Thiamine deficiency in patients with congestive heart failure receiving long-term furosemide therapy: a pilot study.
PURPOSE: To test the hypothesis that long-term furosemide therapy in patients with congestive heart failure (CHF) is associated with clinically significant thiamine deficiency via urinary loss. DESIGN: (1) Biochemical evaluation of thiamine status in hospitalized patients with CHF treated with long-term furosemide and in age-matched control patients. (2) Uncontrolled trial of the effect of intravenous thiamine on cardiac performance in a subset of six patients with CHF. SETTING: General medical ward of a teaching community hospital. PATIENTS: Twenty-three patients with chronic CHF receiving furosemide, and 16 age-matched control patients without heart failure and not taking diuretics. Daily furosemide doses were 80 to 240 mg, and duration of furosemide therapy was 3 to 14 months. Patients with identifiable causes of inadequate thiamine intake, absorption, or utilization or increased metabolic requirements were excluded. INTERVENTION: A 7-day course of intravenous thiamine, 100 mg twice daily, in six consenting patients with CHF. RESULTS: A high thiamine pyrophosphate effect (TPPE), indicating thiamine deficiency, was found in 21 of 23 furosemide-treated patients and in two of 16 controls (p less than 0.001). The mean (+/- SE) TPPE (normal: 0% to 15%) in furosemide-treated and control patients was 27.7 +/- 2.5% and 7.1 +/- 1.6%, respectively (p less than 0.001). Despite the high TPPE, the mean (+/- SE) urinary thiamine excretion in the furosemide-treated patients (n = 18) was inappropriately high (defined as greater than 130 micrograms/g creatinine), 410 +/- 95 micrograms/g creatinine, even in comparison with that in the controls (n = 14): 236 +/- 69 micrograms/g creatinine. In six patients treated with intravenous thiamine, the elevated TPPE decreased to normal, from a mean (+/- SE) of 27.0 +/- 3.8% to 4.5 +/- 1.3% (p less than 0.001), indicating normal thiamine utilization capacity. Left ventricular ejection fraction increased in four of five of these patients studied by echocardiography. CONCLUSIONS: These preliminary findings suggest that long-term furosemide therapy may be associated with clinically significant thiamine deficiency due to urinary loss and contribute to impaired cardiac performance in patients with CHF. This deficit may be prevented or corrected by appropriate thiamine supplements.