Broad P is associated with heart failure rehospitalization in patients with a history of heart failure

A broad P wave in electrocardiography reflects atrial remodeling and is associated with cardiovascular events. Although there are several reports on the association between a broad P wave and cardiovascular events, the association between automatically assessed P-wave duration and cardiovascular events in patients with or without heart failure has not been clarified.

To investigate broad P wave and hospital admission for heart failure in patients with and without a history of heart failure.

We included 4,634 patients enrolled in the Cardiovascular Prognostic Coupling Study in Japan (a registry of subjects with ≥1 cardiovascular risk factors, excluding patients with atrial fibrillation) for whom automated electrocardiography (ECG) data were available. There were 233 patients with a history of heart failure and 4,401 patients without. Twelve-lead ECG was conducted, and the P-wave duration in each lead was analyzed automatically using a 12-lead ECG analysis system (Fukuda Denshi, Tokyo). We selected the maximum P wave duration (Pmax) among the 12 leads. The primary endpoint was heart failure hospitalization. We assessed two P-wave cut-offs (Pmax ≥140 or ≥150 msec) in relation to cardiovascular events.

There were 324 patients with Pmax ≥140 ms and 122 patients with Pmax ≥150 ms. N-Terminal pro brain natriuretic peptide (NT-proBNP) was significantly higher in patients with Pmax ≥140 ms than in those with Pmax <140 ms (median 134.0 vs. 65.7 pg/mL, p<0.001), and also higher in patients with Pmax ≥150 ms than in those with Pmax <150 ms (median 192.0 vs. 66.9 pg/mL, p<0.001). The mean follow-up period was 53 ± 17 months, and the primary endpoint occurred in 73 cases. In a group of patients with heart failure, a broad P-wave was significantly associated with the primary endpoint (≥140 ms: log rank 3.89, p=0.048; ≥150 ms: 9.90, p=0.002); a broad P-wave defined as Pmax ≥150 ms was independently associated with the primary endpoint (HR 3.96, 95% CI 1.64–14.54, p=0.004); and a broad P-wave defined as Pmax ≥140 ms was marginally but significantly associated with the primary endpoint (HR 2.46, 95% CI 0.96–6.29, p=0.061) after adjustment for covariates including NT-proBNP. There were no significant associations between a broad P-wave and the primary endpoint in patients without heart failure (≥140 ms: log rank 0.60, p=0.807; ≥150 ms: 0.93, p=0.336).

Automatically assessed P-wave duration of Pmax ≥150 ms was a significant predictor of readmission for heart failure in patients with a history of heart failure.

Figure