Prevalence and Associated Prognosis of Electrocardiographic Findings in Hypertensive Patients. An Analysis of a 5-Year Follow-Up Study

Objective: Various electrocardiographic abnormalities may be identified in hypertensive patients, but their prognostic significance has not been sufficiently studied. We investigated the presence of diverse ECG findings in a hypertensive population and their predictive ability for cardiovascular disease. Design and method: We studied 1851 white Caucasian hypertensive patients (age 58±12years, 51% females) free of cardiovascular disease for a mean period of 5.3±3.4 years. At baseline, all patients underwent standard 12-lead electrocardiography. Left ventricular hypertrophy (LVH) criteria were assessed using both the Sokolow and the Cornel-Voltage formulas. An interventricular conduction delay (IVCD) was defined as a QRS >110msec in the absence of a bundle branch block (BBB). QRS fragmentation was identified as notches in the QRS complex not related to BBB. The QTc interval was considered prolonged if > 440ms in men and >460ms in women. ST depression was defined as >1 mm in depth. The outcome studied was the incidence of cardiovascular morbidity set as the composite of non-fatal coronary artery disease and stroke. Results: LVH criteria were fulfilled in 7% with the Sokolow formula and in 5.3% with Cornell Voltage criteria. A prolonged PR was evident in 6.3%. An IVCD, a left BBB and a right BBB were observed in 9%, 4.1% and 1% of the population respectively. QRS fragmentation was identified in 18% of the study population. An early transition zone was found in 30% of the population and a late transition zone in 19%. QTc was prolonged in 9.8% of cases. Prevalence of TWI was 3.8%, of which 39% was anterior TWI, 73% was lateral TWI and 11% was inferior TWI. Incidence of the composite endpoint during follow-up was 4%. Cox regression analysis revealed that only prolonged QTc (HR: 1.86, 95% CI: 0.97-3.54, p=0.06) and presence of TWI (HR: 2.6, 95% CI: 1.1-5.9, p=0.025) were associated with the composite endpoint. In multivariate models controlling for standard confounders the association with TWI was overall sustained. Conclusions: Among hypertensive patients without cardiovascular disease, ECG abnormalities are frequent. A prolonged QTc and T wave inversion seem to exhibit prognostic significance for cardiovascular events.