Background: There is wide variability of visit-to-visit (V2V) B-type natriuretic peptide (BNP) in patients with chronic heart failure (CHF), even when they are stable. The prognostic significance of V2V-BNP variability has not been investigated. We aimed to test whether V2V-BNP variability during the stable period of CHF has prognostic value regardless of BNP level. Methods: In 278 stable outpatients (75 ± 10 years, 65% male) with CHF, we studied V2V-BNP variability, which was defined as the coefficient of variance of BNP values measured during 1 year before enrollment. All-cause death and rehospitalization due to HF were considered the primary endpoint. Results: The median V2V-BNP variability was 25.7% (IQR: 19.2–34.4%). During the follow-up period (median 3.2 years), 100 patients reached the endpoint and those with high V2V-BNP variability (≥25.7%) had a significantly higher rate of events (p = 0.001). CHF severity in terms of BNP level and MAGGIC risk score was not significantly different between those with high and low V2V-BNP variability. Multivariable analysis showed that high V2V-BNP variability was independently associated with increased event rates even after adjustment for other known prognostic predictors, including BNP (hazard ratio 1.90, p = 0.003), or for MAGGIC risk score and BNP (hazard ratio 1.72, p = 0.010). The hazard for the outcome consistently increased as V2V-BNP variability increased, with a marked increase up to about 30%. Conclusions: Even in the stable phase of CHF, V2V-BNP variability was associated with worse long-term outcomes, independent of BNP level.

Background: There is wide variability of visit-to-visit (V2V) B-type natriuretic peptide (BNP) in patients with chronic heart failure (CHF), even when they are stable. The prognostic significance of V2V-BNP variability has not been investigated. We aimed to test whether V2V-BNP variability during the stable period of CHF has prognostic value regardless of BNP level. Methods: In 278 stable outpatients (75 ± 10 years, 65% male) with CHF, we studied V2V-BNP variability, which was defined as the coefficient of variance of BNP values measured during 1 year before enrollment. All-cause death and rehospitalization due to HF were considered the primary endpoint. Results: The median V2V-BNP variability was 25.7% (IQR: 19.2–34.4%). During the follow-up period (median 3.2 years), 100 patients reached the endpoint and those with high V2V-BNP variability (≥25.7%) had a significantly higher rate of events (p = 0.001). CHF severity in terms of BNP level and MAGGIC risk score was not significantly different between those with high and low V2V-BNP variability. Multivariable analysis showed that high V2V-BNP variability was independently associated with increased event rates even after adjustment for other known prognostic predictors, including BNP (hazard ratio 1.90, p = 0.003), or for MAGGIC risk score and BNP (hazard ratio 1.72, p = 0.010). The hazard for the outcome consistently increased as V2V-BNP variability increased, with a marked increase up to about 30%. Conclusions: Even in the stable phase of CHF, V2V-BNP variability was associated with worse long-term outcomes, independent of BNP level.