Summary

A causal inference analysis comparing the postoperative outcomes of femoroacetabular impingement patients treated with primary hip labral reconstruction versus labral repair.

Abstract

Introduction

Since its creation, labral repair has become the preferred method among surgeons for the arthroscopic treatment of acetabular labral tears resulting in pain and dysfunction for patients. Labral reconstruction is performed mainly in revision hip arthroscopy but can be used in the primary setting for irreparable or calcified labra. The purpose of this study was to compare minimum 2-year patient reported outcomes (PROs) and risk for revision or arthroplasty between primary labral reconstruction and labral repair.

Methods

Patients who underwent primary hip labral repair or reconstruction with the senior author between 2005-2018 were identified from a prospectively enrolled patient outcomes registry. Exclusion criteria included patients not between 18-65 at the time of surgery, confounding injuries (Leggs Calves Perthes, avascular necrosis, femoral head fracture, etc.), prior ipsilateral hip surgery, or minimum joint space less than 2mm. Labral repairs were performed when adequate tissue was available for repair and labral reconstruction was performed when tissue was absent, ossified or torn beyond repair. Inverse propensity score weighted multiple linear (or logistic) regression (IPSW+MLR) was conducted to estimate the causal average treatment effect (ATT) of choosing labral reconstruction versus labral repair upon postoperative PROs and likelihood for subsequent surgery. PRO endpoints included HOS-ADL, Harris Hip Score, WOMAC, SF-12 PCS and patient satisfaction. Covariates included in the IPSW+MLR models included age, sex, baseline PRO score, BMI, year of surgery, joint space, center edge angle, alpha angle, Sharp angle, Tönnis angle, microfracture or grade 3/4 cartilage lesions of the acetabulum or femoral head, hypoplastic labrum, ossified labrum, labral cyst, labral flattening, and labral tear size.

Results

151 primary labral reconstruction and 1000 primary labral repair cases met inclusion criteria. Median follow-up time was 5.4 and 5.8 years for the reconstruction and repair cohorts, respectively. Compared to labral repair, labral reconstruction was associated with a significantly elevated crude (X2-test OR=3.4, 95%CI=[2.0,5.6], p<0.001) and adjusted risk for THA conversion (IPSW+MLR, OR=3.2, 95%CI=[1.2,8.7], p=0.025). The risk for revision arthroscopy was not statistically significantly elevated for labral reconstruction (crude X2-test OR=1.7, 95%CI=[0.9,3.2], p=0.123; IPSW+MLR, OR=1.4, 95%CI=[0.5,3.9], p=0.538). Among patients that did not undergo subsequent surgery, IPSW+MLR estimated a causal effect of choosing labral reconstruction vs labral repair that was significantly negative for for HOS-ADL (ATT=-3.6, 95%CI=[-1.1,-6.1], p=0.005) and WOMAC (ATT=+3.1, 95%CI=[0.6,5.6], p=0.015), while the causal effect for HHS (ATT=-0.2, 95%CI=[-3.0,2.5], p=0.870), SF-12 PCS (ATT=-0.5, 95%CI=[-1.9,1.0], p=0.519) and patient satisfaction (ATT=0.2, 95%CI=[-0.3,0.6], p=0.398) were not statistically significant.

Conclusion

In conclusion, primary labral reconstruction resulted in worse PRO and increased conversion to THA when compared to primary labral repair when using a multivariable causal modeling approach. These data support the use of labral repair in the primary setting of labral tears and reserves the labral reconstruction for more advanced labral pathology or in revisions cases.