Abstract
OBJECTIVE: To investigate serum triglycerides in relation to all-cause, cardiovascular, and non-cardiovascular mortality in an elderly Chinese population. METHODS: The study participants (n = 3565) were elderly (≥ 60 years) community dwellers living in a suburban town of Shanghai. Hypertriglyceridemia was defined as a serum triglycerides concentration ≥ 2.30 mmol/L (definite) and ≥ 1.70 mmol/L (borderline), respectively. RESULTS: The prevalence of definite and borderline hypertriglyceridemia at baseline was 7.5% and 29.5%, respectively. It was higher in women (n = 1982, 9.0% and 33.8%, respectively) than men (n = 1583, 6.2% and 27.9%, respectively), in obese and overweight participants (n = 1566, 10.5% and 36.4%, respectively) than normal weight participants (n = 1999, 5.6% and 27.1%, respectively), and in diabetic participants (n = 177, 11.9% and 39.0%, respectively) than non-diabetic participants (n = 3388, 7.5% and 30.8%, respectively). During a median of 7.9 years follow-up, all-cause, cardiovascular and non-cardiovascular deaths occurred in 529, 216 and 313 participants, respectively. In analyses according to the quintile distributions of serum triglycerides concentration, the sex- and age-standardized mortality rate was lowest in the middle quintile for all-cause, cardiovascular and non-cardiovascular mortality (18.6, 7.8 and 11.9 per 1000 person-years, respectively, versus 21.5, 10.5 and 12.7 per 1000 person-years, respectively, in the two lower quintiles and 21.7, 9.5 and 14.0 per 1000 person-years, respectively, in the two higher quintiles). The fully adjusted hazard ratios (95% CI) for the middle quintile versus the combined two lower with two higher quintiles were 0.85 (95% CI: 0.67-1.07, P = 0.17), 0.81 (95% CI: 0.54-1.19, P = 0.28) and 0.87 (95% CI: 0.64-1.17, P = 0.35) for all-cause, cardiovascular and non-cardiovascular mortality, respectively. CONCLUSIONS: Our study showed high prevalence of hypertriglyceridemia, especially when defined as borderline and in obese and overweight participants, and mildly but non-significantly elevated risks of cardiovascular mortality relative to the middle level of serum triglycerides.