Abstract 14854: A Prognostic Study of the J wave Patterns on the Resting Electrocardiogram in Ambulatory Patients
Introduction: The prognostic value of J waves and terminal QRS slurs remains controversial. Population studies have demonstrated differing results, with some suggesting that J waves and slurs are associated with cardiovascular death (CVD). Hypothesis: We tested the hypothesis that J waves or terminal QRS slurs on the electrocardiogram (ECG) were associated with cardiovascular death. Methods: Resting digitized 12-lead ECGs from 20,661 ambulatory individuals at the VA Palo Alto Health Care System were visually coded using computer display. Using ≥ 0.1 mV as the amplitude criteria, J-waves were measured at the top of the upward deflection and QRS slurs at the top of the conduction delay on the QRS down-stroke in at least two contiguous leads. Multivariate Cox hazard regression analyses were used to study the associations between J waves, terminal QRS slurs, ST segment slope and CVD. Results: Of the 20,661 subjects, 90% were male (43 ± 8 years of age) and 10% were female (40 ± 10 years of age); 16% were African-American and 7% were Hispanic. Over a median follow-up of 17 years, there were 859 cardiac deaths. A total of 4219 (20%) ECGs had J waves and/or slurs present in the inferior and/or lateral territories: 3318 (16%) ECGs had J waves and/or slurs only in inferior leads (II, III, aVF), 1701 (8%) only in lateral leads (V4-6, I, aVL) and 1198 (6%) in both inferior and lateral leads. None of the J wave/terminal QRS slur patterns were associated with higher rates of cardiovascular mortality (Figure). The subset of ECGs with J wave/terminal QRS slur patterns that also had downward ST slopes were also not associated with CVD. Conclusion: We found no association between higher rates of CVD and any pattern of J waves/terminal QRS slurs, with or without downward ST slope. While previous studies have shown that J waves and terminal QRS slurs are present at higher rates in patients with idiopathic ventricular fibrillation, we did not find them to be predictive of CVD in long-term follow up of a clinical cohort.