BUN and High-Dose Loop Diuretics

Abstract & Commentary

By Michael H. Crawford, MD, Editor

Source: Testani JM, et al. Interaction between loop diuretic-associated mortality and blood urea nitrogen concentration in chronic heart failure. J Am Coll Cardiol 2011;58:375-382.

High-dose loop diuretics are often necessary to reduce elevated filling pressures in patients with heart failure. However, they are known to activate neurohormonal mechanisms that may be harmful. Neurohormonal activation can increase blood urea nitrogen (BUN). Thus, these investigators from the University of Pennsylvania hypothesized that BUN would predict loop diuretic-associated mortality. They examined the Beta-Blocker Evaluation of Survival Trial (BEST) database. This was a study of the effect of adding bucindolol to ACE I in class III-IV systolic heart failure patients. All patients with a baseline BUN and loop diuretic therapy were included in this analysis (2456 of 2708 patients). High-dose loop diuretic was defined as 160 mg/day or more of furosemide or its equivalent. The primary outcome was all-cause mortality, which was not different in the bucindolol vs placebo overall results. High-dose diuretics were used in 680 patients (28%). Heart failure severity measures were higher in the high-dose group and they had a higher mortality (hazard ratio [HR] 1.56). After controlling for confounders, this association was no longer significant (HR = 1.06). Baseline BUN also was associated with higher mortality (HR = 1.28) and with other markers of high mortality. However, after controlling for confounders it was still significantly related to survival (HR = 1.3). In those taking high-dose diuretics, an elevated BUN predicted a greater risk of death (HR = 3.09) than in those not on high-dose diuretics. The authors concluded that the risk of mortality associated with high-dose diuretic use is strongly related to BUN concentrations, which suggests that neurohormonal activation is the mechanism.


Patients with more severe heart failure are more likely to receive high-dose loop diuretics, which confounds the data showing that mortality is higher in patients receiving high-dose loop diuretics. In this study, adjustment for covariates eliminated the higher mortality in the high-dose diuretic patients. However, BUN levels in those receiving high-dose diuretics separated those who benefitted from those who had a higher mortality, and this difference persisted after adjustment for other covariates. The cut point was the median BUN level, which was 21 mg/dL.

Urea is freely filtered at the glomerulus and is then subject to considerable tubular reabsorption. Loop diuretics block the tubular absorption of sodium chloride, which results in diuresis, but also renin release, which initializes a neurohormonal cascade. However, the net effect of all the physiologic variables in heart-failure patients make it difficult to predict who will be harmed by high-dose loop diuretics. This study suggests that BUN may be a marker for those patients with high neurohormonal activation in whom high-dose loop diuretics may be harmful. Perhaps patients with high BUN should be treated with other drugs or ultrofiltration.

There are several limitations to this study. It was retrospective and only the baseline BUN was considered. All the patients were class III-IV heart failure. The study was not blinded. Renal function at baseline varied markedly and protein intake was not known. Also, there may have been unmeasured covariates that influenced the data. Finally, this study was one of the few negative studies for beta-blockers in heart failure, so perhaps there is something atypical about the patients. Interestingly, by design, BEST had a high percentage of African American patients. Thus, before these data can be used for clinical purposes, a prospective study to see if a treatment strategy in severe heart failure subjects based on BUN improves outcomes will need to be performed.