FGF21 and GDF15 alongside conventional cardiac biomarkers in myocardial dysfunction and cardiovascular risk in type 2 diabetes: A systematic review and meta-analysis

Subclinical myocardial dysfunction may precede overt heart failure in type 2 diabetes mellitus (T2DM), while conventional cardiac biomarkers largely reflect myocardial wall stress or cardiomyocyte injury. We systematically reviewed circulating fibroblast growth factor 21 (FGF21) and growth differentiation factor 15 (GDF15) across two distinct evidence domains: direct diagnostic or associative evidence for early myocardial dysfunction and longitudinal prognostic evidence for subsequent cardiovascular outcomes in T2DM. MEDLINE/PubMed, Embase, Scopus and Web of Science were searched from inception, and eligible studies related FGF21 and/or GDF15 to objective cardiac phenotypes or cardiovascular outcomes. Four GDF15 cohorts harmonised to a per-1-SD increase in log-transformed GDF15 showed higher cardiovascular risk (pooled HR 1.23, 95% CI 1.10-1.37; I2 = 16.2%), with stable leave-one-out analyses. Three FGF21 cohorts yielded a pooled HR of 1.60 (95% CI 1.08-2.39; I2 = 69.9%). In an exploratory leave-one-out analysis, omission of the Lenart-Lipińska 2013 cohort yielded HR 1.32 (95% CI 1.15-1.52; I2 = 0%); this was interpreted as an influence analysis rather than an alternative primary estimate. Diagnostic evidence remained limited: FGF21 AUC ranged from 0.70 to 0.88, whereas one GDF15 study reported AUC 0.83 for asymptomatic diabetic cardiomyopathy. The current evidence therefore supports prognostic association rather than validated early diagnostic performance or demonstrated incremental predictive utility beyond established biomarkers; prospective head-to-head validation in subclinical disease is required.