Risk markers and incidence of atrial fibrillation in northern Sweden

Abstract: Background: Atrial fibrillation (AF) is the most common clinically significant arrythmia with a prevalence of approximately 3% in the general population. Less is known about the incidence of AF. In order to reduce the incidence of AF, it is of essence to identify modifiable risk factors for the disease.Aims: The aims of this thesis were (1) to estimate the incidence of AF and to assess the prevalence of provoking factors and risk factors for stroke and systemic embolism at the time of AF diagnosis, (2) to study the association between alcohol consumption and risk of AF, (3) to study the association between weight, height, weight change, and risk of AF, and (4) to study the association between normal or high normal blood pressure (BP), compared to optimal BP, and risk of AF.Methods: To determine the incidence of AF and the prevalence of provoking factors and risk factors for stroke and systemic embolism at AF diagnosis, an observational study was performed between January 1, 2011, and December 31, 2012, in the municipalities of Skellefteå and Norsjö, Västerbotten, Sweden. Diagnosis registries were searched for cases of incident AF. All AF diagnoses were verified by electrocardiogram. Data regarding provoking factors, type of AF and presence of risk factors for stroke and systemic embolism (as assessed by the CHA2DS2-VASc score) was obtained from medical records. Incidence was calculated by dividing the number of incident AF cases by the time at risk for the population.The association between alcohol consumption, weight, height, weight change, normal BP, high normal BP, and risk of AF was investigated in a population-based cohort study of participants of the Västerbotten Intervention Programme (VIP). Residents of Västerbotten County aged 30, 40, 50 and 60 years who had participated in the VIP health examinations between January 1, 1988, and September 5, 2014, were included. Individuals who had been diagnosed with AF before participating in the VIP were excluded. Study participants were followed until a diagnosis of AF, death, migration from the study area, or the end of the study on September 5, 2014. Incident AF cases were identified using the Swedish National Patient Registry. The health examinations included measurements of height and weight, systolic BP, diastolic BP, fasting glucose, oral glucose tolerance, and cholesterol. Participants also answered a questionnaire addressing any history of diabetes and myocardial infarction, alcohol use, education level, smoking habits, medications, and physical activity. Optimal BP was defined as BP < 120/80 mm Hg. Normal BP was defined as BP 120–129/80–84 mm Hg. High normal BP was defined as BP 130–139/85–89 mm Hg. Hypertension was defined as BP ≥ 140/90 mm Hg. Cox proportional hazards regression analysis was used to assess the associations between alcohol consumption, weight, height, weight change, normal BP, high normal BP, and risk of AF. These were performed with crude, age-adjusted, and multivariable models adjusted for other cardiovascular risk factors (age, sex, body mass index, hypertension, cholesterol, previous myocardial infarction, diabetes, problem drinking, smoking, education level, and leisure-time physical activity).Results: The incidence of AF was 4.0 cases of AF per 1000 person-years. In approximately one fifth of participants, a provoking factor was present at the first episode of AF. The CHA2DS2-VASc score was 2 or higher in 81% of participants. Permanent AF was the most common type of AF and was seen in about one third of the participants.When studying the association between alcohol consumption and risk of AF, 109,230 health examination participants were included in the study cohort and were followed for 1,484,547 person-years. During the study period, 5230 participants developed incident AF. Men with alcohol consumption in the highest quartile (≥ 4.83 standard drinks weekly) had a hazard ratio (HR) of 1.21 (95% confidence interval [CI] 1.09-1.34) for AF in a multivariable model when compared to men with an alcohol consumption in the lowest quartile (< 0.90 standard drinks weekly). In men, problem drinking was also associated with an increased risk of AF (HR 1.24, 95% CI 1.10–1.39 in a multivariable model). Among women, no significant association between alcohol consumption, problem drinking, and risk of AF was identified.In a fully adjusted model, height, weight, body mass index, and body surface area were positively associated with risk of incident AF. Among participants who underwent two health examinations 10 years apart, 1142 persons developed AF. The mean weight change from baseline was a weight gain of 4.8%. There was no significant association between either weight gain or weight loss and risk of AF.In the study of the association between normal BP, high normal BP, and risk of AF, 109,697 persons with a total of 162,982 health examinations were included. Incident AF was diagnosed in 5260 participants. There was an increased risk of incident AF in persons with normal BP (HR 1.11, 95% CI 1.01–1.27) and in persons with high normal BP (HR 1.23, 95% CI 1.10–1.38) compared to optimal BP in a multivariable model. Systolic BP and diastolic BP were associated with risk of incident AF, with HRs of 1.11 (95% CI 1.10–1.13) and 1.13 (95% CI 1.09–1.16) per 10 mmHg, respectively, in multivariable models. A restricted cubic spline analysis indicated a gradually increasing risk of AF with increasing systolic BP and diastolic BP.Conclusions: AF has an incidence of 4.0 per 1000 person-years. Alcohol consumption and problem drinking were associated with higher risk of AF in men, but not in women. Weight, height, body mass index, and body surface area were associated with increased risk of incident AF. Normal BP and high normal BP, when compared to optimal BP, were associated with increased risk of incident AF.