Natriuretic peptide levels are valuable in ruling out HF exacerbation as a cause of dyspnea, though results must be considered in the context of obesity, renal dysfunction, and angiotensin receptor neprilysin inhibitor (ARNI) use.

Breathing Not Properly¹ was a multicenter prospective study that demonstrated the value of B-type natriuretic peptide (BNP) levels among 1586 patients presenting to the ED with acute dyspnea. The diagnosis of HF exacerbation was made clinically by ED physicians and confirmed by two cardiologists who had access to medical records but were blinded to BNP levels. Compared to diagnostic criteria based on history, physical examination findings, or other
laboratory tests, BNP was more able to identify (accuracy of 83.4% at a cut-off of ≥100 pg/mL) and rule out HF exacerbation as the cause of dyspnea (negative predictive value of 96% at a cut-off of <50 pg/mL, 95% CI 94%-97%).

A 2015 meta-analysis, which included Breathing Not Properly, provided additional evidence about the diagnostic utility of BNP and of N-terminal pro B-type natriuretic peptide (NT-proBNP, a biologically inactive prohormone that becomes active BNP when its N-terminal fragment is cleaved) in patients presenting to acute care with dyspnea.² Based on 15,263 natriuretic peptide test results, lower thresholds (BNP <100 pg/mL, NT-proBNP <300 pg/mL) have high sensitivity (95% and 99%, respectively), high negative predictive value (94% and 98%, respectively), but low specificity (67% and 43%, respectively) for diagnosing HF exacerbation. There was no statistically significant difference in diagnostic accuracy between BNP and NT-proBNP. These findings support the use of BNPs in ruling out HF exacerbation as a cause of acute dyspnea.

A 2015 meta-analysis investigated the diagnostic accuracy of NT-proBNP for ruling out HF exacerbation in patients with chronic kidney disease,³ suggesting that higher cutoff values should be used in this setting. Because BNP is metabolized by adipose tissue, levels may be inappropriately low in obese patients with an elevated body mass index.⁴ Because BNP is a substrate for neprilysin, use of an ARNI will increase BNP levels, such that a rising BNP level could reflect treatment response, exacerbation, and/or treatment failure.⁵ These findings suggest that BNP levels should be interpreted in view of renal dysfunction, obesity, and ARNI use.

Outcomes are comparable between bolus and continuous intravenous furosemide dosing, with high doses associated with greater initial weight and fluid loss without long-term effect on renal function. Patients on high outpatient doses may benefit more from bolus than continuous administration.

DOSE-AHF⁶ was a multinational, double-blind, randomized controlled trial assessing intravenous furosemide dosing in HF exacerbation. In a factorial design, 308 patients were randomized to continuous infusion versus twice daily bolus dosing, as well as low-dose (equivalent to the outpatient oral dose) versus high-dose (2.5 times the outpatient oral dose) furosemide. Patients were eligible to participate if they had presented to the hospital within the prior 24 hours with at least one sign and one symptom of HF exacerbation and had received an oral loop diuretic (furosemide 80-240 mg daily or equivalent dose) for at least 1 month prior to hospitalization. Patients were excluded for signs of cardiogenic shock such as systolic blood pressure <90 mm Hg, serum Cr >3 mg/dL, or requirement of vasodilators or inotropic agents other than digoxin. Primary outcomes included symptomatic improvement (assessed via patients’ global assessment of symptoms quantified via the area under the curve) and increase in creatinine (from baseline to 72 hours). Secondary outcomes included net fluid loss.

There was no significant difference between continuous and bolus dosing in terms of symptomatic improvement (P = .47) or Cr increase (mean change 0.05 + 0.3 mg/dL with bolus dosing vs. 0.07 + 0.3 mg/dL with continuous infusion; P = .45). There was greater net fluid loss at 72 hours in the high-dose group (4899 + 3479 mL vs. 3575 + 2635 mL; P = .001). Compared to low-dose group, the high-dose group also had a greater proportion of patients who had a >0.3 mg/dL Cr increase at 72 hours (23% vs. 14%; P = .04) but not at 60 days (P > .05).

To evaluate whether these results were a function of outpatient furosemide dose prior to hospitalization, a follow-up analysis of DOSE-AHF compared bolus versus continuous intravenous furosemide administration and stratified outcomes by higher (≥120 mg/d) versus lower (<120 mg/d) outpatient furosemide equivalent doses.⁷ Multivariate regression analysis revealed that for every 10 mg more of outpatient furosemide equivalent dose, bolus administration was associated with more net fluid loss at 72 hours (95 mL more vs. 88 mL less in the continuous infusion group, P = .02).

A meta-analysis of 10 randomized clinical trials (including DOSE-AHF) comparing continuous and bolus administration among 518 total
patients found no difference in the incidence of increased creatinine (weighted mean difference [WMD] 0, 95% CI −0.09 to 0.09; P = .31), length of hospitalization (WMD −1.06, 95% CI −3.88 to 1.76; P = .06), or all-cause mortality (RR 1.13, 95% CI 0.61-2.10; P = .49).⁸ The difference in weight loss significantly favored continuous infusion (WMD 0.78 kg, 95% CI 0.03-1.54; P = .04).