Differential correlation network analysis identified novel metabolomics signatures for non-responders to total joint replacement in primary osteoarthritis patients

INTRODUCTION: Up to one third of total joint replacement patients (TJR) experience poor surgical outcome. OBJECTIVES: To identify metabolomic signatures for non-responders to TJR in primary osteoarthritis (OA) patients. METHODS: A newly developed differential correlation network analysis method was applied to our previously published metabolomic dataset to identify metabolomic network signatures for non-responders to TJR. RESULTS: Differential correlation networks involving 12 metabolites and 23 metabolites were identified for pain non-responders and function non-responders, respectively. CONCLUSION: The differential networks suggest that inflammation, muscle breakdown, wound healing, and metabolic syndrome may all play roles in TJR response, warranting further investigation. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (10.1007/s11306-020-01683-1) contains supplementary material, which is available to authorized users.