Association of loop diuretics use and dose with outcomes in outpatients with heart failure: a systematic review and meta-analysis of observational studies involving 96,959 patients.

There is ongoing controversy regarding the association between loop diuretics (LD), especially in high doses, and adverse clinical outcomes in outpatients with heart failure (HF). We performed a systematic review of the evidence for LD in outpatients with HF. We searched MEDLINE, EMBASE, and Cochrane Clinical Trial Collection to identify controlled studies, evaluating the association between LD and morbidity and mortality in patients with HF. The primary endpoint was all-cause mortality and secondary endpoint HF hospitalizations. Quantitative analysis was performed by generating forest plots and pooling adjusted risk estimates across studies using random effects models. Between-study heterogeneity was assessed through Q and I2 statistics. Twenty-four studies with a total of 96,959 patients were included. No randomized studies were identified. Use of LD was associated with increased all-cause mortality compared with non-use (pooled adjusted risk estimates, 1.18; P = 0.001) and increased HF hospitalization rates (pooled adjusted risk estimates, 1.81; P < 0.001). These associations remained significant after excluding studies that included HF patients at discharge from hospital (pooled adjusted risk estimates, 1.31 and 1.89, respectively; P < 0.001 for both). High-dose LD (median dose 80 mg) were also associated with increased all-cause mortality (pooled adjusted risk estimates, 1.99; P < 0.001) compared with low-dose LD. Again, this association remained significant after excluding studies that included HF patients at discharge from hospital (pooled adjusted risk estimates, 1.33; P < 0.001). Existing evidence indicates that LD, especially in high doses, are associated with increased all-cause mortality and HF hospitalization rates. For this reason, prospective, randomized studies are warranted to clarify whether these associations indicate causality or are merely an epiphenomenon due to disease severity. Systematic review registration: PROSPERO database registration number CRD42020153239. Date of registration: 28 April 2020.