Arthroscopic evaluation of knee articular cartilage: a comparison with plain radiographs and magnetic resonance imaging.

Objective: To compare radiographic, magnetic resonance imaging (MRI), and outpatient direct arthroscopic evaluation of cartilage in patients with osteoarthritis (OA) of the knee.

Methods: Thirty-three patients with OA of the knee were evaluated by plain weight bearing radiographs and arthroscopy using a 1.9 mm arthroscope under local anesthesia. Sixteen of these patients also had MRI of the knee performed. Knee compartments were evaluated using AP weight bearing and lateral radiographs of the knee. MRI and outpatient arthroscopic grading of cartilage were performed within 2 weeks of the plain radiographs. The MRI and arthroscopic evaluations were performed independently and were graded without knowledge of the other.

Results: In patients in whom plain radiographs, MRI, and arthroscopy were compared, the plain radiographs and MRI significantly underestimated the extent of cartilage abnormalities. There was a moderate correlation between imaged cartilage scores and the arthroscopy scores (Pearson correlation coefficient = 0.40). The arthroscopic scores were reproducible with good intra and inter-observer reliability. The arthroscopic procedure was well tolerated and actually preferred over the MRI by the majority of patients. No significant complications were noted as a result of arthroscopy.

Conclusions: Our results indicate that outpatient arthroscopic evaluation is a useful method in evaluating surface cartilage abnormalities and is more sensitive in detecting these abnormalities than either plain radiographs or MRI. Outpatient arthroscopic evaluation of cartilage appears to be a safe, sensitive, and a well tolerated tool for evaluating patients with OA of the knee and may prove to be particularly useful in evaluating response to therapeutic interventions.