Prognostic value of albumin-based indices for mortality after heart failure: a systematic review and meta-analysis

Albumin-based indices have been used to predict mortality after heart failure (HF). We hereby examined the utility of albumin-globin ratio, lactate-albumin ratio, blood urea nitrogen (BUN)-albumin ratio, fibrinogen-albumin ratio, and albumin-bilirubin score (ABS) in predicting mortality after HF. We searched Embase, PubMed, Web of Science, and Scopus databases up to 2nd October 2024 for studies examining association between albumin-based indices and mortality after HF. A random-effect meta-analysis was conducted using multivariate adjusted data to calculate pooled hazard ratio (HR) with 95% confidence intervals (CI). Subgroup analysis was conducted for short-term (< 90 days) and long-term mortality (≥ 1 year) where possible. There were seven studies on ABS, three studies each on BUN-albumin and fibrinogen-albumin ratio, and two studies each for albumin-globulin and lactate-albumin ratio. Meta-analysis showed that ABS was predictive of both short-term (HR: 2.01 95% CI: 1.01, 4.00 I 2 = 62% p = 0.05) and long-term (HR: 1.36 95% CI: 1.05, 1.76 I 2 = 49% p = 0.02) mortality after HF. Meta-analysis also showed that high fibrinogen-albumin ratio (HR: 1.12 95% CI: 1.09, 1.14 I 2 = 0% p < 0.00001), BUN-albumin ratio (HR: 1.84 95% CI: 1.28, 2.64 I 2 = 64% p = 0.001), and lactate-albumin ratio (HR: 1.84 95% CI: 1.68, 2.01 I 2 = 0% p < 0.00001) were predictors of mortality after HF. Very low quality of evidence indicates that albumin-based indices can predict mortality after HF. ABS was found to independently predict short and long-term mortality after HF. Other ratios like the albumin-globin ratio, lactate-albumin ratio, BUN-albumin ratio, and fibrinogen-albumin ratio were also found to predict mortality after HF albeit with limited data.