Efficacy and Safety of Intravenous Chlorothiazide versus Oral Metolazone in Patients with Acute Decompensated Heart Failure and Loop Diuretic Resistance

Study objective: To assess the efficacy and safety of intravenous (IV) chlorothiazide versus oral metolazone when added to loop diuretics in patients with acute decompensated heart failure (ADHF) and loop diuretic resistance.

Design: Retrospective cohort study.

Setting: Large urban academic medical center.

Patients: Adults admitted with ADHF between 2005 and 2015 who had loop diuretic resistance, defined as administration of IV furosemide at a dose of 160 mg/day or higher (or an equivalent dose of IV bumetanide), during hospitalization, and who then received at least one dose of IV chlorothiazide (88 patients) or oral metolazone (89 patients) to augment diuresis.

Measurements and main results: The primary efficacy end point was a change in 24-hour net urine output (UOP) from before to after thiazide-type diuretic administration, and the study was designed to test for the noninferiority of metolazone. Safety end points included changes in renal function and electrolyte concentrations. The mean dose of IV loop diuretic therapy (in IV furosemide equivalents) at baseline (before thiazide-type diuretic administration) was higher in the chlorothiazide group (mean ± SD 318.9 ± 127.7 vs 268.4 ± 97.6 mg/day in the metolazone group, p=0.004), but net UOP was similar (mean ± SD 877.0 ± 1189.0 ml in the chlorothiazide group vs 710.6 ± 1145.9 ml in the metolazone group, p=0.344). Mean doses of chlorothiazide and metolazone were 491 ± 282 mg and 5.8 ± 3.5 mg, respectively. Following thiazide-type diuretic administration, net UOP improved to a similar degree (2274.6 ± 1443.0 ml vs 2030.2 ± 1725.0 ml in the chlorothiazide and metolazone groups, respectively, p=0.308). For the primary efficacy end point, metolazone met the threshold for noninferiority by producing a net UOP of 1319.6 ± 1517.4 ml versus 1397.6 ± 1370.7 ml for chlorothiazide (p=0.026 for noninferiority). No significant differences in renal function were observed between the groups. Although hypokalemia was more frequent in the chlorothiazide group (75% with chlorothiazide vs 60.7% with metolazone, p=0.045), no significant differences in the rates of severe hypokalemia or other electrolyte abnormalities were observed between the groups.

Conclusion: Oral metolazone was noninferior to IV chlorothiazide for enhancing net UOP in patients with ADHF and loop diuretic resistance and was similarly safe with regard to renal function and electrolyte abnormalities. Given the significant cost disparity between the two agents, these findings suggest that oral metolazone may be considered a first-line option in this patient population.