Ankylosing spondylitis is a chronic and progressive inflammatory disease of unknown etiology, characteristically involving the sacroiliac joints, spine, and hip.1 Ankylosing spondylitis affects the hip in more than 20% of cases, leading to severe deformities, functional impairment, and decreased quality of life.2,3 Progressive spinal deformity in ankylosing spondylitis is more prevalent among those with hip arthritis compared with those without hip involvement.4
Little data exist regarding the effectiveness of treatment strategies specifically for ankylosing spondylitis patients with end-stage refractory hip pain. Total hip arthroplasty (THA) is considered effective for relieving pain, improving mobility, restoring function, and improving quality of life.5,6 However, contracture of the hip joint is inevitable during the long-term course of ankylosing spondylitis, making THA more technically difficult.
There are also undesired complications associated with the traditional THA procedure, including incomplete correction of joint deformities, heterotopic ossification, and damage of the femoral nerve.7 Thus, the optimal treatment for patients with ankylosing spondylitis and end-stage hip involvement remains a clinical challenge. Improving joint function should be the primary goal of surgery.
To improve joint function with minimal complications for patients with ankylosing spondylitis and hip involvement, the authors' department has developed a 2-stage surgical treatment. Stage I consists of soft tissue release of the hip joint, femoral osteotomy, and supracondylar bone traction. Stage II is THA. This retrospective study evaluated the outcomes of this phased treatment relative to traditional THA alone in patients with ankylosing spondylitis and severe hip flexion contracture.
Materials and Methods
Patients and Clinical Data
This retrospective study received institutional review board approval. A review was conducted of a consecutive cohort of patients with ankylosing spondylitis who underwent THA in the authors' department between 2011 and 2017. All of the patients met the following criteria: received a diagnosis of ankylosing spondylitis that was based on the 1984 modified New York criteria,8 had severe hip involvement, and had complete radiographic and clinical data. Patients with any of the following were excluded: history of revision arthroplasty or other spine or lower-limb surgery. Patients who met the inclusion criteria received either the 2-stage surgery (experimental group) or THA only (control group). The experimental group consisted of 12 patients (16 hips). The control group consisted of 12 age-matched patients (15 hips).
All of the surgeries were performed by the same senior surgeon (Y.Z.), and all of the patients received epidural anesthesia. For patients with bilateral involvement, the hip with the more severe deformity was included in this analysis.
Stage I: Joint Soft Tissue Release, Femoral Osteotomy, and Supracondylar Bone Traction. With the patient in the lateral decubitus position, the iliopsoas tendon on the lesser trochanter and the reflected head of the rectus femoris were exposed and stripped via the lateral approach. Partial resection of the hip articular capsule was then performed, and the tensor fascia lata was partially incised, exposing the greater trochanter. A segment of the femoral neck was removed, conducive to correction of the joint contracture and to achieve 40° abduction.
Femoral supracondylar bone traction was performed after the surgery, gradually adjusting the hip flexion to avoid injury of the femoral nerve due to excessive stretching. Twenty degrees of hip flexion was maintained for 5 to 6 days (Figure 1).9,10
A representative patient with ankylosing spondylitis in the experimental group. Preoperative range of motion in hip flexion is 90° (A). Preoperative range of motion in hip extension is 60° (B). Range of motion of the hip is 40° after soft tissue release of the hip joint and femoral osteotomy (C). Supracondylar bone traction (D). Range of motion of the hip is 20° after supracondylar bone traction (E). Range of motion of the hip is 10° after total hip arthroplasty (F).
Stage II: THA. The interval between the 2 stages was approximately 2 weeks. Biological fixation implants (Duraloc; Johnson & Johnson, Jacksonville, Florida) were used as the fixation method for both the femoral stem and the acetabular cup in all cases. The patient was placed in the lateral decubitus position. Via the anterolateral lateral approach, the acetabulum and the femoral neck were exposed through the muscles compartment. Segments of the femoral neck and the femoral head were removed, and appropriate acetabular cups and femoral stems were implanted (Figure 2). Relevant data were recorded.6,11,12
Representative anteroposterior radiographs of patients in the experimental group. Preoperative (A), postoperative (B), and 6-month postoperative (C) radiographs of a 39-year-old man. Preoperative (D), postoperative (E), and 6-month postoperative (F) radiographs of a 45-year-old man.
Clinical data were collected before surgery and at the latest follow-up. Hip range of motion (ROM) was assessed (extension, flexion, and adduction). Patient-reported outcomes of quality of life were determined by visual analog scale (VAS, 0–10; 0=no pain) and Harris Hip Score (HHS, 0–100; 100=best function).13,14 Complications were recorded, including femoral nerve injury and heterotopic ossification.
All data were analyzed using SPSS version 20.0 software (IBM, Armonk, New York). For variables collected preand postoperatively, paired t tests were used to determine whether there was a significant change between time points. Differences in HHS, VAS score, and hip ROM between the experimental and the control groups were analyzed by Mann–Whitney U test. The chi-square test was used to analyze categorical data. Statistical significance was set at P<.05.
The mean ages at surgery of patients in the experimental and the control groups were 37.7±5.6 years (range, 24–51 years) and 35.4±2.1 years (range, 29–60 years), respectively. The durations of ankylosing spondylitis were 20.6±4.9 years (range, 7–30 years) and 23.3±5.2 years (range, 7–30 years), respectively.
Both the experimental and the control groups experienced significant corrections after surgery (Table 1). In the experimental group, the HHS increased from the preoperative 26.67±2.39 to 90.92±5.66 at the final follow-up (P=.031), while the VAS score decreased from 7.3±1.9 to 2.3±1.1 (P=.024). Range of motion of the hip in extension improved from the preoperative 30.5°±5.7° to postoperative 8.4°±2.6° and was 5.2°±3.6° at the latest follow-up (P=.037). Hip ROM in flexion improved from 80.9°±10.7° preoperatively to 103.8°±15.6° postoperatively and was well maintained at 94.7°±9.7° at the latest follow-up (P=.021).
Comparison of Clinical Outcomes Between the Experimental and the Control Groups
In the control group, the HHS increased from 24.59±3.12 preoperatively to 85.92±4.82 at the final follow-up (P=.047), while the VAS score decreased from 8.0±1.5 to 4.3±1.8 (P=.019; Table 1). The hip ROM in extension improved from 26.8°±6.6° preoperatively to 11.7°±3.1° postoperatively and was 15.2°±4.3° at the latest follow-up (P=.014). The hip ROM in flexion improved from 71.9°±8.5° preoperatively to 95.1°±11.7° postoperatively and was 64.3°±6.5° at the latest follow-up (P=.009). At the latest follow-up, compared with the control group, the experimental group had significantly higher HSS and ROM in extension.
No patient in the experimental group and 5 patients in the control group had postoperative femoral nerve injury (P=.014). Heterotopic ossification occurred in 5 patients (Brooker class III, 3; Brooker class II, 2) in the experimental group (41.7%) and in 2 patients (both Brooker class III) in the control group (16.7%) (P=.031).
This study described and assessed the feasibility of a novel 2-stage surgical treatment for patients with ankylosing spondylitis and severe hip involvement. In this treatment, THA is preceded by surgical soft tissue release of the hip joint, femoral osteotomy, and supracondylar bone traction. The new treatment (experimental) was evaluated relative to traditional THA alone (control) in an age-matched group of 12 patients. At the latest follow-up, the patients in the experimental group had a higher mean postoperative HSS and ROM in extension and a lower incidence of femoral nerve injury compared with the control group.
There are multiple choices of surgery for patients with ankylosing spondylitis, including femoral osteotomy, total hip surface replacement, and THA. Total hip arthroplasty has been the optimal choice. However, it remains a clinical challenge in patients with ankylosing spondylitis with severe hip arthrogryposis and joint ankylosis, in which fragile bones break easily. In addition, THA in these patients can result in a redundant procedure, malformation residue, and injury of neural function, which may compromise the surgical outcome and patient satisfaction.11,15,16
To achieve hip ROM at a maximum and decrease the risk of femoral nerve injury due to overstretch, the authors perform soft tissue release of the hip joint during the first stage of this method and then gradual correction of hip arthrogryposis through femoral osteotomy and femoral supracondylar bone traction. In this study, none of the patients who underwent this strategy experienced injury of neural function. However, at the 6-month follow-up, 5 patients (42%) who had received traditional THA only (control group) had experienced injury of neural function to varying degrees. Therefore, the authors conclude that the phased treatment was important for protecting the femoral nerve.
Further, the postoperative hip ROM of the experimental group was significantly larger than that of the control group. The VAS scores also showed that this 2-stage treatment satisfied the patients more and decreased pain intensity significantly. The greater hip ROM may have been due to the time interval between the 2 stages (approximately 2 weeks). This may also have been enough time for the nerve to adapt to the changed pathway, avoiding the pain caused by overstretch.
However, the incidence of heterotopic ossification was significantly greater in the experimental group than in the control group. This is a worrying result. Previous studies suggested that heterotopic ossification was a common complication after THA, but the pathogenesis remains unclear. Regarding the current study, the authors speculate that the higher rate of heterotopic ossification in the experimental group may have been because the incremental surgical procedures stimulated the stem cells, promoting their proliferation and differentiation. However, there are currently no efficacious preventive measures for this.
Based on the authors' observations, two major issues require attention in THA surgery. Most patients with ankylosing spondylitis have poor bone quality. Therefore, to prevent a fracture of the femur, a femur-matched prosthesis should be chosen, and the operation should be meticulous when expanding the marrow and implanting the femoral stem. Furthermore, patients with ankylosing spondylitis usually present with obvious kyphosis in the sagittal plane, which leads to a posture in which the trunk bends forward. Placement of an acetabular prosthesis according to conventional anteversion may lead to a risk of dislocation, so it is necessary to reduce the anteversion of the acetabular cup accordingly. In addition, to avoid undesirable contact between the stem and the cup, the anteversion of the acetabular prosthesis should be reduced and the anteversion of the femoral prosthesis increased.
The statistical power of this study may have been limited by its relatively small sample. However, this deficit may have been compensated for by the appropriately matched patient groups and the uniform surgical strategy. This was a retrospective study conducted within 1 year after surgery, which may have increased the potential for bias. The authors' results warrant a further longitudinal study with a larger population.
This 2-stage surgery for patients with ankylosing spondylitis and severe hip involvement, in which THA is preceded by soft tissue release of the hip joint, femoral osteotomy, and supracondylar bone traction, may significantly improve hip ROM while decreasing the incidence of femoral nerve injury, compared with THA alone. The 2-stage method was of greater benefit to the patients' quality of life.
- Braun J, Sieper J. Ankylosing spondylitis. Lancet. 2007;369(9570):1379–1390. doi:10.1016/S0140-6736(07)60635-7 [CrossRef]17448825
- Vander Cruyssen B, Muñoz-Gomariz E, Font P, et al. ASPECT-REGISPONSER-RESPONDIA Working Group. Hip involvement in ankylosing spondylitis: epidemiology and risk factors associated with hip replacement surgery. Rheumatology (Oxford). 2010;49(1):73–81. doi:10.1093/rheumatology/kep174 [CrossRef]
- Ward MM. Health-related quality of life in ankylosing spondylitis: a survey of 175 patients. Arthritis Care Res. 1999;12(4):247–255. doi:10.1002/1529-0131(199908)12:4<247::AID-ART3>3.0.CO;2-H [CrossRef]
- Marks JS, Hardinge K. Clinical and radiographic features of spondylitic hip disease. Ann Rheum Dis. 1979;38(4):332–336. doi:10.1136/ard.38.4.332 [CrossRef]496447
- Kim Y, Oh HC, Park JW, et al. Diagnosis and treatment of inflammatory joint disease. Hip Pelvis. 2017;29(4):211–222. doi:10.5371/hp.2017.29.4.211 [CrossRef]29250494
- He C, He X, Tong W, et al. The effect of total hip replacement on employment in patients with ankylosing spondylitis. Clin Rheumatol. 2016;35(12):2975–2981. doi:10.1007/s10067-016-3431-6 [CrossRef]27734230
- Blizzard DJ, Penrose CT, Sheets CZ, Seyler TM, Bolognesi MP, Brown CR. Ankylosing spondylitis increases perioperative and postoperative complications after total hip arthroplasty. J Arthroplasty. 2017;32(8):2474–2479. doi:10.1016/j.arth.2017.03.041 [CrossRef]28438449
- Calin A. Comment on article by van der Linden et al. Evaluation of diagnostic criteria for ankylosing spondylitis: a proposal for modification of the New York criteria. Arthritis Rheum. 1985;28(3):357–359. doi:10.1002/art.1780280321 [CrossRef]3977980
- van Bosse HJ, Saldana RE. Reorientational proximal femoral osteotomies for arthrogrypotic hip contractures. J Bone Joint Surg Am. 2017;99(1):55–64. doi:10.2106/JBJS.16.00304 [CrossRef]28060234
- Ding L, Gao YH, Li YR, Liu JG, Li SQ, Qi X. Determinants of satisfaction following total hip arthroplasty in patients with ankylosing spondylitis. Int Orthop. 2018;42(3):507–511. doi:10.1007/s00264-017-3589-9 [CrossRef]
- Feng DX, Zhang K, Zhang YM, et al. Bilaterally primary cementless total hip arthroplasty for severe hip ankylosis with ankylosing spondylitis. Orthop Surg. 2016;8(3):352–359. doi:10.1111/os.12254 [CrossRef]27627719
- Celiktas M, Kose O, Turan A, Guler F, Ors C, Togrul E. Conversion of hip fusion to total hip arthroplasty: clinical, radiological outcomes and complications in 40 hips. Arch Orthop Trauma Surg. 2017;137(1):119–127. doi:10.1007/s00402-016-2597-3 [CrossRef]
- Söderman P, Malchau H. Is the Harris hip score system useful to study the outcome of total hip replacement?Clin Orthop Relat Res. 2001;384:189–197. doi:10.1097/00003086-200103000-00022 [CrossRef]
- McCormack HM, Horne DJ, Sheather S. Clinical applications of visual analogue scales: a critical review. Psychol Med. 1988;18(4):1007–1019. doi:10.1017/S0033291700009934 [CrossRef]3078045
- Li J, Zhou Y, Jing J, Zhan J. Comparison of effects of two anticoagulants on hidden blood loss after total hip arthroplasty [in Chinese]. Zhongguo Xiu Fu Chong Jian Wai Ke Za Zhi. 2013;27(4):432–435.23757870
- Hu Y, Jiang WZ, Huang ZF, Torsha TT, Yang DS, Xiao J. Perioperative blood loss during total hip arthroplasty among patients with ankylosing spondylitis. Orthopedics. 2017;40(5):e904–e910. doi:10.3928/01477447-20170831-01 [CrossRef]28877325
Comparison of Clinical Outcomes Between the Experimental and the Control Groups
|Experimental Group||Control Group|
|Visual analog scale scorea|
|Harris Hip Scoreb|
|Range of motion in hip extension|
|Range of motion in hip flexion|