Plasma-free fatty acids (FFAs) are largely derived from adipose tissue. Elevated levels of FFA and fatty acid–binding protein 4 (FABP4), a key cytoplasmic chaperone of fatty acids, have been associated with adverse cardiovascular outcomes, but limited data are available on the relation of these biomarkers with cardiovascular and total mortality. We studied 4,707 participants with a mean age of 75 years who had plasma FFA and FABP4 measured in 1992 to 1993 as part of the Cardiovascular Health Study, an observational cohort of community-dwelling older adults. Over a median follow-up of 11.8 years, 3,555 participants died. Cox proportional hazard regression was used to determine the association between FFA, FABP4, and mortality. In fully adjusted models, FFA were associated with dose-dependent significantly higher total mortality (hazard ratio [HR] per SD: 1.14, 95% confidence interval [CI] 1.09 to 1.18), but FABP4 levels were not (HR 1.04, 95% CI 0.98 to 1.09). In a cause-specific mortality analysis, higher concentrations of FFA were associated with significantly higher risk of death because of cardiovascular disease, dementia, infection, and respiratory causes but not cancer or trauma. We did not find evidence of an interaction between FFA and FABP4 (p = 0.45), but FABP4 appeared to be associated with total mortality differentially in men and women (HR 1.17, 95% CI 1.08 to 1.26 for men; HR 1.02, 95% CI 0.96 to 1.07 for women, interaction p value <0.001). In conclusion, in a cohort of community-dwelling older subjects, elevated plasma concentrations of FFA, but not FABP4, were associated with cardiovascular and noncardiovascular mortality.
Plasma-free fatty acids (FFA), a by-product of lipolysis, are largely derived from adipose tissue. Several studies have shown that elevated levels of FFA are associated with insulin resistance and diabetes. In addition to diabetes, FFA have also been associated with hypertension, atrial fibrillation, coronary heart disease, and cardiovascular disease (CVD). Fatty acid–binding protein 4 (FABP4) serves as a carrier protein in the transport of FFA and other lipophilic substances. FABP4 has also been associated with insulin resistance and diabetes and incident heart failure and poor outcomes after acute ischemic stroke. Despite the association of FFA and FABP4 with cardiovascular risk factors and CVD, their relation with mortality in older adults is unclear. Studies evaluating FFA and mortality have produced conflicting results for CVD mortality and 1 study indicated an increase in noncardiovascular deaths. Studies analyzing the association between FABP4 and mortality have been limited to patients with ischemic stroke and end-stage renal disease. The present study aimed to evaluate the association of FFA and FABP4 with total and cause-specific mortality in a cohort of community-dwelling older men and women.
Methods
The Cardiovascular Health Study (CHS) is a prospective, population-based cohort consisting of 5,888 men and women aged ≥65 years who were recruited from a random sample of Medicare-eligible residents from 4 US communities (Forsyth County, North Carolina; Sacramento County, California; Washington County, Maryland; and Allegheny County, Pennsylvania). A detailed description of the methods and procedures has previously been published. From 1989 to 1990, 5,201 participants were enrolled and in 1992 to 1993, a supplemental cohort of 687 predominantly African-Americans was recruited at 3 of the original sites (except for Washington County). Participants were eligible if they were not wheelchair dependent or institutionalized, did not require a proxy for consent, were not receiving treatment for cancer, and were expected to remain in their respective region for the upcoming 3 years. Participants were contacted every 6 months for follow-up, alternating between a telephone interview and a clinic visit until 1989 to 1999 and by telephone interview only after that. Each participant gave informed consent, and the institutional review board at each center approved the study. For this analysis, the 1992 to 1993 clinic visit was used as baseline. Of the 5,265 participants who participated in the 1992 to 1993 examinations, 558 subjects had missing data on FFA and/or FABP4 and were excluded from the analysis. Thus, 4,707 participants were included in the analysis.
Plasma samples collected at the 1992 to 1993 examinations were stored at −70°C in the central laboratory at the University of Vermont. Plasma FFA concentration was measured by the Wako enzymatic method, which requires the acylation of CoA by the fatty acids in the presence of added acyl-CoA synthetase. Acyl-CoA produced is oxidized by added acyl-CoA oxidase with generation of hydrogen peroxide and in the presence of peroxidase permits the oxidative condensation of 3-methy-N-ethyl-N(B-hydroxyethyl)-aniline with 4-aminoantipyrine to form a purple-colored adduct. The latter is then measured colorimetrically at 550 nm. Two measurements were taken, and the average was used in the current analysis. The interassay coefficient of variation was 3.54% to 8.17% (detectable range 0.0156 to 1.50 mEq/L). Plasma FABP4 concentration was measured using standard enzyme-linked immunosorbent assay kits (Biovendor ELISA, Ashville, North Carolina). The interassay coefficient of variation was 2.61% to 5.32% (detectable range 5 to 250 ng/ml).
Surveillance for mortality occurred during alternating telephone interviews and clinical examinations every 6 months through 1999 and then exclusively through telephone contacts every 6 months thereafter as previously described. Briefly, deaths were confirmed and classified by a mortality review committee using information from hospital records, death certificates, autopsy and coroner reports, insurance records, obituaries, and interviews with physicians or next of kin. By these methods, ascertainment of vital status was complete for 100% of participants.
Covariate data from the 1992 to 1993 examinations were used in the analysis. Information on age, gender, race, educational attainment, physical activity, hormone replacement therapy, alcohol consumption, and smoking status were based on self-report. Weight, height, and waist circumference were measured using standardized protocols. Leisure-time activity (kcal/wk) was assessed using a modified Minnesota Leisure-Time Activities Questionnaire. Body mass index (BMI) was calculated as weight in kilograms divided by the square of height in meters. Missing values for smoking and height were carried forward from previous years if available. Seated blood pressure at rest was measured in the right arm using a habitus appropriate cuff. Diabetes was defined as a use of insulin or oral hypoglycemic agents, a fasting glucose level of ≥7 mmol/L (126 mg/dl), or a nonfasting glucose level of ≥11.1 mmol/L (200 mg/dl). Lipids, fasting glucose, C-reactive protein, cystatin C, and albumin were measured in fasting blood specimens as previously described.
Baseline characteristics of the study participants were summarized according to quartiles of FFA; continuous variables are presented as mean ± SD and categorical variables as percentages. Incidence rates for mortality were calculated per 1,000 person-years. Cox proportional hazards regression was used to estimate the association of FFA and FABP4 with mortality to allow adjustment for covariates. Subjects were censored for death, loss to follow-up, or end of mortality ascertainment (December 31, 2010). Cubic splines were used to assess the form of the association of FFA and FABP4 with mortality. Because the association of both biomarkers with mortality appeared approximately linear, FFA and FABP4 were modeled continuously. Confounders included in model 1 were age, gender, race, clinic, and education (less than high school vs high school or more). Model 2 included the variables in model 1 with the addition of BMI, cystatin C, albumin, kilocalories of physical activity, alcohol intake (0, <7, ≥7 drinks/wk), smoking status (never, former, and current), hormone replacement therapy for women, and self-reported health status (excellent, very good, good, fair, and poor). The correlation between FFA and FABP4 was determined using Spearman’s rank correlation coefficient. To evaluate intermediate pathways by which FFA and FABP4 might lead to mortality, we fit additional models that included diabetes, prevalent CVD, and C-reactive protein. We tested for interaction between FFA and FABP4 by including both FFA and FABP4 and their cross-product term (FFA × FABP4) in the model. We also looked for evidence of effect modification of the relation between each biomarker and mortality by testing interaction terms for age (continuous), gender, prevalent CVD, diabetes, and BMI (continuous). We also evaluated the association between FFA and cause-specific mortality, including deaths because of CVD, cancer (any type), dementia, infection (pneumonia, sepsis, or other infection), respiratory disease, trauma (including fractures), and other (liver disease, gastrointestinal disease, renal failure, amyotrophic lateral sclerosis, Parkinson’s disease, bladder disease, metabolic conditions, amyloid, failure to thrive, myelodysplastic syndrome, and other musculoskeletal disease). Schoenfeld residuals and plots of the residuals over time were used to evaluate proportional hazard assumptions; there were no meaningful violations.
Results
The mean age of the study participants was 75 years (range 65 to 98), and 58.3% were women. Over a median follow-up of 11.8 years, 3,555 participants died. Baseline characteristics of the study population are listed in Table 1 . FFA and FABP4 levels had a modest correlation (r = 0.18). Mortality rates per 1,000 person-years were 61.6, 62.9, 67.4, and 71.3 across consecutive quartiles of FFA. Corresponding values for FABP4 quartiles were 65.7, 65.4, 61.1, and 70.7, respectively.
| Variable | Quartiles of FFA (mEq/L) | |||
|---|---|---|---|---|
| Q1 (1180) | Q2 (1177) | Q3 (1173) | Q4 (1177) | |
| Age (years) | 74.1 ± 4.9 | 74.7 ± 5.3 | 75.2 ± 5.4 | 75.4 ± 5.6 |
| Men | 61.2% | 46.1% | 34.5% | 25.1% |
| African-American | 14.9% | 17.0% | 17.5% | 17.5% |
| Education: high school or more | 77.0% | 72.5% | 73.8% | 68.8% |
| Body mass index (kg/m 2 ) | 26.2 ± 4.0 | 26.8 ± 4.6 | 27.1 ± 5.0 | 27.3 ± 5.3 |
| Waist circumference (cm) | 96.3 ± 11.3 | 97.1 ± 13.4 | 97.9 ± 13.6 | 98.4 ± 14.3 |
| Glucose (mg/dl) | 104.7 ± 30.7 | 105.4 ± 29.3 | 106.9 ± 31.8 | 116.4 ± 46.0 |
| Cystatin-C (mg/L) | 1.14 ± 0.38 | 1.11 ± 0.29 | 1.12 ± 0.33 | 1.11 ± 0.37 |
| Albumin (g/dl) | 3.92 ± 0.27 | 3.96 ± 0.27 | 3.98 ± 0.27 | 4.02 ± 0.27 |
| Physical activity (kcal/d) | 1642 ± 1920 | 1462 ± 1787 | 1329 ± 1612 | 1290 ± 1687 |
| Alcoholic drinks/week | ||||
| None | 50.6% | 54.5% | 55.0% | 58.9% |
| ≤7 | 38.7% | 35.3% | 36.2% | 29.1% |
| >7 | 10.7% | 10.2% | 8.8% | 12.0% |
| Smoking | ||||
| Never | 38.1% | 44.4% | 47% | 52.5% |
| Former | 51.1% | 44.3% | 43.8% | 39.3% |
| Current | 10.9% | 11.3% | 9.2% | 8.2% |
| Estrogen use | 9.2% | 12.9% | 13.0% | 16.7% |
| Health status | ||||
| Excellent | 9.2% | 7.6% | 4.8% | 4.8% |
| Very good | 35.3% | 32.8% | 29.7% | 25.5% |
| Good | 37.1% | 40.7% | 45.1% | 43.8% |
| Fair | 16.4% | 17.5% | 18.2% | 22.6% |
| Poor | 2.0% | 1.4% | 2.2% | 3.2% |
| Prevalent diabetes mellitus | 12.0% | 12.9% | 14.6% | 21.7% |
| Metabolic syndrome | 38.5% | 43.6% | 48.2% | 55.9% |
| Systolic blood pressure (mm Hg) | 131.7 ± 20.9 | 135.6 ± 21.6 | 137.2 ± 21.4 | 140.8 ± 21.4 |
| Diastolic blood pressure (mm Hg) | 70.5 ± 10.8 | 71.7 ± 11.2 | 71.2 ± 12.2 | 71.7 ± 11.5 |
| Low-density lipoprotein cholesterol (mg/dl) | 119 ± 3 | 121 ± 3 | 121 ± 4 | 118 ± 4 |
| High-density lipoprotein cholesterol (mg/dl) | 50 ± 1 | 52 ± 1 | 54 ± 2 | 57 ± 2 |
| Triglycerides (mg/dl) | 131.6 ± 73.5 | 141.1 ± 79.1 | 148.1 ± 94.9 | 155.7 ± 91.9 |
| Free fatty acids (mEq/L) | 0.27 ± 0.06 | 0.41 ± 0.03 | 0.54 ± 0.04 | 0.76 ± 0.14 |
| Fatty acid binding protein-4 (ng/ml) | 29.7 ± 17.5 | 32.4 ± 17.1 | 35.3 ± 17.7 | 39.5 ± 21.6 |
In a multivariable Cox proportional hazards model, FFA, but not FABP4, was associated with total mortality ( Table 2 ). The addition of components of the metabolic syndrome not already in the model (triglycerides, fasting glucose, waist circumference, and diastolic blood pressure) did not appreciably alter the results (results not shown). Simultaneous adjustment for both FFA and FABP4 did not appreciably alter either variable’s association with mortality in the fully adjusted model (hazard ratio [HR] 1.13, 95% confidence interval [CI] 1.09 to 1.18 for FFA; HR 1.02, 95% CI 0.96 to 1.07 for FABP4). We did not find evidence of an interaction between FFA and FABP4 (p = 0.45).
| Number of Events | Model 1 | Model 2 | |||
|---|---|---|---|---|---|
| HR | 95% CI | HR | 95% CI | ||
| FFA Q1 (reference) | 3555 | 1.00 | 1.00 | ||
| FFA Q2 | 1.02 | (0.93–1.12) | 1.05 | (0.95–1.15) | |
| FFA Q3 | 1.16 | (1.06–1.28) | 1.20 | (1.08–1.32) | |
| FFA Q4 | 1.30 | (1.18–1.43) | 1.34 | (1.21–1.48) | |
| FFA per SD | 1.12 | (1.08–1.16) | 1.14 | (1.09–1.18) | |
| FABP4 Q1 (reference) | 3555 | 1.00 | 1.00 | ||
| FABP4 Q2 | 1.16 | (1.05–1.27) | 1.07 | (0.97–1.18) | |
| FABP4 Q3 | 1.19 | (1.08–1.32) | 1.05 | (0.94–1.17) | |
| FABP4 Q4 | 1.51 | (1.36–1.68) | 1.13 | (0.99–1.28) | |
| FABP4 per SD | 1.21 | (1.17–1.25) | 1.04 | (0.98–1.09) | |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
