Background:

Weight loss is a surrogate marker of volume changes with therapy in acute heart failure. Large drops (>50%) in B‐type natriuretic peptide (BNP) levels correlate with changes in wedge pressures. Wedge pressures correlate strongly with orthopnea. Pre‐discharge BNP levels < 250 pg/ml predict excellent 6‐month outcomes. We determined relationships between symptomatic improvement and changes in BNP levels and weight loss with diuresis in telemetry patients with acute decompensated heart failure (ADHF).

Methods:

Patients were shown a visual analog scale (VAS) with word anchors at 10 cm representing “unbearable” dyspnea, at 7 “very bad” dyspnea, at 4 “moderate” dyspnea, and at 0 “no” dyspnea close to presentation in the emergency department and then daily. Rest dyspnea and orthopnea were separately tested as dichotomized outcome variables with a score < 3 being mild or better dyspnea and a score ≥ 4 moderate or worse dyspnea. BNP levels and weight were log‐transformed due to skewness. Models were adjusted for ejection fraction (EF), creatinine clearance, age, sex, body mass index, and comorbidity. Nonparametric analyses and the t test with unequal variance assessed relationships between changes in BNP, weight, and dyspnea scores as appropriate. Patients with rapid atrial fibrillation were excluded because rate‐related BNP elevations may occur. Dialysis patients, patients with sepsis, non‐English speakers, patients with acute coronary syndromes, and patients with dementia were also excluded.

Results:

Paired data for BNP and weight available for 96 patients were analyzed. Mean age was 65 ± 14 years, 50% were male, and mean EF was 36% ± 19%. Median BNP on day 1 was 697 pg/mL (381, 1352 pg/mL). Mean length of stay was 4.9 ± 6.2 days. Change in BNP levels from day 1 to day 2 was 5.9% ± 59%. Weight change from day 1 to day 2 was 2.2% ± 3.3%. Weight changes from day 1 to day 2 in patients with BNP drops of at least 50% compared with those with less were similar. Mean day 2 rest dyspnea and orthopnea scores were similar. BNP levels increased by 165% ± 92% (95% Cl ‐53–21) in patients with day 2 rest dyspnea scores of 0 compared with a decrease of 16% ± 32% (95% Cl 7–25) in the rest of the patients (P = 0.15). Age, sex, and EF were similar in the 2 groups (P = 0.3). No correlation between BNP or weight changes from day 1 to day 2 and improvement in rest dyspnea or orthopnea were found. Day 2 orthopnea scores were lower for patients with day 3 BNP levels < 250 pg/mL compared with those with BNP levels > 250: 2.8 ± 2.9 (95% Cl 1.6–4.0) versus 4.6 ± 3.6 (95% Cl 3.6–5.6), respectively (P = 0.02). Rest dyspnea scores were not different (2.8 vs. 2.7), P = 0.98. A trend toward significance was noted in the difference in orthopnea scores on day 2 between patients with day 4 BNP < 250 and BNP > 250 (P = 0.08); there were no differences in rest dyspnea scores (P = 0.4)

Conclusions:

Changes in BNP levels or weight are unreliable markers of symptom improvement in ADHF. Greater early improvement in orthopnea predicts lower predischarge BNP levels better than improvements in rest dyspnea. Assessment of orthopnea is an important part of assessment of improvement in ADHF.

Author Disclosure:

V. Sharma, SUNY Downstate, support from Biosite for BNP test equipment; N. Rojas, none; E. Ko, none; R. Kumar Amipara, none; P. Khullar, none; S. Zehtabchi, none; R. H. Birkhahn, Methodist Hospital of Brooklyn, material support from Biosite Inc.