EP4.13 MRI and MRA in the investigation of Femoroacetabular Impingement: Are they reliable?
Abstract
Abstract MRI scans have variable sensitivity and specificity for detecting labral tears and transitional zone cartilage injuries in the hip. This study aimed to assess the sensitivity and specificity of pre-operative conventional MRI (cMRI) and direct MR arthrogram (dMRA) for each grade of chondrolabral injury and whether the location or timing of scan influences their effectiveness as a diagnostic tool. This retrospective study reviewed all patients treated for femoroacetabular impingement (FAI) with hip arthroscopy between 2017-2024 at Queen Elizabeth University Hospital, Glasgow. cMRI and dMRA reports were reviewed on Picture Archiving and Communication System (PACS) and compared with intra-operative findings. All hip labral and transitional zone cartilage injuries were classified using the University College London Hospital (UCLH) classification system at the time of hip arthroscopy. Sensitivity and specificity were calculated for each grade of injury. Comparison was made between scans reported by musculoskeletal radiologists at the specialist tertiary centre and those reported in referring hospitals. All scans took place on 1.5T MRI scanners. Three hundred and eighty-five scans (166 males and 219 females) were included. Mean age was 34 years (range 16-57 years). dMRA had improved sensitivity (dMRA 40% vs. cMRI 39%) but were less specific (dMRA 79% vs cMRI 88%) for all injuries. UCLH grade 1 cartilage injuries had low sensitivity (23%) compared to grade 3 (43%) and 4 (63%). Scans reported by specialist musculoskeletal radiologists were more sensitive (42% v 33%) than those reported by district general hospital radiologists. The sensitivity of all scans decreased with time from scan to surgery (43% - 20%) while the specificity increased (70% - 91%). cMRI and dMRA scans had moderate sensitivity and specificity for every grade of labral and cartilage injury. There was better sensitivity for higher grade injuries and poor sensitivity for UCLH grade 1 injury. This study supports the hypothesis that cMRI and dMRA scans are a poor diagnostic tool for FAI associated labral and chondral injuries. A negative cMRI or dMRA scan in the presence of clinical signs of FAI should not be used to offer false reassurance.