Introduction:
Obesity has been linked to increased risk of sudden cardiac death and ventricular arrhythmia. Whether the metabolically healthy obese phenotype is a benign condition, is debatable and few studies examined its ventricular repolarization profile.
Purpose:
To examine the association of metabolically healthy/unhealthy obesity phenotypes with prolonged corrected QT (QTc) interval in a large population-based study.
Methods:
Cross-sectional data from an ongoing cohort study in Poland. Data was collected using a standardized protocol. The QT intervals were obtained from digital standard 12-lead resting ECG and were corrected for heart rate by Bazett’s formula. Plasma lipids and glucose were measured in a fasting state. After excluding drugs known to affect the QT interval (antiarrhythmics, digoxin, antipsychotics), the analytic sample size was 11068 participants, ages 45 to 64 years. Based on the presence of obesity (BMI ge 30 kg/ m
2
) and metabolic syndrome (per the AHA/NHLBI harmonized definition), we defined four phenotypes: metabolically healthy non-obese (MHNO), metabolically unhealthy nonobese (MUNO), metabolically healthy obese (MHO) and metabolically unhealthy obese (MUO). Multivariable linear and logistic regression models were used for analyses.
Results:
The prevalence of the 4 phenotypes was: MHNO: 51.34%, MUNO: 18.07%, MHO: 10.07%, MUO: 20.52%. The prevalence of an increased QTc interval (greater than >430ms in men/450ms in women) was 14.28%, and the prevalence of a highly prolonged QTc interval (greater than 450ms in men/470ms in women) was 5.2%. The age- and sex-adjusted mean QTc across the 4 phenotypes was: MHNO: 417.05ms (416.39- 417.71; MUNO: 418.82 ms (417.71-419.92); MHO: 420.46 ms (418.99 -421.93); MUO: 422.45 ms (421.41-423.49). The age- and sex adjusted odds (OR, 95% CI) of an increased QTc interval (greater than 430/450ms in men/women) were increased in MUNO (1.11, 0.96-1.29), MHO (1.44, 1.20-1.72) and MUO (1.47, 1.28-1.69), compared to MHNO phenotype. These estimates were minimally attenuated after additional adjustment for prevalent CVD, LVH on ECG, smoking, alcohol intake, physical activity and education: MUNO (1.10, 0.94-1.28), MHO (1.45, 1.21-1.73) and MUO (1.44, 1.26-1.66). We did not detect effect modification by sex. We obtained similar results in subgroup analyses restricted to those without diabetes and after excluding those with third degree atrioventricular blocks or conduction abnormalities with QRS>120ms.
Conclusion:
Both metabolically healthy- and non-healthy obese phenotypes had a higher likelihood of an increased/borderline QTc interval compared to the MHNO phenotype. Our study furthered our understanding of ventricular repolarization as reflected in the QTc interval, in the setting of different obesity phenotypes.
Indexes for ventricular repolarization abnormality: Re-evaluation of QT interval.
