BACKGROUND

Recent findings in population-based studies suggest that abnormal P wave terminal force in lead V1 (PTFV1), a marker of left atrial abnormalities such as fibrosis, dilatation and elevated filling pressures, is associated with incident ischemic stroke, even in the absence of atrial fibrillation. However, whether PTFV1 predicts incident stroke in hypertensive patients during blood pressure lowering has not been examined.

METHODS

Risk of incident stroke was examined in relation to abnormal PTFV1 on a baseline ECG in 7778 hypertensive patients with ECG left ventricular hypertrophy, no history of atrial fibrillation, in sinus rhythm on their baseline ECG with no incident atrial fibrillation during follow-up, who were randomly assigned to losartan-based or atenolol-based treatment. Results focused on the subset of patients between 55 and 60 years old (n = 1879) because of a significant interaction between PTFV1 and age in Cox analyses. Abnormal PTFV1 was defined by the presence of a negative terminal P wave in lead V1 with amplitude × duration ≥ 4000 μV ms.

RESULTS

During mean follow-up of 4.8 ± 0.9 years, 364 patients (4.7%) of the overall study population and 45 patients (2.4%) in the subset of patients aged 60 years or less experienced a definite stroke. In the overall population, abnormal PTFV1 was not a significant predictor of incident stroke [hazard ratio 1.12, 95% confidence interval (CI) 0.91-1.38, P = 0.301], but there was a highly significant interaction of PTFV1 with age stratified at 60 (P = 0.009, hazard ratio 2.30, 95% CI 1.27-4.13, P = 0.006 for abnormal PTFV1 in the interaction model). Further analyses in the subset of patients aged 60 years or less revealed a higher incidence of stroke occurred in those with abnormal than normal baseline PTFV1: incidence rate per 1000 person-years, 7.8 (95% CI 5.2-11.4) vs 3.4 (95% CI 2.2-5.2; P = 0.004); a greater than two-fold increased risk of incident stroke (hazard ratio 2.31, 95% CI 1.28-4.16, P = 0.005) in univariate Cox analysis; and in multivariable Cox regression models that adjusted for other significant predictors of incident stroke in this population (sex, history of stroke or transient ischemic attack, ischemic heart disease or diabetes, baseline creatinine and in-treatment SBP), that abnormal PTFV1 remained associated with a greater than two-fold increased risk of incident stroke (hazard ratio 2.06; 95% CI 1.14-3.74, P = 0.017).

CONCLUSION

Abnormal PTFV1, a marker of left atrial abnormality, was strongly associated with incident stroke in hypertensive patients, independent of in-treatment SBP and other predictors of incident stroke. This association, in the absence of detectable atrial fibrillation, suggests that an underlying atrial cardiopathy may cause left atrial thrombus formation and a subsequent stroke without intervening clinically recognized atrial fibrillation.