Transverse subtrochanteric shortening osteotomy with double tension-band fixation during THA for Crowe III-IV...

When performing total hip arthroplasty in patients with severe developmental dysplasia, shortening the femur facilitates reduction while also preventing sciatic or crural nerve injury and excessive length of the operated limb. No consensus exists about the optimal type of implant and best internal fixation procedure, two parameters that directly govern the risk of the most common intraoperative and postoperative complications (diaphyseal fractures, dislocation, non-union). To minimise these risks, we developed a technique combining a transverse subtrochanteric shortening osteotomy, a long ovoid-profiled, cementless stem anchored in the metaphysis, and double tension-band wiring for internal fixation. The primary objective of this study was to evaluate the outcomes of this technique with emphasis on (i) complications and femoral implant survival, (ii) clinical outcomes (functional scores and lower-limb length inequality [LLLI]), and time to healing.

Our technique is associated with low rates of intraoperative and postoperative complications.

This single-centre retrospective cohort study included patients who underwent THA with a transverse subtrochanteric shortening osteotomy and fixation using double tension-band wiring to treat severe (Crowe III or IV) developmental hip dysplasia. The femoral implant was a long, ovoid, cementless stem fully coated with hydroxyapatite. We collected the intraoperative and postoperative complications, survival, LLLI, functional scores (Harris Hip Score [HHS] and Forgotten Joint Score [FJS]), patient satisfaction, and radiographic outcomes.

The study included 31 hips in 25 patients. Two patients (2/31 hips, 6.5%) were lost to follow-up, leaving 29 hips for the analysis of postoperative outcomes. Mean follow-up was 13.7±4.2 years (range, 5.8-18.3 years). The four intraoperative complications (4/31, 12.9%) consisted fracture of the diaphysis (2/31, 6.5%), fracture of the greater trochanter (1/31, 3.2%), and sciatic nerve injury followed by a full recovery (1/29, 3.4%). The 8 (8/29, 27.5%) postoperative complications consisted of dislocation (2/29, 6.9%), stem subsidence (2/29, 6.9%), and non-union (4/29, 13.8%). Femoral implant survival at last follow-up was 87.1% (95% CI, 76.1-99.7). The mean HHS increased from 39.6±12.0 (range, 14-61) before surgery to 81.7±13.2 (range, 48-100) at last follow-up (p<0.01). The FJS at last follow-up indicated that the joint was forgotten in 14/29 (48.2%) cases and caused only acceptable symptoms in 9/29 (31.0%) of cases. Clinically significant (≥ 1cm) LLLI was present in 8/29 (27.6%) patients postoperatively compared to 19/29 (65.5%) preoperatively. The mean LLLI decreased from 20.8±19.7mm (range, 0-60mm) to 5.0±7.3mm (range, 0-30mm). Mean time to healing was 4.3±2.4 months (range, 2-11 months).

Regarding these complex procedures, this technique was associated with low rates of intraoperative fractures and early postoperative complications. However, femoral stem survival was shorter than in earlier studies and the non-union rate was high, despite satisfactory functional and clinical outcomes.