Aims It is common in heart failure (HF) trials, especially in HF with preserved (HFpEF) and mildly reduced ejection fraction (HFmrEF), to select for increased risk of outcomes ('enrichment'). We investigated the association between loop diuretic use and common trial outcomes. Methods and results Patients in the Swedish HF Registry with HFmrEF and HFpEF were divided into three groups: no loop diuretic use, 1-40 mg furosemide equivalent, and >40 mg. Outcomes were assessed for all patients and high risk for exacerbation patients (previous HF hospitalization [HFH] or N-terminal pro-B-type natriuretic peptide >= 300 ng/L). Among 25 986 patients, 41.3% were not taking loop diuretics, 33.5% had 1-40 mg, and 25.2% had >40 mg; 80.7% were at high risk. With increasing diuretic dose, the crude risk of all outcomes increased in both analysis groups. Event rates for cardiovascular death (CVD)/HFH were as follows for total: no loop diuretic: 6, dose 1-40 mg: 17, dose >40 mg: 34 and as follows for high risk: no loop diuretic: 8, dose 1-40 mg: 19, dose >40 mg: 36. Hazard ratios (HRs) for CVD/HFH with no loop diuretic as reference for the total population were: 1-40 mg HR 2.59, >40 mg HR 4.91 (all p < 0.001), and for high risk were: 1-40 mg HR 2.21, >40 mg HR 4.05 (all p < 0.001). Conclusion In HFpEF and HFmrEF, loop diuretics were used in 58.7% of patients. Use of and higher doses were strongly associated with HF trial-relevant outcomes. These associations were similar regardless of other risk markers. In HFpEF/HFmrEF trials, requiring loop diuretics as an enrichment factor will reduce the number of eligible patients but increase event rates considerably.