplasty for developmental dysplasia of the hip. J Arthroplasty 2013;28: 1741. 44. Tozun IR, Akgul T, Sensoy V, et al. The results of monoblock stem with step-cut femoral shortening osteotomy for developmentally dislocated hips. Hip Int 2016;26:270. 45. Li C, Zhang C, Zhang M, et al. Comparison of transverse and modified subtrochanteric femoral shortening osteotomy in total hip arthroplasty for developmental dysplasia of hip: a meta-analysis. BMC Musculoskelet Disord 2014;15:331. 46. Ogawa H, Ito Y, Shinozaki M, et al. Subtrochanteric transverse shortening osteotomy in cementless total hip arthroplasty achieved using a modular stem. Orthopedics 2011;34:170. 47. Eggli S, Hankemayer S, Muller ME. Nerve palsy after leg lengthening in total replacement arthroplasty for developmental dysplasia of the hip. J Bone Joint Surg Br 1999;81:843. 48. Reikeras O, Haaland JE, Lereim P. Femoral shortening in total hip arthroplasty for high developmental dysplasia of the hip. Clin Orthop Relat Res 2010;468: 1949. 49. Hasegawa Y, Iwase T, Kanoh T, et al. Total hip arthroplasty for Crowe type developmental dysplasia. J Arthroplasty 2012;27:1629.