Background:
Nonalcoholic fatty liver disease (NAFLD) is a common condition driven by the obesity epidemic. It is associated with cardiometabolic risk factors including diabetes, obesity, and hyperlipidemia, but also cardiovascular disease events, independent of these factors. No prospective studies have investigated the association of NAFLD with stroke risk.
Hypothesis:
NAFLD is associated with the risk of stroke in the REasons for Geographic and Racial Differences in Stroke (REGARDS) case-cohort study.
Methods:
The REGARDS study recruited 30,239 participants from the contiguous U.S., in order to study the reasons for regional and racial differences in stroke mortality. The REGARDS case-cohort study consists of 569 cases of incident stroke with 5.4 years follow up and a cohort random sample of 1,104. The Fatty Liver Index (FLI) was used a surrogate marker for NAFLD. It is calculated as e
X
/(1 + e
X
) x 100, where x = 0.953*log(triglycerides) + 0.139*BMI + 0.718*log(γ-glutamyltransferase) + 0.053*waist circumference - 15.745. An FLI >60 is considered a positive score, 20-60 an intermediate score, and <20 a negative score. After excluding 68 participants who reported heavy alcohol consumption and 87 with baseline stroke, Cox proportional hazards models were used to calculate the hazard ratio (HR) and 95% confidence interval (CI) of stroke for FLI category, adjusting for age, race, sex, and the Framingham stroke risk factors and stratified by body mass index (BMI).
Results:
In the cohort sample, 44% of participants had NAFLD based on the FLI and 19% had a negative score. Compared to those without NAFLD, individuals with a positive score were more likely to be male (51% vs. 28%), have hypertension (69% vs. 40%), dyslipidemia (68% vs. 37%) diabetes (35% vs. 8%), and higher BMI (mean 33.7 vs. 23.0 kg/m
2
; all p<0.001). No participant with BMI < 20 kg/m
2
had NAFLD by FLI. NAFLD was not associated with risk of stroke in a model adjusted for age, race and sex; HR 1.00 (95% CI 0.69-1.46), or a model further adjusted for Framingham stroke risk factors; HR 0.71 (95% CI 0.45-1.11). Stratifying by BMI group (20-30 kg/m
2
), there was no association between NAFLD and stroke risk in those with BMI 20-<25 or 25-30 kg/m
2
. We were unable to analyze NAFLD in the BMI >30 group, due to low number of negative scores. When analyzed as a continuous variable among those with BMI 30 kg/m
2
, the HRs for a 10 unit higher FLI score were 0.92 (95% CI 0.84-1.01) and 1.17 (95% CI 0.97-1.42), respectively, adjusted for age, sex, race, and stroke risk factors.
Discussion:
NAFLD, as determined by a positive FLI score, was not associated with risk of stroke although FLI score was borderline associated with stroke risk in those with a BMI >30. Results raise the possibility that NAFLD represents end organ damage from an adverse metabolic profile, and is not a mediator of stroke risk.
Premature Atrial Contractions on the Screening Electrocardiogram and Risk of Ischemic Stroke: The Reasons for Geographic and Racial Differences in Stroke Study