**2.1. Study selection**

A comprehensive electronic search of the MEDLINE (National Library of Medicine) database was conducted to identify English-written studies published between January 1, 2004, and January 1, 2015, using ("Atherosclerosis"[Mesh]) AND "Rabbits"[Mesh] AND (Ani‐ mals[Mesh:noexp]) in PubMed as Medical Subject Headings (MeSH). Two investigators (IK and SSM) coincidentally assessed titles and abstracts of articles generated by the literature research after removal of duplicated records. Discrepancies in opinion between the two investigators were adjudicated by a third investigator (GB). We considered the following criteria in the selected studies for meta-analysis: (1) inclusion of cholesterol in diet was the main part of the dietary intervention; (2) comparison groups could include a control diet or an intervention in which cholesterol intake was quantitatively different from cholesterol intake in the experimental groups; (3) among pharmacological studies, only results of the control group and the cholesterol-fed group were selected, and the rest of the drug-treated groups were excluded; (4) we included studies that reported net quantity of lipid and lipoprotein profiles and studies that reported changes graphically have been excluded; (5) studies that solely considered other end points such as inflammatory end points of atherosclerosis, histopathologic descriptive end points, and vascular imaging of atherosclerosis were excluded; (6) all studies that reported dyslipidemia in genetic engineered models and strains were excluded; and (7) all studies that focused on lipid, and lipid profiles of cholesterol-fed New Zealand rabbits (NZRs) were included to do meta-analysis over their results. In this continu‐ um, 32 articles met the aforementioned eligibility criteria for inclusion in meta-analysis [10– 41]. Figure 1 summarizes information on the study selection process. To show substantial heterogeneity among studies, we report fixed-, random-, and mixedeffects meta-analysis [43]. In this continuum, random-effects meta-analysis takes into account the precision of discrete studies and the variation among studies and weights of each study accordingly.

Figure 1. Flow diagram of publication inclusion. **Figure 1.** Flow diagram of publication inclusion.

causes in the etiopathogenesis of atherosclerosis [3]. However, all aspects of atherosclerosis are not known at the present time, and more human and animal studies are requested to decode the black box of atherosclerosis. Since we are not able to do interventional studies in humans,

Many seminal reviews (e.g., see [4–6]) have shown that rabbits are good, reliable, and cheap animal models of atherosclerosis, with a high degree of comparability to human's atheroscle‐ rosis. In this context, low-density lipoprotein cholesterol (LDL-C) is predominant plasma

Despite these data, the amount of dietary cholesterol, the diet formula, and the duration of cholesterol feeding necessary to induce translated atherosclerosis are not conclusive among experimental studies (e.g., see [7,8]). In this line, Prof. Watanabe and colleagues [9] partially solved this problem by producing a genetic rabbit model of hypercholesterolemia many years ago. However, this expensive model is not globally available in all laboratories and is not suitable for translating nutritional interventions that lead to atherosclerosis. Watanabe rabbit

In addition, there is not any concert about the methodology used for inducing atherosclerosis in nongenetic animal models like New Zealand rabbits (NZRs) among various studies. In this sense, reliable and loyal translation of atherosclerosis in rabbits were hampered by methodo‐ logical limitations, including suitable sample size, suboptimal biomarker assay, amount of cholesterol intake, routes of cholesterol intake, and duration of cholesterol intake, among others. The aforementioned limitations result in low statistical power to translate atheroscle‐

To obtain a more truthful and defined estimation of association between cholesterol intake and atherosclerosis (i.e., the occurrence of dyslipidemia), to do root cause analysis, and to explore the source of heterogeneity among randomized controlled trials (RCTs), we conducted a metaanalysis of RCTs by evaluating the associations of dietary cholesterol with the dyslipidemia

A comprehensive electronic search of the MEDLINE (National Library of Medicine) database was conducted to identify English-written studies published between January 1, 2004, and January 1, 2015, using ("Atherosclerosis"[Mesh]) AND "Rabbits"[Mesh] AND (Ani‐ mals[Mesh:noexp]) in PubMed as Medical Subject Headings (MeSH). Two investigators (IK and SSM) coincidentally assessed titles and abstracts of articles generated by the literature research after removal of duplicated records. Discrepancies in opinion between the two investigators were adjudicated by a third investigator (GB). We considered the following

animal models are good simulated and translated tools in this endeavor.

is still not a pet animal model for studying atherosclerosis in all laboratories.

lipoprotein in rabbits and humans [6].

4 Lipoproteins - From Bench to Bedside

rosis in animal models like rabbits.

component of atherosclerosis in rabbits.

**2. Materials and methods**

**2.1. Study selection**
