Purpose: The actual effects of L-carnitine administration on leptin serum level is inconsistent. In order to assess the efficacy of L-carnitine supplementation on serum leptin we conducted a meta-analysis of randomized controlled trials (RCTs).
Methods: Seven studies with 325 cases and 330 controls were included. The pooled weighted mean difference (WMD) was calculated by random-effects model. The heterogeneity across studies was evaluated by using Cochrane's Q and I2 tests. In addition, we carried out the metaninf command to test the effect of each individual study on the overall result.
Results: L-carnitine supplementation seemed to have no significant effect on serum leptin concentrations (WMD: -0.565 ng/mL; 95% CI: -2.417 to 1.287, p = 0.550). However, between-study heterogeneity was higher across all studies (I2 = 84.3%, p < 0.0001). Subgroup analysis to find the sources of heterogeneity showed that L-carnitine dosage (g) ( < 2 g: I2 = 00.0%, p = 0.408), and study population (diabetes: I2 = 46.7%, p = 0.153, and non-diabetes: I2 = 15.1%, p = 0.317) were the potential sources of heterogeneity. Besides, a more significant reduction in serum leptin concentration was observed with a daily dose of ≥ 2 mg L-carnitine (WMD: -2.742 ng/mL; 95% CI: -3.039 to -2.444, p < 0.001), in diabetic patients (WMD: -2.946 ng/mL; 95% CI: -3.254 to -2.638, p < 0.001), and with intervention duration <12 weeks (WMD: -2.772 ng/mL; 95% CI: -3.073 to -2.471, p < 0.001).
Conclusion: L-carnitine consumption does not reduce serum leptin significantly. However, a significant effect on leptin was observed in diabetic patients and patients who received doses more than 3 mg per day in the course of <12 weeks.
Keywords: L-carnitine; Leptin; Meta-analysis.