Near-infrared spectroscopy for arthroscopic evaluation of cartilage lesions: results of a blinded, prospective, interobserver study.

Abstract

Mechanical tests to grade cartilage damage are limited by the instruments used and by the ability to access all areas of cartilage within a joint. Better methods to diagnose cartilage injury or degeneration are needed. Purpose/ To detect the interobserver variance of arthroscopic cartilage grading by subjective judgment using the International Cartilage Repair Society (ICRS) score and by objective measurement using near-infrared (NIR) spectroscopy. We hypothesized that objective measurement of cartilage lesions by NIR spectroscopy will yield more valid results than routine grading using the ICRS score. Cohort study (diagnosis); Level of evidence, 2. Fifteen patients undergoing arthroscopic knee operations were evaluated by 4 experienced arthroscopists independently. The cartilage lesions within the medial knee compartment were estimated by each observer using the ICRS grade and by measurements with a special arthroscopic NIR spectroscopy probe. The ICRS grading had a poor interobserver agreement, with a mean Fleiss kappa index of κ = 0.173. Only in 10% (6 of 60) of judged cartilage areas did all 4 surgeons grade the cartilage areas with the same result. In 17 areas (28.3%), the surgeons had a variance of 2 or more grades. In the remaining cases, the surgeons varied within 1 grade. The objective NIR spectroscopy-obtained measurements of cartilage resulted in a significant correlation within the observers of R = 0.885 ± 0.036 (P < .001). Our results of interobserver evaluation in real-time arthroscopic cartilage grading suggest that this subjective grading is not satisfactory. This study emphasizes the need for objective measurement techniques for arthroscopic cartilage grading. Near-infrared spectroscopy has a good interobserver correlation. Thus, this method could be developed in the future as a precise method of measuring cartilage lesions.