The association between serum lipids and risk of premature mortality in Latin America: a systematic review of population-based prospective cohort studies

Objective To synthetize the scientific evidence on the association between serum lipids and premature mortality in Latin America (LA). Methods Five data bases were searched from inception without language restrictions: Embase, Medline, Global Health, Scopus and LILACS. Population-based studies following random sampling methods were identified. The exposure variable was lipid biomarkers (e.g., total, LDL- or HDL- cholesterol). The outcome was all-cause and cause-specific mortality. The risk of bias was assessed following the Newcastle-Ottawa criteria. Results were summarized qualitatively. Results The initial search resulted in 264 abstracts, five ( N  = 27,903) were included for the synthesis. Three papers reported on the same study from Puerto Rico (baseline in 1965), one was from Brazil (1996) and one from Peru (2007). All reports analysed different exposure variables and used different risk estimates (relative risks, hazard ratios or odds ratios). None of the reviewed reports showed strong association between individual lipid biomarkers and all-cause or cardiovascular mortality. Conclusion The available evidence is outdated, inconsistently reported on several lipid biomarker definitions and used different methods to study the long-term mortality risk. These findings strongly support the need to better ascertain the mortality risk associated with lipid biomarkers in LA.