The study was approved by the medical ethical committee of Zhongnan Hospital of Wuhan University. All participants were Han population from China, who gave their informed, written consent. From 2012 to 2014, we consecutively recruited patients who underwent selective coronary angiography for suspected coronary atherosclerosis and who were admitted within 24 h from chest pain at cardiovascular department. The diagnosis of CAD was defined as the presence of coronary lesions ≥50 % in at least one major artery segment assessed by two independent cardiologists who had no information of the patients’ clinical characteristics and biochemical results. We excluded all patients with additional complications including viral hepatitis, tumor, autoimmune disease, serious liver or renal dysfunction, myocardiopathy or valvular heart disease and patients received with heparin treatment.

Serial blood samples were taken after admission for routine laboratory parameter tests, including liver function, kidney function, lipids and lipoproteins and myocardial enzymes. Estimated glomerular filtration rate (eGFR) was calculated using the Modification of Diet in Renal Disease (MDRD) study equation: eGFR = 186 × ((serumcreatinine)^(‐ 1.154)) × ((age)^(‐0.203)) × [0.742 if female]. AMI was defined by typical clinical symptoms, electrocardiograph changes and positive serum biomarkers [11, 12]. Unstable angina (UA) was established based on the criteria of Brawmwald using clinical symptoms, non-ST segment elevation and negative serum biomarkers [13].

Serum samples were collected, separated and stored at −80 °C within 24 h of admission. Samples were incubated for 15 min at room temperature and then a single determination of FFAs content was tested by enzymatic assay (Clinimate NEFA, Tokyo, Japan).

Catheterization and multiple views were imaged and maximal stenosis in each coronary artery segment was evaluated by a cardiologist according to the segmental classification system of the Coronary Artery Surgery Study. The extent of angiographically documented CAD was quantified in the left anterior descending artery, the left circumflex artery, and the right coronary artery, the criteria were as follows: normal coronary arteries (smooth, with either no stenosis or stenosis of <10 % of the luminal diameter), mild coronary arteries (stenosis of 10 % to 50 % of the luminal diameter in ≥ 1 coronary artery), or 1, 2 or 3-vessel disease, defined as stenosis of more than 50 % of the luminal diameter.

To assess coronary artery disease burden based on coronary angiogram, the Gensini score was calculated according to Gensini scoring system [14]. The Gensini score equaled the sum of all segment scores (where each segment score equaled the segment weighting factor multiplied by the severity score). Severity scores assigned to the specific percentages of luminal diameter reduction of the coronary artery segment were 32 for 100 %, 16 for 99 %, 8 for 90 %, 4 for 75 %, 2 for 50 %, and 1 for 5 %.

Variables were expressed as the mean ± SD, median (25th–75th percentile) or proportions and compared using Mann–Whitney U-test, Kruskal-Wallis H-test or the χ² test in univariable analysis. Binary logistic regression analysis was used to identify correlated risk factors. All analyses were performed using PASW Statistics, version 13.0 (SPSS, Chicago, IL, USA).

The study population consisted of 404 SCAD (stable coronary artery disease) patients, including 294 chronic stable angina patients (male 56.1 %, aged 68 ± 11),110 old myocardial infarction patients (male 76.4 %, aged 66 ± 12) and 488 ACS patients, including 108 UA patients (male 56.6 %, aged 68 ± 11), 191 NSTEMI (non-ST segment elevation myocardial infarction) patients (male 76.2 %, aged 67 ± 13) and 189 STEMI (ST segment elevation myocardial infarction) patients (male 82.4 %, aged 62 ± 13).

The baseline demographic, clinical characteristics and laboratory parameters according to the quartiles of the FFAs levels (Q1 0.12–0.37, n = 223; Q2 0.38–0.54, n = 218; Q3 0.55–0.78, n = 228; Q4 0.79–4.57, n = 223) for the study population were summarized in Table 1. The distribution of FFAs quartiles were slightly unbalanced with traditional cardiovascular risk factors such as age, smoking, fasting blood glucose, hs-CRP and the atherogenic lipid parameters including TC, LDL-C, TG (P < 0.05). In addition, the necrosis and ischemia parameters dramatically increased according to FFAs quartiles such as cTnT, CK-MB, maximal stenosis and Gensini score (P < 0.05).

The FFAs levels in ACS patients were higher than in SCAD patients ((0.68 mmol/l (0.45, 0.911) vs. 0.44 mmol/l (0.3, 0.63), P < 0.001, respectively)) (Fig. 1a), which remained a significant correlation after adjustment for age and gender (P < 0.001). Among patients with ACS, the higher FFAs levels remained significant between STEMI and UA + NSTEMI patients ((0.77 mmol/l (0.55, 1.1) vs. 0.58 mmol/l (0.4, 0.83), P < 0.001, respectively)) (Fig. 1b), which was significant after adjustment for age and gender (P < 0.001). Meanwhile, the FFAs levels in UA + NSTEMI patients were higher than in SCAD patients ((0.58 mmol/l (0.4, 0.83) vs. 0.44 mmol/l (0.3, 0.63), P < 0.001, respectively)) (Fig. 1c), and the difference still remained significant after adjustment for age and gender (P < 0.001), suggesting that FFAs content increased closely with the severity of coronary artery lesions in CAD.

Meanwhile, the occurrence of ACS and STEMI increased according to the FFAs quartiles (P value for trend = 0.003, 0.001) (Fig. 1d, e). To explore the relationship of FFAs and the prevalent ACS and STEMI, a binary logistic regression analysis was performed. After adjusted for traditional cardiovascular risk factors including age, SBP, DBP, smoking, history of hypertension, diabetes mellitus and hyperlipidemia, the FFAs levels remained positively associated with prevalent ACS and STEMI ((for the occurrence of ACS, OR(95 % CI) = 9.956(5.21, 19.023), P < 0.001; for the occurrence of STEMI, OR(95 % CI) = 8.428(4.619, 15.376), P < 0.001)) (Tables 2 and 3).

To further explore whether the FFAs levels could reflect the severity of ischemia and necrosis focus in ACS patients, we performed Spearman’s correlation analysis. It showed that FFA level was positively correlated with Gensini score and maximal stenosis (r = 0.224, P < 0.001 and r = 0.224, P < 0.001) and remained significant after adjustment for age and gender (P = 0.001 and P = 0.003). The occurrence of a higher Gensini score with more than 40 increased according to the FFAs quartiles (P value for trend < 0.001) (Fig. 1f), after adjusted for traditional cardiovascular risk factors including age, SBP, DBP, smoking, history of hypertension, diabetes mellitus and hyperlipidemia, the FFAs levels remained a risk factor for a higher Gensini score with more than 40 ((OR(95 % CI) = 3.741 (1.826, 7.664), P < 0.001)) (Table 4). Correlations of FFAs content with the infarct size in ACS patients, estimated by CK-MB and cTnT were performed, showing that FFA levels had positively weak correlations with infarct size (r = 0.09, P < 0.001 and r = 0.022, P < 0.001).

We constructed the multivariate linear regression analysis to elucidate the association of FFAs with inflammation parameters, after adjusting for traditional cardiovascular risk factors including age, SBP, DBP, smoking, history of hypertension, diabetes mellitus and hyperlipidemia, which showed after natural logarithm transformation that FFAs were associated with hs-CRP (β = 0.003, p = 0.048) and WBC (β = 0.042, p = 0.008), respectively.

The whole study population were divided into four groups according to the FFAs quartiles and each group was sectionalized into 2 subgroups according to their high/low levels of WBC (≤10×10⁹, >10×10⁹) or hs-CRP (≤3.0 mg/l, >3.0 mg/l) respectively [15]. As shown in Fig. 2a, among all FFAs quartile groups, patients with hs-CRP more than 3.0 mg/l were more susceptible to ACS, compared to patients with hs-CRP less than 3.0 mg/l (P <0.05); the trend for ACS occurrence in both high and low hs-CRP groups was significantly elevated with the distribution of higher FFA level (P value for trend < 0.05) (Additional file 1: Figure S1A). The occurrence of STEMI was elevated with higher hs-CRP, except in the lowest FFAs quartile Q1 group (P < 0.05) (Fig. 2b); a similar trend was observed in STEMI susceptibility and FFAs level in both high and low hs-CRP groups (P value for trend < 0.05) (Additional file 1: Figure S1B). Meanwhile, higher hs-CRP level was associated with a higher Gensini score in the FFAs quartile Q2 group (P < 0.05) (Fig. 2c); the percentage of Gensini score for more than 40 dramatically increased according to FFAs quartiles in both high and lowe hs-CRP groups (P value for trend <0.001) (Additional file 1: Figure S1C). The similar trendsexist in WBC counts and ACS susceptibility (Fig. 2d, Additional file 1: Figure S1D), WBC counts and STEMI susceptibility (Fig. 2e, Additional file 1: Figure S1E), WBC counts and higher Gensini score (Fig. 2f, Additional file 1: Figure S1 F), respectively.

The trend of ACS occurrence, STEMI attack and higher Gensini score with higher FFAs in high hs-CRP and WBC levels versus low hs-CRP and WBC levels were suggestive of an interaction between hs-CRP, WBC counts and FFAs in relation to the progress of myocardial ischemia. After performed by logistic regression analysis, we found a multiplicative interaction between hs-CRP, WBC and FFAs in ACS and STEMI susceptibility and a higher Gensini score ((in relation to ACS susceptibility, FFAs*hs-CRP, OR(95 % CI) = 1.029(1.016, 1.043), P < 0.001, FFAs*WBC, OR(95 % CI) = 1.357(1.281, 1.438), P < 0.001; in relation to STEMI susceptibility, FFAs*hs-CRP, OR(95 % CI) = 1.016(1.008, 1.024), FFAs*WBC, OR(95 % CI) = 1.246(1.194, 1.3), P < 0.001; in relation to a higher Gensini score, FFAs*hs-CRP, OR(95 % CI) = 1.033(1.013, 1.054), P = 0.001, FFAs*WBC, OR(95 % CI) = 1.065(1.02, 1.112), P = 0.004)) (Fig. 2g, h, i), which suggests that FFAs influence the progress of ischemia involved by inflammation processes.