We ran a double-blind randomized, placebo-controlled study to inspect the endotoxemia, and inflammation levels in 44 CAD participants. This trial was done at Shahid Madani hospital, as cardiology and heart surgery hospital of Tabriz University of Medical Sciences, Tabriz, Iran. Our study was in agreement with the Helsinki Declaration of the World Medical Association (2000) and was accepted by our local ethics committee of Tabriz University of Medical sciences (IR.TBZMED.REC.1397.184) and also was listed in the Iranian Registry of Clinical Trials (IRCT) (IRCT20121028011288N15). All patients admitted to this hospital with a diagnosis of CAD were considered for participation in the study and screened by a cardiologist for eligibility. Patients were selected a few days after the Percutaneous coronary intervention (PCI) and/or conditions were stable. The inclusion criteria were as follows: having CAD; maintaining them during the course of the trial; normal diet; and Body Mass Index (BMI) =25-35 kg/m2. Patients were excluded if they refused to participate or if they had a history of gastrointestinal disorders, if they had thyroid, renal, pancreatic, or liver diseases; if they were lactating or pregnant; if they are taking antibiotics, probiotics, prebiotics, and inflammatory drugs for one month before the intervention or during the intervention. A third party who was blind to the study gave the randomization sequence extracted from allocation software. The contributors, investigators and the medical providers were blinded after assignment to interventions. Informed consent was obtained from all subjects. All 44 contributors completed the study and none of them dropped their participation (Fig 1. CONSORT Flow diagram).
To calculate sample size, mean (standard deviation; SD) of LPS was used from a previous clinical trial on the effect of prebiotic on endotoxmia of obese patients . Calculated based on a conﬁdence interval (CI) of 95%, and power of 80% in two-sided tests using Power analysis and sample size software (PASS; NCSS, LLC, US) version 15, the sample size was 20 per group, which was increased to 22, considering a probable 10% dropout rate
A blind trial at two levels (researcher, participants), placebo-controlled, randomized, clinical design was used to examine the effect of probiotic supplementation on endotoxemia in patients with CAD. Each qualified contributor was randomly allocated into probiotic or placebo group, according to 1:1 equal proportion rule. The order of random allocation was made by random sequence software. The random numbers were kept by a free person not complicated in the assessment of the patient, or in the data collection and analysis. In the current study, patients were randomly assigned into two groups to receive whichever probiotic supplements (n=22) or placebo (n=22) for 3 months. Both participants and researchers were unaware of the treatment allocation.
All the participants received a moderate calorie restricted dietary plan during 12 weeks’ intervention period. In this study, the program was designed to enable weight loss of 7–10% of weight, at a rate of 0.5–1 kg/wk throughout the intervention. Calorie intake was planned based on individual features of the participants and with the aim of daily energy restriction (500 kcal fewer than the total energy requirements [TEE] estimated by Mifflin–St Jeor equation). The diets were provided 55 ~ 60% of TEE from carbohydrate, 10 ~ 15% from protein, and 25 ~ 35% from fat.
Patients in the probiotic group received one probiotic capsule daily containing a Lactobacillus rhamnosus (LGG)) 1.6 ×109 colony-forming unit (CFU) with their lunch. In the placebo (control) group, the capsules contained maltodextrin (Tak Gen Zist Pharmaceutical Company, Tehran, Iran). The physical properties of the placebo were identical in terms of shape, color, size, packaging, and smell but contained no bacteria. Phone contacts were made to ensure adherence twice a month. Compliance to supplementation was established by requesting participants to return the medication containers. The remaining capsules were counted and subtracted from the number provided to determine the number taken. To increase the compliance, all subjects received short messages on their cell phones reminding them to take the supplements every week. For ethical issues, the subjects were allowable to take their routine medications. However, taking any antioxidants and/or vitamin supplements were prohibited during the trial. The participants were allowed to discontinue the trial if they were unwilling to complete or experience any adverse effect during the intervention. Adverse effect evaluation was complete during the trial by questioning the participants and assessments for any adverse effect related to the intervention. At the beginning of the study, participants were recommended not to change their level of physical activity.
Dietary intake was assessed using a dietary record at month 0, 1 and 3 of the intervention. We used Nutritionist IV software adjusted for Iranian diets to acquire nutrient intakes of participants based on the average of three-day food diaries.
Physical Activity Assessment
The physical activity assessment was gotten to monitor patient’s usual physical activity levels throughout the study. The validated short-form International Physical Activity Questionnaire (IPAQ) was used to measure the participant’s physical activity. Based on previous studies, physical activities were classified as low, moderate, and high.
Assessment of anthropometric indices
Body weight was assessed via a scale with 250 gr accuracy (Seca, Hamburg, Germany) and patients were measured while wearing a minimum dress and without shoes. Height without shoes was measured by a tape with 0.5-cm accuracy. BMI was computed by dividing weight (Kg) by height2 (m2). To avoid measurement bias, all measurements were taken by a trained dietitian.
After an overnight fasting (12 hours), blood samples were taken by venipuncture of the antecubital vein, using vacuum tubes. Blood serum was obtained from whole blood through centrifugation at 2500 rpm for 10 min. Fasting blood sugar (FBS) and lipid profile (including High density lipoprotein cholesterol (HDL-C), Low density lipoprotein cholesterol (LDL-C), Triglycerides (TG), and Total cholesterol (TC)) were examined using commercial kits (Pars Azmoon Inc., Tehran, Iran) and a Selectra Pro M autoanalyzer (Vital Scientific, Spankeren, the Netherlands). Endotoxin, and inflammatory markers were measured using enzyme-linked immunosorbent assay  kits as follows: Interleukin 1 beta (IL-1β) (intra-assay variation = 5.8%, normal range = 0.02-6 pg/mL, inter-assay variation = 9.06%, sensitivity = 0.01 pg/mL), Toll-Like Receptor 4 (TLR-4) (normal range = 0.05-15 ng/mL, intra-assay variation = 4.58%, inter-assay variation = 7.8%, sensitivity = 0.027 ng/mL), IL-10 (normal range = 0.2-100 ng/dl, intra-assay variation = 10%), and, LPS (inter-assay CV = 10.0% , detectable range = 12.00 – 1000 ng/ml). Sera Trimethylamine-N-oxide (TMAO), transforming growth factor beta (TGF-β), and matrix metallopeptidase 9 (MMP-9) also were measured via an enzyme-linked immunosorbent assay (ELISA) (Crystal Day, China) as stated by the manufacturer’s information. Serum malondialehyde (MDA) was measured based on reaction with Thiobarbituric Acid (TBA). The number of Bacteroidetes and Firmicutes was measured by Real-time PCR (SYBR green method).
The data were analyzed via SPSS software (version 21; SPSS Inc., Chicago, IL) and the outcomes were stated as mean ± SD. To determine the normal distribution of variables, we used skewness and kurtosis test. Paired samples t-test was applied for within-group comparisons (end-point vs. baseline). Afterward adjusting for the confounders (weight, PAL, and calorie intakes) and baseline levels, we did analysis of covariance (ANCOVA) in which the confounding effect of these variables were taken into account which was used to determine the statistically significant pairwise differences. The analyses were conducted using an intention-to-treat approach . Circulating endotoxin level was intended as a primary outcome, while inﬂammatory markers were defined as secondary outcomes. For all statistical tests, a P value less than 0.05 was interpreted as statistically significant.