Chondrolabral junction pathology in Femoroacetabular Impingement Syndrome is underestimated by pre-operative MRI and have no significant correlation with baseline outcomes.

Abstract

To assess correlation of pre-operative symptoms, labral damage and chondrolabral junction pathology on MRI with intra-operative findings in patients undergoing hip arthroscopy for femoroacetabular impingement (FAI). Consecutive patients with FAI treated surgically were included. Data were collected on patient demographics and preoperative clinical outcome measures. Three radiologists performed retrospective independent reviews of pre-operative MRIs in 96 hips. The presence of cartilage fraying, chondrolabral separation, labral tear pattern (intrasubstance, linear, complex), chondral flap, cartilage delamination, and ligamentum teres (LT) pathology (degeneration/partial or complete tear) were recorded on preoperative MRI and intraoperatively during arthroscopy. Inter-reader and inter-method agreement were measured using weighted-Cohen's or Fleiss'-kappa, or Kendall's coefficient. Associations between MRI and arthroscopic findings and preoperative modified-Harris-Hip-Score (mHHS) and Non-Arthritic-Hip-Score (NAHS) were assessed using Mann-Whitney-U or Kruskal-Wallis test. All readers detected lower rates of chondrolabral separation, cartilage delamination, and chondral flap, and higher rates of LT tear on MRI compared to arthroscopy. Linear tears were the most common tear pattern detected by readers on MRI (29%-50%) and arthroscopy (71%). The inter-reader agreement was moderate for LT findings and fair for remaining MRI variables. Between MRI and arthroscopy, there was moderate agreement for labral fraying for one reader (k = 0.303, 95% CI[0.106-0.501], p = 0.004) and fair agreement for labral tear pattern for one reader (k = 0.187, 95% CI[0.037-0.337], p = 0.006). There was poor-to-no agreement between MRI and arthroscopy for the remaining MRI variables for all readers (p = 0.05). Preoperative mHHS and NAHS scores demonstrated no significant association with preoperative MR or intraoperative findings (p > 0.05). MRI labral and chondrolabral findings in patients with FAI underestimated the intraoperative pathology identified during hip arthroscopy and were not associated with preoperative patient reported outcome measures. III (retrospective case series).