Association of Serum Leptin with All-Cause and Disease Specific Mortality: A Meta-Analysis of Prospective Observational Studies.

Leptin levels may affect mortality through its link to inflammation and obesity. However, data are inconclusive. This meta-analysis was performed to investigate the association of leptin with mortality. The PubMed, Scopus, Google scholar, and reference lists of the included studies were searched for all published prospective observational studies that described the associations of serum leptin and mortality up to August 2017. Two reviewers independently assessed all potentially relevant studies for inclusion and methodological quality using standardized abstraction forms. Of 518 studies identified, 19 studies considered the association of leptin with all-cause mortality (16 208 subjects) and 12 studies evaluated the association of leptin with disease-specific mortality (13 680 subjects). In the overall analysis, no significant association was found between leptin and all-cause mortality (HR=1.028; 95% CI: 0.908-1.165; p=0.659), with a high between study heterogeneity (p˂0.001, I2 =61). Subgroup analysis found that high leptin is associated with increased all-cause mortality in healthy people (HR=1.159 (95% CI, 1.032, 1.302), p=0.012), in men (HR=1.162; 95% CI: 1.036-1.302; p=0.010), and in subjects with ≥60 years old (HR=1.129; 95% CI: 1.030-1.238; p=0.010); whereas, increased leptin levels was related to the decreased cancer-specific mortality ((HR=0.550; 95% CI: 0.418-0.724), p˂0.001). Higher leptin levels in healthy people, old people, and males were associated with increased all-cause mortality, while increased leptin level was related to the reduced risk of death among patients with cancer.