Seventeen RCTs were identified (8,961 participants, range 315 to 1,306). Only 13 studies were included in the meta-analysis as four studies did not report dispersion measures.
DPP-4 inhibitor versus controls significantly reduced total cholesterol (WMD -0.18 mmol/L, 95% CI -0.29 to -0.06; 13 studies) and triglycerides (WMD -0.15 mmol/L, 95% CI -0.26 to -0.03; 10 studies) but not HDL cholesterol (13 studies).
Subgroup analyses supported the main findings.
Meta-regression for the placebo-controlled trials found significant correlations that greater cholesterol reduction was associated with younger patients, lower diabetes duration, lower BMI, higher baseline cholesterol and higher HbA1c.
The authors stated that there was evidence of major publication bias.