In this prospective, singe-center study, consecutive patients who underwent coronary angiography for diagnostic purposes at our center between February 2013 and March 2014 were enrolled. Patients with congenital heart disease, coronary artery anomaly, cancer, obstructive pulmonary illness, or renal insufficiency and those already on hypolipidemic drugs were excluded from the study. A signed form indicating informed consent was obtained from each participating patient. The protocol of the study was approved by the Ethics Committee of our institution and the study was conducted in accordance with the ethical standards of Good Clinical Practice and the Helsinki Declaration.

All patients were asked to report their age and gender. Body-mass index (BMI) was estimated based on the height and weight measurements. Patients were defined as having hypertension if they displayed a systolic blood pressure ≥ 140 mmHg, diastolic blood pressure ≥ 90 mmHg, or self-report of physician diagnosis of hypertension and/or current use of antihypertensive medications. Diabetes mellitus was considered present if the patient presented with fasting glucose levels > 126 mg/dL, glycated hemoglobin (HbA1c) levels > 6.5%, or self-report of physician diagnosis of diabetes and/or use of antidiabetic medications. Patients were categorized as smokers if they reported smoking within last one year.

Each participating patient was evaluated for total cholesterol, triglycerides, high-density lipoprotein (HDL) cholesterol, low-density lipoprotein (LDL) cholesterol, very-low-density lipoprotein (VLDL) cholesterol, and non-HDL cholesterol levels using standard techniques. In addition, lipid ratios of triglycerides to HDL cholesterol and triglycerides to non-HDL cholesterol were calculated. Patients with abnormal lipid levels were identified by cutoff points of: >200 mg/dL for total cholesterol, >150 mg/dL for triglycerides, >130 mg/dL for LDL cholesterol, ≤40 mg/dL for HDL cholesterol, >4 for triglycerides/HDL cholesterol ratio, and >0.9 for triglycerides/non-HDL cholesterol ratio.

All patients underwent coronary angiography through transfemoral/transradial route using standard percutaneous techniques in the typical orthogonal projections. The FR index was used to assess the extent of coronary disease. The scores in this index range from 0 to 15. Each of the three main coronary arteries was scored separately from zero to five. The scores were: 0 for no arteriographic abnormalities, 1 for trivial luminal narrowing of <29%, 2 for localized luminal narrowing of 30% - 68%, 3 for multiple luminal narrowings of 30% - 68%, 4 for luminal narrowing of 69% - 100% without 100% occlusion of proximal segments; and 5 for total obstruction of a proximal segment. [9] Based on these, patients were classified in to four groups: patients with 0 FR index score, patients with 1 - 4 FR index score, patients with 5 - 10 FR index score, and patients with 11 - 15 FR index score.

Based on the collected data, we evaluated the differences in mean lipid levels (e.g. total cholesterol, triglycerides, HDL cholesterol, LDL cholesterol, VLDL cholesterol, non-HDL cholesterol, triglycerides/HDL cholesterol, and triglycerides/non-HDL cholesterol) across the groups of patients classified according to their FR index scores. We also determined the correlation between FR index and lipid levels. Subsequently, we assessed the association between severity of FR index and frequencies of patients classified according to the cut-off levels for each lipid variable as well as the association between FR index and demographic variables such as age, gender, BMI, hypertension, diabetes, and smoking.

The collected data were analyzed using the Statistical Package for Social Sciences (SPSS for Windows version 20.0; Chicago, IL, USA). Continuous variables were described by mean and standard deviation. Categorical variables were presented as frequency and percentage. Non-parametric ANOVA (Kruskal-Wallis) test was used to assess the significant differences in mean lipid levels across FR index groups (e.g. difference in mean total cholesterol levels across FR index groups). Non-parametric spearman’s rho correlation test was used to assess the correlation between two continuous variables (e.g. triglycerides and FR index). Non-parametric Pearson Chi-square test was used to assess the association between categorical variables (e.g. gender and FR index groups). In addition, a binary logistic regression analysis was conducted to predict the extent and severity of cornary artery lesions, in terms of FR index, using lipid variables and demographic characteristics as predictors. The P value < 0.05 was considered to indicate statistical significance.

A total of 566 patients were included in the study. The mean age of study participants was 56.17 ± 9.99 years. Of them, 435 (76.9%) were males, 214 (37.8%) had diabetes, 370 (63.4%) had hypertension, and 359 (63.4%) were current smokers. Mean total cholesterol was 182.72 ± 52.47 mg/dL, triglyceride was 152.57 ± 78.21 mg/dL, HDL cholesterol was 35.37 ± 10.25 mg/dL, LDL cholesterol was 116.00 ± 45.11 mg/dL, TG/HDL cholesterol ratio was 4.68 ± 4.66, and TG/non-HDL cholesterol ratio was 1.09 ± 0.60. The mean FR index was 5.40 ± 3.78. Table 1 describes the demographics of patient sample, as well as their distribution according to the FR index.

The mean lipid levels across patients classified according to their FR index have been depicted in Figure 1. Corresponding mean lipid ratios of triglycerides/HDL cholesterol and triglycerides/non-HDL cholesterol across FR index groups are depicted in Figure 2. Analysis of various lipid variables indicated that there were significant differences between total cholesterol, triglycerides, HDL cholesterol, VLDL cholesterol, and triglycerides/non-HDL cholesterol across the groups of patients according to their FR index (Table 2). In addition, a significantly positive correlation was observed between FR index and total cholesterol, triglycerides, VLDL cholesterol, non-HDL cholesterol, triglycerides/HDL cholesterol ratio, and triglycerides/non-HDL cholesterol ratio. On the contrary, an inverse correlation was observed between FR index and HDL cholesterol levels (Table 3).

Analysis of the lipid profile showed that 189 (33.4%) patients had hypercholesterolemia, 194 (34.3%) had high LDL-cholesterolemia, 437 (77.21%) had low HDL-cholesterolemia, 232 (40.1%) had hypertriglyceridemia, 178 (31.5%) had elevated triglycerides/HDL cholesterol ratio, and 317 (56.0%) had elevated triglycerides/non-HDL cholesterol ratio (Table 4). A further analysis of frequency of patients classified according to the cut-off lipid levels revealed that variables like HDL cholesterol levels, triglycerides levels, and triglycerides/non-HDL cholesterol ratio had a significant association with severity of FR index (Table 4).

The association between FR index and demographic variables such as age, gender, BMI, hypertension, diabetes,

^aContinuous variables are described by mean and standard deviation; categorical variables are presented as frequency and percentage.

^aData are presented as mean ± standard deviation; ^bKruskal-Wallis Test.

and smoking is described in Table 5. We found that age or gender played no role in governing the FR index severity. However, characteristics such as BMI, hypertension, diabetes, and smoking showed significant association with FR index.

Table 6 represents the major findings of the binary logistic regression analysis conducted in the study. Among the variables analyzed, presence of diabetes displayed the strongest correlation with FR index (odds ratio: 7.710, P < 0.001). Other variables showing significant correlation with FR index were BMI (odds ratio: 1.106, P =

^aSpearman’s rho correlation test.

^aData are presented as number of patients (percentage of patients); ^bPearson Chi-square test for association.

0.032), status of hypertension (odds ratio: 3.713, P < 0.001), and smoking (odds ratio: 5.784, P < 0.001). On the other hand, age and gender showed a weak insignificant correlation with FR index. Of note, the relationship between FR index and all lipid variables was found to be statistically insignificant.