Introduction: Obesity is associated with higher risk of incident AF, but the underlying mechanisms are not well understood. Increased pericardial fat deposition may lead to atrial fibrosis and renin-angiotensin system activation due to free diffusion of cytokines into the thin atrial wall, promoting AF development. Little is known about the association of pericardial fat volume with incident AF.
Hypothesis: We assessed the hypothesis that greater pericardial fat volume is associated with higher AF risk in MESA and JHS, overall and in four race/ethnic groups.
Methods: Pericardial fat volume was measured on chest CT scans (performed 2000-02 in MESA, 2007-09 in JHS) in 18 2.5-mm slices, from 1.5 cm above to 3.0 cm below the superior extent of the left main coronary artery, using Volume Analysis software (GE Healthcare, Waukesha, WI). Data were combined across the 2 studies. Participants with prevalent AF before the scan were excluded. Incident AF was identified by hospital discharge diagnosis codes for AF or atrial flutter, by study ECG at a follow-up visit, or, for those enrolled in fee-for-service Medicare, by an inpatient or outpatient claim with an AF diagnosis in any position. We used Cox regression to estimate adjusted hazard ratios for incident AF.
Results: A total of 8056 participants (6681 in MESA; 1375 in JHS) had pericardial fat volume measured and were followed for clinical events. Among MESA participants, 1855 were AA, 2568 white, 1470 Hispanic, and 788 Chinese; all JHS participants were AA. In the combined data, the average age was 62 years; 55% were women. Greater pericardial fat volume was associated with male sex, older age, white or Hispanic race/ethnicity, greater BMI and systolic blood pressure (SBP), treated hypertension (HTN), impaired fasting glucose, and diabetes mellitus. Despite more obesity, AA participants had on average the lowest pericardial fat volume. During an average of 9 years of follow-up in MESA and 4 years in JHS, a total of 614 cases of incident AF were identified. Whites had the highest unadjusted AF incidence and AA the lowest. In all 4 race/ethnic groups, pericardial fat volume was positively associated with unadjusted AF incidence. After adjustment for age, sex, race/ethnicity, and study, greater pericardial fat volume was associated with higher risk of incident AF (HR=1.17 per SD pericardial fat volume [41 ml], 95% CI 1.09-1.26). After further adjustment for BMI, height, diabetes, SBP, and treated HTN, the association was attenuated (HR 1.06 per SD, 95% CI 0.97-1.16). Associations did not differ in subgroups defined by sex, race/ethnicity, or study.
Conclusion: Greater deposition of fat in the pericardium is associated with higher AF incidence and higher adjusted risk of incident AF. Much of this association appears to be related to obesity, diabetes, and HTN. Lower average pericardial fat volume may explain in part the observed lower AF incidence in AA than in whites.