Introduction

Metabolic syndrome (MetS) is an asymptomatic disorder

that includes a cluster of metabolic abnormalities associated with obesity, hyperlipidemia, hypertension, and insulin resistance (Alberti et al., 2009). The causative factors of

MetS are central obesity and insulin resistance, which lead

to cardiovascular diseases (CVDs), diabetes, and stroke

(Srikanthan et al., 2016). Oxidative stress and inflammation

also contribute to the etiology of MetS (Soares and Costa,

2009). Metabolic markers such as triglyceride levels, highdensity lipoprotein-cholesterol (HDL-C), low-density

lipoprotein-cholesterol (LDL-C), hypertension, blood

pressure, obesity, insulin, and oxidative stress are the criteria used to diagnose MetS. This non-communicable disease has become a significant major cause of mortality

worldwide and increases the mortality rate of patients with

type 2 diabetes and CVDs, coronary heart disease, and

stroke (Ford, 2004). The American Heart Association

reported that about 35% of adults and 50% of 60 years

older in the US have MetS (Aguilar et al., 2015). The

International Diabetes Federation stated that nearly 25% of

the world’s population suffers from MetS (O’neill and

O’driscoll, 2015). However, the prevalence varies by age,

ethnicity, gender, and variation in the definition of MetS.

Based on the International Diabetes Federation definition,

the eastern country of Tunisia showed a MetS prevalence

of 45.5%; in Iran, this value was 37.4% (Delavari et al.,

2009).

Data extraction

In the initial search, two researchers (HL and EJ) independently reviewed the title and abstracts of the articles

under the PICOS framework. Next, descriptive data

screened based on full-text articles were assessed for eligibility. A standard form included the following information from the selected articles: bibliographic details, study

design, study origin, participants’ health status, age, sex,

body mass index, groups description, a form of wheat

germ, intervention period, washout period, dose amount,

intake direction, physical and dietary intake details during

an intervention, functionality of wheat germ, biomarker

readings at baseline and post-intervention, outcomes measures, statistical results, compliance, and dropout rate.

There were insufficient data on dichotomous outcomes

in the included studies. To utilize the available data in a

meta-analysis, we included data for three metabolic

markers (cholesterol, triglycerides, and glucose) in the

meta-analysis as continuous outcomes.

Quality assessment

The quality of the selected trials was measured by

Cochrane Collaboration’s tool to evaluate the risk of bias in

the randomized trials (Higgins et al., 2011). The bias tools

have the following respective domains: random sequence

generation (selection bias), allocation concealment (selection bias), blinding of participants and personnel (performance bias), blinding of outcome assessment (detection

bias), incomplete outcome data (attribution bias), selective

reporting (reporting bias), and other sources of bias. Each

domain was rated as a low, high, and unclear risk. If at least

one of the domains showed a high or unclear risk, we

classified the overall result as a high or unclear risk,

respectively. The overall evaluated result was considered

as low risk if all domains showed a low risk in the

respective study.

Statistical analysis

To conduct the meta-analysis, we used the review manager

(RevMan) version 5.3 (Collaboration, 2016). Data in the

included articles were continuous outcomes within the

studies related to different metabolic markers. In the analytical method, we analyzed the random effects model by

DerSimonian and Laird methods (DerSimonian and Laird,

1986). Follow-up from baseline in the experimental group

was compared to that in the control group using the standard mean difference (SMD) as a primary effective

measure. To identify the parametric relationship between

the intervention group (wheat germ) and control group, we

calculated the inverse of variance (IV) as the study weight

in analysis and 95% confidence intervals (CIs) among the

categories of metabolic markers. To more precisely

examine the effect of cholesterol, we stratified cholesterol

into subgroups: HDL-C and LDL-C.

Results and discussion

Studies included in the analysis

The detailed search strategy was performed, as shown in

the PRISMA flow chart (Moher et al., 2009) (Fig. 1). We

initially identified 14,888 studies in the three different

databases, with 9776, 2705, and 2407 articles from

PubMed, Google Scholar, and Web of Science, respectively. All references from these databases were imported

to an Endnote library. After deleting duplicate references

using EndNote x7, 8611 studies remained. Next, 2823 fulltext articles remained after eliminating abstract, proceeding, and review papers. Forty-three articles were further

reviewed after eliminating 2780 studies that failed to meet

the inclusion criteria. In the preparatory mapping review,

we tested many studies that demonstrated the health outcomes of wheat germ consumption. Most of these studies

were dropped out because of unrelated functionality and

study design.