Factors Associated With Complications in 176 Crowe IV Hips Treated With Total Hip Arthroplasty.

Objective: Total hip arthroplasty (THA) in Crowe IV hip was a challenge for surgeons, along with some complications. Currently, most evaluations focused on the surgical techniques and radiographs, which were indirectly measured parameters and easily affected by the operators. Objective factors were not considered. This study aimed to investigate objective factors to predict the complications.

Methods: We retrospectively reviewed a series of Crowe IV patients who received THA between July 2010 and December 2019 in our hospital. Demographics and surgical parameters were collected: gender, age, height, weight, sides of preoperative knee valgus, limb length discrepancy, hip surgical history, osteotomy length, acetabular prosthesis position, acetabular prosthesis sizes, femoral prosthesis, femoral head sizes, femoral prosthesis distal sizes, and acetabular liner. The complications periprosthetic fractures, periprosthetic infection, polyethylene lining wear, postoperative dislocation, limited flexion, limp, knee valgus, knee pain, thigh pain (distal femoral prosthesis), and hip abnormal noise were recorded. Univariable and multivariable logistic regression analyses were used to identify the predictors of complications.

Results: A total of 136 Crowe IV patients (176 hips) were included in this study. The mean follow-up time was 8.87 ± 2.60 (5-14) years. The mean limb length discrepancy was 2.73 ± 2.05 cm. Thirty-two hips had a surgical history. One hundred and three hips underwent intraoperative osteotomy, and the mean osteotomy length was 3.42 ± 1.22 (1-7) cm. Complications occurred in 45% (79/176) Crowe IV hips. The height (odds ratio [OR]: 0.00; 95% confidence interval [CI]: 0.00-0.26), and preoperative left knee valgus (OR: 0.37; 95% CI: 0.16-0.88) were identified as independent significant factors for complications in Crowe IV hips. A residual limp was observed in 34.09%, knee valgus was 23.3%, hip abnormal noise was 7.39%, knee pain was 1.7%, thigh pain (distal femoral prosthesis) was 1.14%, and limited flexion was 0.57%. The incidence of polyethylene lining wear was 6.3%, periprosthetic fractures was 1.7%, postoperative dislocation was 1.14%, and periprosthetic infection was 0.57%.

Conclusion: Our model provided a framework to guide decision-making in Crowe IV hips for surgeons. A tall Crowe IV patient with preoperative left knee valgus was found to have a lower rate of complications.