Strongest evidence: High-strength evidence supports a moderate effect of L-Carnitine (2–3 g/day) on reducing C-reactive protein (CRP) levels (5 of 6 studies beneficial) and a small effect on reducing triglycerides (5 of 6 studies) and fasting blood glucose (5 of 5 studies). Moderate-strength evidence shows consistent small-to-moderate reductions in body weight (4 of 4 studies), HOMA-IR (3 of 3), and TNF-α (3 of 3), typically at doses of 1–3 g/day over 6–8 weeks. Additionally, moderate evidence supports small reductions in LDL cholesterol, hemoglobin A1c, and body mass index, as well as increases in serum albumin and total protein in critically ill patients.
Mixed or weaker evidence: No outcomes with low or very low evidence strength were identified in this synthesis. However, several moderate-strength outcomes show population-specific variability: triglyceride reduction was neutral in a meta-analysis of hemodialysis patients, and BMI reduction was neutral in 2 of 6 studies (hemodialysis and type 2 diabetes subgroups). The effect on total cholesterol was mixed (small to large), with one neutral meta-analysis in type 2 diabetes.
Effective dose patterns: Across all cardiovascular and metabolic outcomes, the most commonly studied effective dose is 2–3 g/day. For glycemic and anthropometric outcomes (fasting glucose, HbA1c, BMI, body weight), doses of 1–2 g/day also show consistent benefit. The median study duration across outcomes is 42–56 days, suggesting benefits appear within 6–8 weeks.
Population insights: The strongest effects are consistently seen in clinical populations with underlying metabolic dysfunction, including adults with PCOS, type 2 diabetes, impaired glucose tolerance, overweight/obesity, and critically ill patients. Benefits in healthy individuals or in certain subgroup conditions (hemodialysis, NAFLD) are less consistent or absent. Publication bias is flagged as a likely issue across multiple outcomes, meaning null results may be underrepresented in the published literature.
Notable caveats: Across syntheses, key caveats include: (1) publication bias is suspected for most positively reported outcomes, (2) many trials are short-term (median 42–56 days) with unknown long-term effects, (3) supplement form (L-carnitine vs. acetyl-L-carnitine) was rarely specified or controlled, which may influence outcomes, and (4) high heterogeneity (I² up to 94%) was observed in several meta-analyses, indicating variable effects across populations.