This retrospective study was approved by the Institutional Review Board of West China Hospital (ID: 2012-268) and was reported in accordance with the STROCSS criteria . The inclusion criteria were patients with severe pain and/or considerable difficulty in performing daily activities refractory to nonoperative management who underwent ipsilateral THA and TKA for rheumatoid arthritis (RA) or osteoarthritis (OA). The exclusion criteria were hip dysplasia, acetabulum or femoral fracture, ankylosing spondylitis, prior lower extremity fracture, posttraumatic arthritis, revision THA or TKA, and incomplete clinical or radiographic records. Additionally, to avoid bias related to prior TJA, patients with an interval time between ipsilateral THA and TKA shorter than 6 months and patients who underwent prior TJA with a functional score at the time of subsequent surgery less than 70 points were also excluded. We identified 88 patients (97 hips and 97 knees) who underwent ipsilateral THA and TKA at our institution between January 2008 and April 2014. Of these, two patients (2 hips and 2 knees) were lost to follow-up and could not be contacted via telephone, while three patients (4 hips and 4 knees) died from diseases unrelated to the operation. According to different sequences of ipsilateral THA and TKA, the remaining 83 patients (91 hips and 91 knees), including 47 patients undergoing TKA followed by ipsilateral THA and 36 as THA subsequent to ipsilateral TKA, were classified into 2 groups. To minimize possible confounding factors, the two groups were statistically matched for age (up to ±10 years), sex, cause of TJA, primary surgery data (up to ±12 months), and THA prosthesis. Because all implants of primary TKA in our center were posterior-stabilized, we did not match the prosthesis of TKA. Ultimately, twenty-eight patients were selected for each group and included in the final analysis.
All THA procedures were performed using a posterolateral approach under general anesthesia. The cementless porous-coat acetabular fixations (Pinnacle implants, DePuy Orthopaedics) were routinely press-fitted into the acetabulum at 15 ± 10° of anteversion and 40 ± 10° of inclination. If necessary, supplemental screws were used to achieve implant stability. Two cementless femoral stems, Corail and Summit (DePuy Synthes, Warsaw, IN), were inserted into the hips. Of these, ceramic-on-ceramic articulations were utilized in 42 hips (75.0%), and ceramic-on-polyethylene was used in 14 hips (25.0%).
For the ipsilateral TKA procedure, the knee was exposed by a standard medial parapatellar approach, and osteophytes, worn meniscus, and posterior cruciate ligament of the knee were resected. After determining the entry point, we inserted a femoral intramedullary alignment rod in the center of the femoral intercondylar notch with the distal femoral cutting guide set for individual degrees measured before surgery. Of note, one knee in the THA-TKA group and 2 knees in the TKA-THA group had a preoperative anatomic valgus of >20° that we released the iliotibial band (ITB) using the “pie-crusting” technique . If the lateral tension on extension and flexion was still tight, the posterlateral capsule was also released, avoiding the collateral ligament (LCL) and popliteus tendon (POP). After assessing the extension gap and balancing the flexion gap, we performed cemented, posterior-stabilized TKAs on all subjects with two total knee implants (DePuy Sigma PFC and Stryker Scorpio NRG). All patients received intraoperative and postoperative prophylactic broad-spectrum antibiotics and tranexamic acid antithrombotic therapy. Postoperatively, active flexion-extension ankle motion and quadriceps strengthening exercises were encouraged. Partial weight-bearing with crutches as tolerated on the second postoperative day and full weight-bearing were allowed from the third day. For patients with THA, simultaneous flexion and internal rotation were avoided after surgery.
Clinical evaluations and radiographic assessments
Clinical follow-up was conducted routinely at 3 weeks, 3 months and 6 months after the procedures and annually thereafter until the final follow-up. The clinical evaluation protocol included the Harris Hip Score (HHS) , Knee Society Score (KSS) , and Short Form-12 scale (SF-12) . Radiographs (serial standardized anteroposterior and lateral radiographs of hip and knee, and full-length weight-bearing anteroposterior films) were taken for patients preoperatively and performed at each follow-up. To minimize the variability of the interobserver, all radiographic measurements were performed independently and averaged by 2 trained investigators in the index surgery. To describe the coronal extremity axis, the hip-knee-ankle (HKA) angle was defined as the medial angle between the mechanical axis of the femur and the mechanical axis of the tibia. The femorotibial angle (FTA) was defined as the lateral angle between the anatomical axis of the femur and the anatomical axis of the tibia. When assessing coronal alignment on full-length films, an HKA less than 177° was considered varus, neutral between 177° and 183°, and valgus greater than 183°. Of note, severe valgus knee alignment was defined as an FTA of < 160° . The femoral offset (FO) was defined as the perpendicular distance from the centerline of the proximal part of the femoral shaft to the center of the femoral head. The inclination angle was defined as the angle formed by the inter-teardrop line and the long axis of the cup opening ellipse. The limb length discrepancy (LLD) was calculated by comparing the vertical distance from the prominent points of the lesser trochanter to the teardrop line. If the deformation of the lesser trochanter was severe, we used the tip of the greater trochanter as the reference. The correction in LLD was defined as the height difference in the preoperative and postoperative LLD  (Fig. 1).
Serial radiographs were also evaluated for evidence of postoperative periprosthetic fracture (PFF), dislocation, subsidence, and femoral component stability. Subsidence of the femoral component was defined as the change in the distance from the center of the femoral head to the lesser trochanter by the method of Heekin et al . The femoral component stability was evaluated and graded as bone stable, fibrous stable, or unstable, according to the criteria described by Engh et al . Complications included periprosthetic infection, deep venous thrombosis, and neurologic injury. Kaplan-Meier survivorship analysis was performed on all THAs and TKAs using a standard case scenario where all arthroplasties were considered to be successful at the final follow-up. Prosthesis failure was defined as any reason for aseptic revision.
Groups of THA-TKA and TKA-THA were matched one to one using case-control matching, and statistical analyses were performed using SPSS statistical software, version 25.0 (IBM Corp., Armonk, NY). The figures were generated by GraphPad Prism version 8.0. The students’ t-test or Mann-Whitney U-test was used to analyze continuous variables between two groups. Pearson’s chi-squared test or Fisher’s exact test was used for discontinuous variables between two groups. Implant survivorships were analyzed by Kaplan-Meier curves with revision for any reason other than infection as the endpoint. The survival rate between the 2 groups was compared by the log-rank test. α= .05, P < .05 indicated statistical significance.