Metabolic health and obesity phenotypes in relation to incident heart failure: a systematic review and meta-analysis

Abstract Background Obesity is an established risk factor for incident heart failure, but whether the associated risk depends on metabolic health rather than body size alone remains uncertain. The cross-classification of obesity status and metabolic health into four phenotypes (metabolically healthy/unhealthy, obese/non-obese) offers a framework to disentangle these effects. Whereas the “obesity paradox” concerns prognosis among individuals with established cardiovascular disease, the present synthesis addresses the prevention question of how metabolic health and obesity phenotypes are associated with incident heart failure. Methods We systematically searched PubMed/MEDLINE, Embase, Web of Science, the China National Knowledge Infrastructure, and ClinicalTrials.gov (PROSPERO: CRD420261474858). We included cohort studies that cross-classified obesity status and metabolic health into the four metabolic phenotypes and reported heart-failure outcomes. Effect estimates were pooled with random-effects models, and the certainty of evidence was rated with the GRADE framework. Results Twenty-three cohort studies representing approximately 12.5 million participants (with some overlapping populations) were included in the qualitative synthesis; nine cohort studies (representing approximately 4.2 million participants) contributed heart-failure-specific hazard ratios to the primary meta-analysis. Compared with metabolically healthy non-obesity (MHNO), the pooled hazard ratios were: metabolically unhealthy non-obesity (MUNO; HR 1.42, 95% CI 1.26–1.59; 5 studies), metabolically healthy obesity (MHO; HR 1.65, 95% CI 1.46–1.86; 9 studies), and metabolically unhealthy obesity (MUO; HR 2.12, 95% CI 1.74–2.59; 6 studies), with MUO showing the highest estimate. Metabolic dysfunction was associated with elevated heart-failure risk even among individuals without obesity (MUNO). However, heterogeneity was substantial for MHO (I² = 69.5%) and considerable for MUO (I² = 84.4%), which may limit the generalisability of the pooled estimates. Conclusions Heart-failure risk differed across metabolic health and obesity phenotypes. Metabolic dysfunction identified elevated heart-failure risk even in the absence of obesity, suggesting that metabolic characteristics may provide risk information beyond BMI alone. However, substantial heterogeneity indicates that, although the direction of the association is largely consistent, the magnitude of risk associated with MHO and MUO remains uncertain across populations. Key Message What is currently known about this topic? Obesity is an established risk factor for heart failure; prior meta-analyses were inconclusive. What is the key research question? How are metabolic health and obesity phenotypes associated with incident heart failure? What is new? Across 23 cohorts (9 in the primary meta-analysis), heart-failure risk was highest in metabolically unhealthy obesity (HR 2.12; 6 studies, I² = 84.4%). Estimates derive from different numbers of studies and heterogeneity was high for MHO and MUO; certainty of evidence is low to very low. How might this study influence clinical practice? Metabolic phenotyping may provide complementary risk information beyond BMI, although its incremental predictive value remains untested.