Objective: To compare the results of MRI Gadolinium Arthrograms and hip arthroscopy in patients with hip pain.

Design: Retrospective analysis

Setting: St. James University Hospital, Leeds

Participants: 15 patients (10 females and 5males) underwent hip arthroscopy between 1999 and 2002.The procedure was performed by a single surgeon. The mean age of the patients was 39 (Range 24–74).

Outcome Measures: A review of the findings of both the MRI and arthroscopy were undertaken. Attention was paid to the both the specificity and sensitivity of the MRI diagnosis when compared to the findings at arthroscopy.

Results: All 15 patients underwent MRI Arthrograms with intra-articular gadolinium prior to arthroscopy. The time that elapsed between radiological evaluation and surgery was a mean of 10 months (Range 7–15).During this time there had been no significant change in any of the patients’ clinical symptoms.

Of the 15 scans performed only 10 were found to provide an accurate diagnosis confirmed at the time of surgery, The presence of osteochondral defects, loose bodies, synovitis and AVN were all correctly identified.

Of the remaining five cases, two labral injuries were identified radiologically, but arthroscopy proved to be normal. MRI was also found to be particularly inaccurate when assessing for articular cartilage defects, failing to demonstrate moderate to advanced changes in three cases, later confirmed at surgery.

Discussion: MRI is the non-invasive investigation of choice for the assessment of both intra and extra articular hip pathology. However our results show that limitations do exist with the sensitivity and specificity of this investigation. Previous studies (Edwards et al, Villar et al) have demonstrated similar findings .The interpretation of MRI Arthrograms should therefore be carefully considered when establishing a diagnosis for hip pain.