The participants were patients who referred to coronary angiography in the Shahid Rajaei Cardiovascular, Medical & Research Center in Tehran, Iran. Eligible subjects were men and post-menopausal women less than 75 years who had at least one of the major cardiovascular risk factors including hypertension, diabetes mellitus, dyslipidemia, or acute cardiac events. None of the subjects had taken anti-inflammatory medication or dietary antioxidants or omega-3 supplements during the previous month before the study. Participants with low adherence to the intervention (who consumed less than 80% of the olive or canola oils delivered at the baseline), any changes in the disease treatment plan including type or dose of medications or coronary artery bypass graft (CABG) and gastrointestinal complications such as diarrhea were excluded from the study.
The present study was a randomized, controlled, parallel-arm, clinical trial that was conducted in the spring and summer 2019 in Tehran, Iran. At baseline, demographic and medical information were obtained through face-to-face interviews and study of medical files, respectively. Next, the participants were randomly assigned to one of the two groups following simple randomization procedures using computerized random numbers and were requested to consume a daily amount of 25 mL of refined olive oil (OO) (Etka, Iran) or canola oil (CO) (CanaPlus, Canada) as raw with meals for 6 weeks. Olive and canola oils were provided to patients in sufficient quantities. To ensure compliance with the intervention and proper oil consumption, participants were followed up weekly by telephone contact.
The procedures followed in this trial were in accordance with the 1964 Helsinki Declaration and the study protocol was approved by the Ethical Committee of National Nutrition & Food Technology Research Institute, Tehran, Iran (The ethical committee No. IR.SBMU.NNFTRI.REC.1398.074) and written informed consent were obtained from all the participants. This clinical trial was registered at the Iranian Registry Center of Clinical Trials (IRCT) (registration number: 20160702028742N5).
Weight and height were measured at the baseline and after 6 weeks of intervention by the study dietitian. Weight was measured without shoes, coats, or jackets using a digital scale. Height was measured without shoes using a wall-mounted stadiometer.
Venous blood samples were obtained from each patient after 12-hr overnight fasting and collected into heparinized tubes at baseline and after 6 weeks. The blood samples were centrifuged (4000 rpm for 20 min) and the resulting plasma was stored at -80°C. Plasma Lp-PLA2 mass and activity were analyzed by a commercially available ELISA kit (ZellBio, Germany) and commercial colorimetric assay kit (Cayman Chemical Co.), respectively. Commercially available ELISA kit (BioLegend, USA) was used to measure plasma IL-6 concentration. Plasma complement C3 and C4 were determined by the turbidimetric method and lipid profile by the colorimetric method by commercial kits (Pars Azmoon, Iran) using an auto-analyzer (Selectra 2, Vital Scientific, Spankeren, The Netherlands).
Dietary intakes and physical activity
Dietary intake and physical activity levels were monitored at baseline and after 6 weeks. Dietary intakes were assessed using the 24-hr dietary recall questionnaire completed in three days (two regular days in the middle of the week and one day at the weekend) by the trained dietitian. Participants were asked to maintain their habitual lifestyle throughout the study. Recall data was analyzed using the Nutritionist software (version IV, N-Squared Computing, San Bruno, CA, USA) to which was added the local food data.
Although our primary outcome was Lp-PLA2, we could not calculate sample size based on this variable because according to our search, there was no study comparing the effects of CO and OO on Lp-PLA2 mass or activity. On the other hand, studies have shown that Lp-PLA2 is transported in plasma predominantly (80%) is associated with LDL-C . Therefore, the study sample size was calculated using LDL-C as the main outcome variable. To detect a change in the mean of LDL-C concentration (10 mg/dL) as reported in a previous investigation  at the 5% level of significance and with 80% power, 24 participants were needed in each arm of the two-arm trial.
The data were analyzed using the SPSS software for Windows version 21 (SPSS Inc., Chicago, IL, USA). All values were reported as Mean ± SD or percentage (%). The per-protocol analysis was performed (i.e. only those who completed the study were included in the analyses). The normality of distribution of the study variables was tested by the Shapiro-Wilk test. When the variables were not normally distributed, raw values were log-transformed. Analysis of covariance (ANCOVA) was used to compare the 6-week values between the groups using the baseline measures as the covariate. Paired samples t‐test was used for comparing the measurements in the beginning and at the end of the intervention within the study groups. The χ2 test was used for the comparison of categorical variables. The statistical significance level was set at p = 0.05 (two tails).