Objective: Correct application of the Kellgren and Lawrence (K&L) classification system is difficult due to inexact wording of the descriptors. We summarized different descriptions and searched for evidence on the impact of such variations on classification of knee osteoarthritis (OA) in epidemiological studies.
Methods: We searched Medline/Pubmed (1966 to August 2006) for studies of epidemiological cohorts that professed use of the original K&L scale (grade 0-4), and recorded their descriptions of the five grades. The descriptions were compared with each other and with the original description.
Results: We identified five different descriptions. In grade 2, often used as cut-off to classify OA, one description replaced ‘definite osteophytes and possible narrowing of joint space’ (K&L) by ‘definite osteophyte, unimpaired joint space’. Another description for grade 2 was ‘minimal osteophytes, possible narrowing, cysts, and sclerosis’. In some cohort studies, descriptions changed during follow-up. None of the included articles studied the impact of the use of different descriptions.
Conclusion: Major OA cohort studies disagree between and even among themselves on the definition and grading of disease according to the original K&L system. The impact of this disagreement warrants further study, but consensus urgently needs to be reached on a single valid and feasible classification system.