The present cross-sectional study was approved by the ethics committee of Isfahan University of Medical Sciences (Ethical approval code: IR.MUI.RESEARCH.REC.1400.200), Iran and was part of a larger research that its protocol was approved by the ethics committee of Shahid Sadoughi University of Medical Sciences, Yazd, Iran. Patients aged 25–75 years with increasing chest pain (angina) or acute-onset angina, who were identified as candidates for angiography after the initial examination (exercise stress testing or electrocardiogram (ECG) or computed tomography (CT) scan) by a cardiologist, were included in the study. Eventually, among the patients admitted under diagnostic coronary angiography in Afshar Hospital in Yazd, based on inclusion and exclusion criteria, 463 patients were recruited. Patients with the following criteria were not included in the study: (1) history of cancer, heart failure, heart attack, percutaneous coronary intervention (PCI), coronary artery bypass grafting (CABG), chronic kidney disease stage 3 and above, specific liver disease or receiving medication for liver disorders, immunodeficiency, AIDS; (2) people with severe obesity (body mass index (BMI) above 40); (3) pregnant and lactating women; (4) people who for any reason have limited food intake by mouth; (5) have a special diet. Non-response to many of food frequency questionnaire items and not detecting the type of genotype led to the patient's exclusion from the study. Finally, data from 428 persons were analyzed (Supplementary material: participant flowchart). Written informed consent was taken from eligible persons. The study was carried out per the Declaration of Helsinki.

The severity of CAD was determined by Gensini and SYNTAX scores. For this purpose, the coronary angiogram was interpreted by an experienced cardiologist blinded for demographic and clinical data except for sex and age. Gensini and SYNTAX scores were computed randomly for several patients by the second cardiologist. The Gensini score calculation starts by allocating a severity score to each identified coronary stenosis as follows: 1 point for ≤25% narrowing, 2 points for 26–50% narrowing, 4 points for 51–75% narrowing, 8 points for 76–90% narrowing, 16 points for 91–99% narrowing, and 32 points for total occlusion (100%). After that, each lesion score is multiplied by a factor that takes into account the importance of the coronary arteries and the lesion's position in the coronary circulation (5 for the left main coronary artery, 2.5 for the proximal segment of the left anterior descending coronary artery, 2.5 for the proximal segment of the circumflex artery, 1.5 for the mid-segment of the left anterior descending coronary artery, 1.0 for the right coronary artery, the distal segment of the left anterior descending coronary artery, the posterolateral artery, and the obtuse marginal artery, and 0.5 for other segments). Finally, the Gensini score was acquired by summating of the coronary segment scores. A higher Gensini score demonstrates a more intensive disease (18–20). The patients were categorized into two groups based on Gensini score: low Gensini score (< 20) and intermediate-high Gensini score (≥ 20) (21).

The SYNTAX score was computed through the internet-based SYNTAX calculator version 2.0 (www.syntaxscore.com). SYNTAX score algorithm containing consecutive and interactional self-guided questions focusing on functional and anatomical parameters of the lesions with ≥50 % stenosis in arteries with a diameter of ≥1.5 mm. The final SYNTAX score was acquired by summation of all lesion scores. The participants were classified into two groups based on SYNTAX score: low SYNTAX score (<23) and intermediate-high SYNTAX score (≥23). A higher SYNTAX score represents a more intensive disease (22, 23).

The usual dietary intakes of patients during the past year were evaluated using the valid and reliable 182-item semi-quantitative food frequency questionnaire (FFQ) and face-to-face interviews with trained nutritionists. The FFQ used in this study was a modified version of a 178-item FFQ, which was previously validated to assess the dietary intake of Yazidis adults (24). Patients were asked to report the frequency (number of times per month, week, or day the food was consumed) and the amount of each food item that was consumed every time in the past year (portion size based on the standard serving sizes commonly consumed by Iranians). The reported values of each food item were converted to g/day using household measures of consumed foods. Then, nutrient intake was calculated using Nutritionist IV software (25).

In this study, a nutritionist measured weight using a portable digital scale and the body analyzer (Omron Inc. Osaka, Japan), with an accuracy of 0.1 kg, with minimal coverage and without shoes.. Height was measured with an accuracy of 0.1 cm, using a wall-fixed measuring tape in a standing position with shoulders in normal alignment and no shoes. BMI was calculated as body weight (kg) divided by height squared (m²). Waist circumference was assessed by a flexible inelastic tape measure (i.e., the tape measure should not stretch when taking the measurement) in the standing position to the nearest 1 cm. The narrowest area between the iliac crest and the last rib was measured (26). The percentage of muscle mass, total body and visceral fat were also assessed using a body composition analyzer (Omron, Japan model: BF511) (27). Blood pressure was also measured using a bedside automated monitor, and the patient's average blood pressure during the day was recorded by nurses working at the cardiovascular care unit (CCU) before patients underwent angiography.

To determine the levels of triglyceride (TG), TC, HDL-C and LDL, after 10–12 h of overnight fasting, a 4 ml blood sample was drawn from the patients. 2 ml of blood samples were centrifuged at 2,500 rpm for 3 mins to separate the serum from the blood cells. Buffy coats and remaining whole blood samples were stored at −80°C for DNA extraction and other biochemical tests. TG, TC, LDL-C and HDL-C concentration are measured using Pars Azmon company kits in the Yazd School of Health laboratory. Then, lipid profile were categorized into normal TG (<150 mg/dl) or high TG (≥150 mg/dl), normal TC (<200 mg/dl) or high TC (≥200 mg/dl), normal LDL-C (<130 mg/dl) or high LDL-C (≥130 mg/dl), normal HDL-C (≥40 mg/dl for men and ≥50 mg/dl for women) or low HDL (<40 mg/dl for men and <50 mg/dl for women), per the Adult Treatment Panel III (28) and the 2013 American College of Cardiology/American Heart Association guidelines (29).

DNA samples were isolated from the white blood cell genome of the complete blood sample of the participants using the SimBiolab Blood Kit, according to the manufacturer's protocol. The rs662 polymorphism (major allele: Q, minor allele: R), a fragment of 520 base pairs (bp) in exon 6 of the PON1 gene, was genotyped by the polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) method. The PCR mixture was provided in a total volume of 20 μl containing 2 μl of genomic DNA, 10 μl of Master Mix (Amplicon, Denmark), 6 μl of water and 1 μl (10 pmol) of each oligonucleotide primer. Forward and reverse primer consists of AAACCCAAATACATCTCCCAGAAT and GCTCCATCCCACATCTTGATTTTA, respectively. PCR is performed by repeating three steps. First, DNA templates were denatured at 95°C for 5 min; amplification consisted of 45 cycles at 95°C for 15 s, annealing at 60°C for 30 s, extension at 72°C for 30 s, with a final extension at 72°C for 5 min. Amplified DNA (10 ml) was digested with 5 U restriction enzyme HinfI (Fermentase, Germany) at 37°C, overnight. All products were visualized by electrophoresis in 2% agarose gel (SinaClon, Iran) at 90 V for 2.5 h.

General demographic data including age, gender, smoking status, multivitamin intake, vitamin C and E supplements intake, the medication used, and medical history were collected using valid and reliable questionnaires. Physical activity was assessed using International Physical Activity Questionnaire (IPAQ). Physical activity level was calculated based on metabolic equivalent task—minutes per week (30). Persian translation validation of IPAQ has previously been confirmed by Moghaddam et al. (31).

Subjects were divided into two genotype groups: R-allele carriers (QR/RR) and non-carriers (QQ). Pearson's chi-square test was used for the Hardy-Weinberg Equilibrium (HEW). Vitamin C and E intake was categorized into two categories based on median intake (higher and lower than the median). Gensini score, SYNTAX score, TG, TC, LDL-C and HDL-C were also categorized into two categories. Continuous and categorical variables were expressed as mean ± standard deviation (SD) and frequencies (percentages), respectively. The normality distribution of continuous variables was checked by the Kolmogorov- Smirnov test. Chi-squared test and independent samples t-test were used for comparing baseline categorical and continuous variables between genotypes (QQ, QR + RR) and categories of vitamin intake, respectively. The Gensini score, SYNTAX score, HDL-c, LDL-c, TC and TG as both categorical and continuous response variables based on conducted statistical analysis were treated. The covariance (ANCOVA) test under the generalized linear model (GLM) framework was used to analyze the independent and interaction effects of PON1 rs662 genotypes and vitamin intake on Gensini and SYNTAX scores and lipid profile. The adjusted means were estimated after controlling for age, gender, total energy intake, physical activity, smoking status, alcohol consumption, multivitamin intake, vitamin C and E supplements intake, BMI and medication used (antihypertension drugs, antidiabetic drugs and antihyperlipidemic drugs). Logistic regression was used to examine the joint effect of vitamin C/E intake and rs662 genotypes (QQ and QR/RR) for predicting being at higher risk of CAD and lipid disorders. The association analyses were done in adjusted logistic regression models and adjustment was made for age, gender, total energy intake, physical activity, smoking status, alcohol consumption, BMI, multivitamin intake, vitamin C and E supplements intake and medication used (antihypertension drugs, antidiabetic drugs, and antihyperlipidemic drugs). The analyses were performed using SPSS software version 24 (IBM Corp., Armonk, NY, USA). P ≤ 0.05 were considered significant.

Data from 428 patients (aged 25–75 years) were analyzed. The prevalence of PON1 rs662 polymorphism genotypes was QQ (47.4%), QR + RR (52.6%). Table 1 presents the characteristics and dietary intake of study participants according to PON1 genotypes and vitamin C and E intake. The mean intake of energy and fat was higher in R-allele carriers, while the average carbohydrate intake was lower than in the non-carriers. Age, BMI, gender, physical activity, medication used (antihypertension and antihyperlipidemic drugs), smoking status, alcohol consumption, mean energy intake, carbohydrate, fat, cholesterol and vitamin C supplement intake was different between the two categories of vitamin E intake. Vitamin E consumption was higher in younger people with higher physical activity. On the other hand, vitamin E intake was higher in men than in women. Also, the average energy and fat intake was higher in people with higher vitamin E intake. On the other hand, BMI, gender, smoking, alcohol consumption, and dietary intake were significantly different in the two categories of vitamin C intake. In people with a higher intake of vitamin C compared with individuals who received less than the median, the consumption of energy, carbohydrates, and folate was higher, while the cholesterol and saturated fatty acids intake were lower.

Table 2 shows the mean values of Gensini and SYNTAX scores and lipid profile across PON1 genotypes (QQ and QR/RR) based on higher and lower than median vitamin E and C intake. After adjustment for age, gender, total energy intake, physical activity, smoking status, alcohol consumption, multivitamin intake, vitamin C and E supplement intake, BMI, and medication used, R allele carriers (RR + RQ) have lower HDL-C levels than non-carriers (QQ) (P ≤ 0.05). On the other hand, a higher vitamin C intake was associated with lower LDL-C (P = 0.007) and HDL-C levels (P = 0.010). Vitamin E intake was not significantly associated with any studied variables (P ≥ 0.05).

The odds of the association of vitamin C and E intake, PON1 rs662 genotypes (RR+RQ vs. QQ), and the interaction of the two variables with CAD severity and lipid disorders based on adjusted logistic regression are reported in Table 3. After adjustment for potential confounders, higher consumption of vitamin C was associated with a reduced risk of high TC (OR: 0.42, 95% CI 0.23–0.75, P = 0.003) and LDL-C (OR: 0.49, 95% CI 0.25–0.96, P = 0.038) and an increased risk of low HDL-C (OR: 1.88, 95% CI 1.03–3.42, P = 0.037). Furthermore, a significant interaction was observed between vitamin C intake and genotypes of rs66 polymorphism on LDL-C (P = 0.050) (Figure 1). In detail, the R-allele carriers with lower vitamin C intake had higher LDL-C levels than QQ genotype carriers. No significant interaction was found between vitamin E intake and rs662 polymorphism genotypes on the Gensini and SYNTAX scores and lipid profile (P > 0.05).