Association of increased plasma adipocyte fatty acid-binding protein with coronary artery disease in non-elderly men

This study included 211 consecutive patients (mean age: 66 years, range: 33-87 years) with coronary artery disease (CAD) recruited to undergo coronary angiography, from April 2008 to March 2009, at Kagawa Prefectural Central Hospital, Japan. Patients with CAD had 75% or greater organic stenosis of at least one major coronary artery, confirmed by coronary angiogram, had suffered a myocardial infarction, or had previously undergone percutaneous transluminal coronary angioplasty or coronary artery bypass graft surgery. Patients with hemodialysis, acute coronary syndrome, recent myocardial infarction within 4 weeks, and malignancies were excluded. Control male subjects (n = 211), matched with CAD patients for age (mean age: 66 years, range: 35-85 years), were selected from patients who visited our affiliated hospitals. Controls were characterized by no history of angina or other heart diseases, a normal resting ECG, and normal exercise ECG stress testing. This study protocol complied with the Declaration of Helsinki and was approved by the Ethics Committees of the institute. Informed consent was obtained from all patients before the study entry.

Blood samples were taken after overnight fasting. The plasma was separated and stored at-80°C, and plasma levels of A-FABP (Biovendor Laboratory Medicine, Modrice, Czech Republic) and high sensitivity C-reactive protein (hsCRP; R&D Systems, Minneapolis, MN) were measured by enzyme-linked immunosorbent assay, as previously described[13]. The performance characteristics of these assays were < 7 and < 8% CV intra-assay, and < 5 and < 7% CV inter-assay for A-FABP and hsCRP, respectively.

Risk factors were defined as follows. Diabetes was confirmed according to the criteria of the American Diabetes Association [14], or from a history of diabetes mellitus treatment. Dyslipidemia was defined as one or more of the following criteria: (1) serum triglyceride ≥150 mg/dL, (2) HDL-cholesterol < 40 mg/dL, (3) LDL-cholesterol ≥130 mg/dL, (4) already on lipid-lowering drugs. Hypertension was defined as a resting blood pressure of ≥140/90 mmHg or on regular antihypertensive medications. Smoking was defined as current smokers. eGFR was calculated by the Modification of Diet in Renal Disease (MDRD) equation [15] with coefficients modified for Japanese patients [16]: eGFR (ml/min/1.73 m² ) = 194 × (serum creatinine) ^-1.094 × (age) ^{- 0.287}Ann Intern Med 130 . Renal dysfunction was defined as eGFR < 60 ml/min/1.73 m² .

Coronary angiography was performed according to standard methods. After intracoronary injection of isosorbide dinitrate, angiograms were obtained in two or more views. The coronary angiogram was scored by two independent investigators and according to three techniques [17]: (1) Vessel score: The number of vessels with significant stenosis defined as 50% or greater luminal diameter narrowing. (2) Stenosis score: A modified Gensini score, which has been previously reported. Briefly, the most severe stenosis in each of eight segments was graded according to severity, from 1 to 4. The scores in each of the eight segments were added to give a total score out of 32. (3) Extent score: According to the proportional length of each vessel segment in the coronary artery tree, segments were graded with different maximum numbers of points, as previously reported. The scores of each vessel were added to give a total score out of 100. Stenosis score and extent score may be regarded to reflect the coronary plaque burden.

Plasma A-FABP levels in CAD patients were significantly higher than in control subjects (median [IQR], 20.6 [15.7-27.8] vs. 15.1 [11.7-19.9], p < 0.001, Figure 1). The clinical characteristics of male CAD patients and control subjects are shown in Table 1. In the fourth quartile (plasma A-FABP level ≥24.5 ng/mL), the number of CAD patients was 2-fold that of control subjects, while in the first quartile, the number of CAD patients were less than that of control subjects. CAD patients had a significantly higher body mass index, fasting blood glucose, LDL cholesterol, and hsCRP, and lower level of HDL cholesterol and eGFR. The presence of hypertension, diabetes mellitus, dyslipidemia, and smoking habit was significantly higher in CAD patients than in control subjects.

Among diabetic patients (n = 155), the presence of CAD was higher in those with over the medium level of plasma A-FABP than in those with the lower plasma A-FABP (84% vs. 64%, p < 0.01). Among patients with dyslipidemia (n = 245), the presence of CAD was higher in those with over the medium level of plasma A-FABP than those with lower plasma A-FABP (78% vs. 54%, p < 0.01).

In all subjects, plasma A-FABP was correlated with the body mass index (r = 0.47, p < 0.001), but not with age (r = 0.01, p = 0.86), LDL-cholesterol level (r = 0.07, p = 0.15), or HbA1c (r = 0.09, p = 0.15). Plasma A-FABP in subjects with a history of smoking was significantly higher than that in subjects without a history of smoking (19.6 [15.1-27.2] vs. 17.2 [12.7-24.1], p = 0.03). In each CAD group or control group, medications did not affect the plasma A-FABP level (data not shown).

The association of plasma A-FABP and three angiographic scores, the vessel score, extent score, and modified Gensini score, were analyzed in CAD patients. The level of plasma A-FABP did not differ among patients with one diseased vessel (20.1 [15.3-28.1]), two diseased vessels (20.9 [15.8-27.4], and three diseased vessels (21.6 [16.7-20.9]). Meanwhile, the extent score and the modified Gensini score were correlated significantly with a stepwise increase in plasma A-FABP (Figure 2).

To assess the relation of each factor with CAD patients, simple logistic regression analysis was performed (Table 2). The body mass index, smoking, plasma A-FABP level, diabetes mellitus, fasting blood glucose, hypertension, dyslipidemia, LDL cholesterol, HDL cholesterol, renal dysfunction, eGFR, and hsCRP were significantly associated with the presence of CAD. The uses of angiotensin-converting enzyme inhibitors/angiotensin II receptor blockers (ACEIs/ARBs), calcium channel blockers (CCBs), β-blockers, and statins were also significantly associated with the presence of CAD statistically; however, the following parameters were dependent on each other; fasting blood glucose and diabetes mellitus, lipid profile and dyslipidemia, and eGFR and renal dysfunction. Therefore, diabetes mellitus, dyslipidemia and renal dysfunction were selected as variates for multiple logistic regression analysis, along with body mass index, smoking, hsCRP, plasma A-FABP, and the use of ACEI/ARBs, CCBs, β-blockers, and statins. Multiple logistic regression analysis, including all the above factors revealed that the increase in plasma A-FABP was independently associated with the presence of CAD (p = 0.01) along with traditional CAD risk factors.

Furthermore, to evaluate the impact of plasma A-FABP on CAD presence according to age, we analyzed patients aged < 65 years and ≥65 years separately (Table 3). For analysis of subjects aged < 65, age-matched control subjects (n = 92) against CAD subjects aged < 65 years (n = 102) were re-selected from the control group. The mean ages of the control group and CAD group were comparable (57 ± 6 years and 57 ± 6 years, respectively). Simple logistic regression analysis demonstrated that plasma A-FABP, smoking, body mass index, diabetes mellitus, hypertension, dyslipidemia, renal dysfunction, hsCRP, and the use of ACEI/ARBs, CCBs, and statins were significant factors associated with the presence of CAD, Multiple logistic regression analysis including all significant factors in subjects aged < 65 years revealed that the increase in plasma A-FABP was independently associated with the presence of CAD (p = 0.001) along with smoking and diabetes mellitus. Next, for analysis of subjects aged > 65, control subjects (n = 93) age-matched against CAD subjects aged > 65 years (n = 109) were re-selected from the control group. The mean ages of the control and CAD groups were 75 ± 5 years and 75 ± 6 years, respectively. Simple logistic regression analysis demonstrated that plasma A-FABP, diabetes mellitus, hypertension, dyslipidemia, renal dysfunction, hsCRP, and the uses of ACEI/ARBs, CCBs, β-blockers and statins were significant factors associated with the presence of CAD; however, multiple logistic regression analysis including all significant variates revealed that plasma A-FABP was not an independent factor associated with CAD.

Several lines of evidence have suggested the role of A-FABP in atherogenesis, and it is possible that the high plasma A-FABP levels in patients with CAD in this study were the cause of progressive coronary atherosclerosis. Studies on the molecular function of A-FABP in macrophages have shown that A-FABP deficiency reduced foam cell formation in response to oxidized LDL and increased the cholesterol efflux pathway[18]. A-FABP-deficient mice also showed a significant decrease of vascular atherosclerosis in the absence of differences in serum lipids or insulin sensitivity in hyperlipidemia model mice, and this effect was attributed to the action of A-FABP in macrophages[5]. Thus, A-FABP in macrophages and adipocytes has pathological effects on vessels. In humans, cohort studies suggested that A-FABP plays an important role in insulin resistance and metabolic syndrome[9]. Our study shows that diabetic patients over the medium level of plasma A-FABP had a significantly greater presence of CAD than diabetic patients with lower plasma A-FABP. These results suggest that plasma A-FABP may have a differential impact on CAD from diabetes mellitus. A previous cross-sectional study, however, showed that serum A-FABP was correlated closely with risk factors characterized by abdominal obesity, including insulin resistance, hyperglycemia, increased serum triglyceride, LDL-cholesterol, and decreased HDL-cholesterol[9]. Our findings also found that plasma A-FABP had a weak correlation with hsCRP and eGFR (data not shown). In addition, previous studies showed that medications affect the circulating A-FABP level. Serum A-FABP was reported to be reduced with statin therapy [19], but increased with the treatment of thiazolidinedione[20]. Despite several confounding factors, multiple logistic regression analysis demonstrated that plasma A-FABP was independently associated with the presence of CAD after adjustment for well-known CAD risk factors, hsCRP, and renal dysfunction. Thus, increased plasma A-FABP can be considered a candidate risk factor for CAD, although a large-scale prospective study in the general population is needed.

A-FABP is expressed in adipocytes and macrophages, and an increase in plasma A-FABP may reflect an increased expression in those cells. Plasma A-FABP concentration has been reported to be correlated with carotid atherosclerotic parameters, such as intima-media thickness[11]and plaque volume of the coronary artery determined with IVUS[13]. In line with previous studies, we demonstrated that plasma A-FABP in CAD patients increased with the severity of angiographic coronary stenosis[12]. Besides the severity of coronary stenosis, previous studies showed that plaque vulnerability is more important for future cardiac events[21]. Extent of atherosclerosis is another marker of an increased risk of coronary events. There is a significant association between coronary artery calcium scores, that reflects the actual presence and severity of atherosclerosis, and cardiovascular events [22]. In our study, A-FABP level was closely associated with modified Gensini score and extent score, that reflect coronary plaque burden. Thus, A-FABP might have potential to be a predictor of cardiovascular outcomes.

Several studies showed that genetic involvement in adipokine regulation was associated with atherosclerosis [23]. Regarding A-FABP, the T-87 C allele in the promoter region of the A-FABP gene has been shown to be associated with lower triglyceride levels, a reduced risk for coronary artery disease and type 2 diabetes [8]. Recent genetic studies identified several candidate genes associated with type 2 diabetes, including KCNQ1 [24], while A-FABP also contributed to the development of diabetes mellitus. In fact, a previous study showed that circulating A-FABP could predict the development of type 2 diabetes [10]. A basic experiment showed that interplay between FABPs and pancreatic islets plays an important role insulin secretion [25]. Thus, even though the direct interaction between A-FABP polymorphism and atherosclerosis has not been fully elucidated, the influence of A-FABP on lipid and glucose metabolism [26]may be one of the important mechanisms of atherosclerosis.

In this study, multivariate analysis using traditional CAD risk factors revealed that the association between elevated plasma A-FABP and the presence of CAD remain significant in patients aged < 65 years, but not in subjects aged ≥65 years. This finding is similar to predictive power of cardiovascular risk factors in relation to aging. Several studies documented that impact of risk factors on mortality decreases with age, partly because of selective survival and the influence of comorbidity on risk factor levels [27‐29]. For instance, increased systolic blood pressure or cholesterol in younger subjects may affect cardiovascular mortality events with low-grade elevation or short-term exposure, more than in subjects of advanced age. On the other hand, the effect of each factor for risk control would be reduced in subjects of advanced age. As the progression of coronary artery disease is the manifestation of a cluster of several risk factors, the impact of A-FABP on the development of atherosclerosis would be reduced in subjects of advanced age; thus, the impact of increased plasma A-FABP can be regarded as a coronary risk factor especially in patients aged < 65 years.

This study focused on men, because male gender is an important risk factor of CAD and gender differences in serum A-FABP have been reported [13]. Previously, we have shown that serum A-FAPB in female patients with CAD was significantly higher than in male patients with CAD, whereas serum A-FABP in men was lower than in women. This contradictory difference in serum A-FABP in women might confuse the analysis of the interaction; therefore, we analyzed only men in this study. This sex difference may be partly the result of the relatively higher fat percentage in women than in men, because adipose tissue is the major contributor of circulating A-FABP. Another explanation is the regulation of A-FABP expression by sex hormones. A study showed that the secretion of adiponectin, which is another major adipokine, was suppressed by testosterone [30]. A similar mechanism of the secretion of A-FABP might be possible. Further studies will be needed to elucidate the clinical implication of increased plasma A-FABP in women in respect of CAD.