Association between indices of cardiac repolarization, adiposity and metabolic syndrome. The Polish Norwegian Study (PONS)

Abstract

BackgroundSeveral cardiac repolarization indices have been associated with cardiovascular disease (CVD) mortality and arrhythmia risk, but little is known about their relation to adiposity in apparently healthy individuals.Study objectiveTo compare prevalence of T‐wave inversion (TWI) and QT interval duration across levels of adiposity and metabolic health.HypothesisWe hypothesized that repolarization indices are associated with body mass index (BMI), waist circumference (WC), waist‐to‐hip ratio (WHR) and metabolic syndrome.MethodsCross‐sectional data of 10531 CVD‐free participants, age 45‐64 years. TWI was considered relevant if present in leads other than V1‐V3, on a standard 12‐lead resting ECG read by study cardiologists. QT intervals were digitally obtained by automated ECG software and were corrected for frequency (QTc, Bazett formula). Metabolic syndrome was defined according to the International Diabetes Federation criteria. We defined metabolically healthy/unhealthy normal weight (MHNW/MUHNW), metabolically healthy/unhealthy overweight (MHOW/MUHOW) and metabolically healthy/unhealthy obesity (MHO/MUO). We used logistic regression and general linear models to adjust for traditional cardiovascular risk factors and beta blockers medication.ResultsIn the overall study population (mean age [SD] 55.2 [5.3] years, 66.4% women), TWI prevalence was 2.59% and ranged from 1.4%, 2.5% to 3.9% in those with normal BMI, overweight and obesity, respectively. There was linear relation between TWI prevalence and BMI decile (p for trend <0.001). Overweight and/or obesity were associated with a 1.47 (95% CI 1.02‐2.12) and 2.09 (95% CI 1.41‐3.08) higher odds of TWI respectively, after multivariable adjustment including left ventricular hypertrophy (Cornell criteria). TWI prevalence was higher for sex‐specific increased vs. normal WHR (3.16% vs. 1.78%) and WC (3.67% vs 1.90%) respectively; these increases persisted after multivariable adjustment. TWI prevalence increased from 2.12% to 3.64% in the presence of metabolic syndrome, but not statistically significant after accounting for BMI. Compared to MHNW, higher adjusted odds (OR, 95%CI) for TWI were observed for those with MHOW (1.52, 1.03‐2.24), MUHOW (1.72, 1.08‐2.77), MHO (1.98, 1.24‐3.16) and MUHO (2.78, 1.87‐4.14). QTc increased from 423.31 to 425.97 ms in the presence of sex‐specific increased WC (delta 2.67 ms, 95% CI 0.82‐4.52), but there were no statistically significant associations with BMI or metabolic syndrome after multivariable adjustments.ConclusionTWI prevalence showed a gradient with BMI and was associated with regional adiposity. Even traditionally considered metabolically healthy overweight/obesity were associated with increased TWI prevalence suggesting early repolarization abnormalities with weight gain. Potential implications: health promotion efforts in conjunction with a simple routine ECG may aid identification of individuals requiring targeted aggressive prevention interventions.