Skip to main content
Evidence-Based Supplement Research
Evidence-Based Supplement Research

Study Design

Type
Meta-Analysis
Sample size
n = 506
Population
type 2 diabetes patients
Methods
systematic review of PubMed, Scopus, Embase; manual search of reference lists; two reviewers extracted data and assessed risk of bias using Cochrane tool; random effects model for meta-analysis

Introduction

Hyperlipidemia, a prevalent comorbidity among type 2 diabetes patients, is a potential risk factor for cardiovascular diseases. It is unclear whether cocoa has beneficial impacts on the serum lipids of patients with diabetes.

Methods

PubMed, Scopus, and Embase databases were systematically reviewed for clinical trials on cocoa intake and blood lipids in type 2 diabetes until January 1, 2024, and the reference list of relevant articles was searched manually. Two reviewers extracted data and determined the risk of bias (RoB) using the Cochrane tool. The random effect model was applied to calculate standardized mean differences (SMDs). Finally, the certainty and clinical importance of the evidence were checked (PROSPERO registration code: CRD42021224931).

Result

Eleven RCTs with 506 participants were included. Different forms of cocoa and various intervention durations were applied. Only two RCTs had a low RoB. Findings showed a significant reduction in serum triglyceride (SMD: - 0.57, 95% CI - 1.05, - 0.10, I2: 82.7%), but not in other blood lipids. There was a severe‎ heterogeneity in results justified with discrepancies in age, designs, durations, interventions, body mass index, baseline blood lipids, and risk of bias. The results showed low certainty and unimportant lipid changes.

Conclusion

Although cocoa may slightly change serum lipids in diabetes, its recommendation for lipids control has fair clinical benefits. Due to the lack of certainty of findings and an inadequate number of studies, further well-designed trials considering possible sources of heterogeneity with low RoB are highly recommended.

Research Insights

Back to top