Diabetes mellitus and obesity are increasing in prevalence and are associated with an elevated risk of atrial fibrillation (AF). Given the aging of the United States population, AF is projected to concomitantly increase in prevalence in the upcoming decades. Both diabetes and obesity are associated with insulin resistance. Whether insulin resistance is an intermediate step for the development of AF is uncertain. We hypothesized that insulin resistance is associated with an increased risk of incident AF. We examined the association of insulin resistance with incident AF using multivariate Cox proportional hazards regression analysis adjusting for the established AF risk factors (i.e., age, gender, systolic blood pressure, hypertension treatment, PR interval, significant heart murmur, heart failure, and body mass index). Of the 3,023 eligible participants (55% women; mean age 59 years) representing 4,583 person-examinations (Framingham Offspring fifth and seventh examination cycles), 279 participants developed AF (9.3%) within ≤10 years of follow-up. With multivariate modeling, insulin resistance was not significantly associated with incident AF (hazard ratio comparing top quartile to other 3 quartiles of homeostatic model assessment index 1.18, 95% confidence interval 0.84 to 1.65, p = 0.34). In a community-based cohort with ≤10 years of follow-up, no significant association was observed between insulin resistance and incident AF.
Insulin resistance is a common metabolic substrate that is associated with several cardiovascular conditions. In addition, insulin resistance is associated with inflammation, diabetes, and obesity, all common risk factors for atrial fibrillation (AF). Given the association of insulin resistance with the metabolic syndrome and the risk factors for AF, we hypothesized that insulin resistance predisposes to AF. We used the homeostasis model assessment index for insulin resistance (HOMA-IR), a validated research tool of insulin resistance to examine the relation of insulin resistance to incident AF in the community.
We evaluated Framingham Heart Study Offspring who attended either the fifth (1991 to 1994) or seventh (1998 to 2001) examination cycles (total possible person examinations 7,338). The enrollment details have been previously and extensively reported. We excluded 1,580 participants <50 years old (low risk of incident AF), 214 with prevalent AF, 588 with prevalent diabetes (for whom the insulin levels would not be informative), and 373 participants with missing HOMA-IR data at both examinations. The institutional review board at Boston University Medical Center approved the study protocols for all examination cycles, and the participants provided written informed consent at each study visit.
The participants attended a routine Framingham Heart Study clinical visit every 4 to 8 years. Blood pressure was measured at rest twice using a mercury column sphygmomanometer. The body mass index was calculated as the weight in kilograms divided by the square of the height in meters. Clinically significant heart murmur was diagnosed by the presence of a grade 3 or greater of 6 systolic murmur, or any diastolic murmur, as determined during standardized examination at the Heart Study. Hypertension treatment was ascertained by self-report of medication. Diabetes was defined as fasting serum glucose ≥126 mg/dl, a history of physician-diagnosed diabetes, or the use of medications for diabetes. Heart failure was adjudicated, incorporating Framingham clinic and outside medical records, on the basis of major and minor clinical criteria, as described previously.
Fasting blood samples were collected at each examination. The standardized insulin levels were measured in plasma as immunoreactive insulin. Insulin resistance was defined as the top quartile of the previously validated HOMA-IR, assessed using the formula: (fasting plasma insulin [μU/ml]) × (fasting plasma glucose [mmol/L])/22.5. Fasting plasma insulin was measured with different assays at the 2 examinations. At examination 5, we used ethylenediaminetetraacetic acid plasma as total immunoreactive insulin and calibrated to serum levels for reporting purposes. The cross-reactivity of this assay with proinsulin at mid curve is approximately 40%; the intra-assay and interassay coefficients of variation ranged from 5.0% to 10.0%. At examination 7, the insulin level was specific, having essentially no cross-reactivity to insulin split-products (Linco Research, St. Charles, Missouri). The assay coefficient of variation was <10% at examination cycle 5 and <6.8% at examination cycle 7.
The identification of AF was made from the records collected from the participants’ Framingham examinations (interim cardiovascular events were routinely ascertained by Heart Study physicians), outside office visits, and hospitalizations. For Framingham Offspring participants, the biennial health history updates included a routine question on AF. The participants were considered to have AF if an electrocardiogram showed either AF or atrial flutter. Incident AF cases underwent review by 1 of 2 Framingham cardiologists (D.L. or E.J.B.).
Descriptive statistics were examined, including percentages for discrete variables and mean and standard deviations for continuous variables. Insulin resistance was analyzed as a dichotomous variable defined as the upper quartile of HOMA-IR (yes/no). We assessed HOMA-IR and clinical risk factors for AF at examination 5 and followed up the participants for development of incident AF for ≤10 years or their seventh examination. We redefined the baseline characteristics at examination cycle 7, excluded those with interim AF and followed forward for the development of incident AF for ≤10 years. Multivariate-adjusted hazard ratios (HR) were estimated using Cox modeling to examine the relation between insulin resistance and AF for ≤10 years of follow-up after confirming the proportionality of hazards. HRs in the first model were adjusted for age and gender. In addition to age and gender, the second model was further adjusted for systolic blood pressure, hypertension treatment, PR interval, significant heart murmur, and heart failure—all established clinical risk factors for AF, as described previously. The third model had the same covariates as the second model plus the body mass index. We determined the primary analysis had 80% power to identify an association between insulin resistance and new-onset AF at a HR of 1.41 with an α level of 0.05. All analyses were generated using SAS software, version 9.1 (SAS Institute, Cary, North Carolina). A 2-sided p <0.05 was considered statistically significant.
The baseline characteristics of the 4,583 person examinations, representing 3,023 unique participants (55% women; mean age 59 years), are listed in Table 1 . A total of 279 participants developed AF during the 10 years of follow-up, including 64 with and 215 without insulin resistance at baseline (7,661 person-years follow-up with insulin resistance vs 27,426 person-years follow-up without insulin resistance).
|Characteristic||n = 3,023|
|Age (years)||59.2 ± 6.9|
|Body mass index (kg/m 2 )||27.4 ± 4.7|
|Systolic blood pressure (mm Hg)||127 ± 18|
|Fasting glucose (mg/dl)||96 ± 10|
|Electrocardiographic PR interval (ms)||164 ± 24|
|Significant precordial heart murmur||1.7%|
|Prevalent heart failure||0.4%|
|Homeostasis model assessment of insulin resistance (mg/dl)||6.1 ± 3.3|
We examined the association of insulin resistance with incident AF, adjusting for established AF risk factors in multivariate Cox proportional hazard models ( Table 2 ). Insulin resistance was not associated with AF in the age- and gender-adjusted models (HR 1.27, 95% confidence interval 0.92 to 1.76, p = 0.15) or in models adjusting for established AF risk factors, regardless of whether the body mass index was included. The cumulative incidence of AF according to the presence or absence of insulin resistance is shown on Figure 1 .