Current evidence for clinical phenotypes in knee osteoarthritis: a systematic review

Purpose: Heterogeneity within the population of knee osteoarthritis (KOA) patients impedes progress in our understanding of the disease and development of effective interventions. It has been suggested that distinct phenotypes, or patient subgroups characterised by distinctive features, are present within the wider KOA population. The purpose of this systematic review was to identify current evidence for the existence of distinct clinical phenotypes in KOA. Methods: Literature databases were systematically searched for clinical studies in humans with knee osteoarthritis aimed at subgroup or phenotype identification. Study quality was scored using a checklist for observational studies according to STROBE guidelines. Results: 19 full-text original papers were included in the review from an initial search result of 625 abstracts. These studies addressed the research areas of genetics; anatomy and medical imaging; biochemistry; biomechanics; and epidemiology. Evidence was found for the existence of an inflammatory phenotype linked to biomarkers such as IL-1; a 'chronic pain' phenotype characterized by high prevalence of psychological distress; a 'mechanical overload' phenotype predominantly prevalent as medial tibiofemoral OA; and an obesity and physical inactivity-linked phenotype with high prevalence of muscle weakness and potential links to a group of biomarkers. These phenotypes may not be mutually exclusive and weaker evidence was found for other phenotypes. Conclusions: Preliminary yet consistent evidence across research areas was found for the existence of KOA phenotypes from exploratory studies. Confirmatory studies are needed to establish the validity of these phenotypes, and their relevance to the course of KOA outcomes and to clinical decision making.