Risk of Mortality Associated With QT and JT Intervals at Different Levels of QRS Duration (from the Third National Health and Nutrition Examination Survey)

Abstract

QT prolongation in the setting of QRS >120 ms is believed to be triggered by prolonged depolarization rather than repolarization. Hence, JT interval is suggested as an alternative to QT interval when QRS duration is prolonged. It is unclear, however, if JT and QT intervals portend similar risk of mortality for different durations of QRS. We examined the association between QT and JT, separately, with all-cause mortality across different levels of QRS duration in 8,025 participants (60 ± 13 years, 41% white and 54% women) from the Third National Health and Nutrition Examination Survey. At baseline (1986 to 1994), 486 participants (6%) had QRS duration ≥120 ms. During a follow-up of up to 18 years, 3,045 deaths (38%) occurred. There were significant nonlinear relations of QT and JT intervals with mortality (p <0.001). Hence, QT and JT were categorized as prolonged (>95th percentile), shortened (<5th percentile), and normal (reference group). In multivariate-adjusted Cox regression models, prolonged JT (hazard ratio [HR] 4.75, 95% confidence interval [CI] 1.86 to 12.11) was associated with increased risk of mortality more than prolonged QT (HR 1.50, 95% CI 1.03 to 2.17) in participants with QRS ≥120 ms (interaction p = 0.02). In participants with QRS duration <120 ms, prolonged QT and JT were equally predictive of all-cause mortality (HR 1.27, 95% CI 1.06 to 1.54, and HR 1.31, 95% CI 1.10 to 1.55, respectively). Similar patterns were observed with shortened QT and JT intervals. In conclusion, although both QT and JT intervals are predictive of mortality, JT is more predictive in the setting of QRS duration >120 ms supporting the use of JT interval in patients with prolonged QRS.

Figure 1. Risk of mortality across JT interval

Cubic regression model adjusted for age, sex, race, heart rate, systolic blood pressure, anti-hypertensive medications, diabetes, HDL, body mass index, total cholesterol, ever smoker, history of cardiovascular disease (history of congestive heart failure, previous stroke, and previous coronary artery disease). Knots were placed at 5^th, 50^th and 95^th percentiles. The non-linear relationship was significant (p <0.001)

Figure 2. Risk of mortality across QT interval

Cubic regression model adjusted for age, sex, race, heart rate, systolic blood pressure, anti-hypertensive medications, diabetes, HDL, body mass index, total cholesterol, ever smoker, history of cardiovascular disease (history of congestive heart failure, previous stroke, and previous coronary artery disease). Knots were placed at 5^th, 50^th and 95^th percentiles. The non-linear relationship was significant (p <0.001)