Evaluation of the clinical age-related maculopathy staging system

Objective: To evaluate a clinical classification system, the Clinical Age-Related Maculopathy Staging (CARMS) system, for age-related maculopathy (ARM) using a simple grading scale designed for clinical practice and clinical research protocols. Participants: Two hundred forty-six male and female participants with various stages of ARM who were enrolled during the first 4 years of a longitudinal study were selected for the evaluation of the CARMS system. Design: Cross-sectional comparison study. Methods: The CARMS system divides patients into 5 mutually exclusive categories based on slit-lamp assessment of drusen, retinal pigment epithelial irregularities, geographic atrophy, retinal pigment epithelial detachment, and choroidal neovascularization. Fundus photographs and clinical data of the subjects were used to evaluate this scale. Clinical grades assigned for 492 eyes of 246 patients with varying stages of ARM were compared with grades obtained from photographs evaluated by a reading center. To compare grades obtained from an inexperienced grader with those of an experienced grader, observations based on photographs from 50 randomly selected patients were reviewed. To quantify intraobserver agreement for photographic grades, observations from one observer were compared with those made at a later date by the same observer. Main Outcome Measures: Reliability and validity of the CARMS system in grading various stages of ARM based on clinical examination and photography. Results: The degree of overall agreement between the clinically assigned grade and photographic assessment for 492 eyes was substantial (exact overall agreement, 75%; unweighted kappa, 0.63; weighted kappa, 0.78). For advanced age-related macular degeneration, the sensitivity was 0.83, and specificity was 0.97. When assessing photographic grades, the degree of agreement between an inexperienced and an experienced grader was very high (unweighted kappa, 0.79; weighted kappa, 0.86), and the degree of intraobserver agreement was excellent (unweighted kappa, 0.92; weighted kappa, 0.97). Conclusions: The CARMS system, a 5-level clinical scale, is a valid and reliable staging system that can be used in both clinical practice and in clinical research protocols involving patients with all stages of ARM.