Natriuretic response prediction equation for use with oral diuretics in heart failure

Background and Aims Limited data are available to assess oral diuretic response in outpatients with heart failure (HF). The natriuretic response prediction equation (NRPE) predicts natriuresis following a loop diuretic dose using a urine sample 2 h after the dose and was validated to accurately predict intravenous diuretic response. The primary aim was to validate the NRPE's assessment of oral diuretic response in patients with HF. Methods The NRPE was evaluated in two HF patient cohorts receiving oral loop diuretics: Mechanisms of Diuretic Resistance (MDR) and TRANSFORM-Mechanism. Participants received their home oral loop diuretic followed by a supervised timed urine collection including spot urine samples at 1 and 2 h. Patients quantified their self-assessed diuretic response (urine volume) via a standardized survey. A poor diuretic response was defined as cumulative natriuresis < 50 mmol over the study visit. Results The MDR cohort included 318 oral diuretic administrations from 237 patients. The NRPE predicted a poor natriuretic response with an area under the curve (AUC) of .87 [95% confidence interval (CI) .83-.91] and similar accuracy to the previously validated intravenous NRPE performance (P = .16). Patient's ability to self-estimate their diuretic response was poor with an AUC of .57 (95% CI .44-.70) and significantly worse than the oral NRPE (P < .001). In TRANSFORM-Mechanism (110 oral diuretic administrations), the NRPE had similar operating characteristics (AUC .89, 95% CI .80-1.0) for poor diuretic response. Conclusions Natriuretic response to an oral diuretic can be rapidly and accurately assessed with a urine sample collected 2 h after an oral diuretic dose and the NRPE.