PurposeCorrelations between maximum flexion and functional outcomes in total knee arthroplasty (TKA) patients are reportedly weak. We investigated whether there are differences between passive maximum flexion in nonweight bearing and other types of maximum flexion and whether the type of maximum flexion correlates with functional outcomes.Materials and MethodsA total of 210 patients (359 knees) underwent preoperative evaluation and postoperative follow-up evaluations (6, 12, and 24 months) for the assessment of clinical outcomes including maximum knee flexion. Maximum flexion was measured under five conditions: passive nonweight bearing, passive weight bearing, active nonweight bearing, and active weight bearing with or without arm support. Data were analyzed for relationships between passive maximum flexion in nonweight bearing by Pearson correlation analyses, and a variance comparison between measurement techniques via paired t test.ResultsWe observed substantial differences between passive maximum flexion in nonweight bearing and the other four maximum flexion types. At all time points, passive maximum flexion in nonweight bearing correlated poorly with active maximum flexion in weight bearing with or without arm support. Active maximum flexion in weight bearing better correlated with functional outcomes than the other maximum flexion types.ConclusionsOur study suggests active maximum flexion in weight bearing should be reported together with passive maximum flexion in nonweight bearing in research on the knee motion arc after TKA.