Introduction: For at least 35 years the associations between insulin resistance, hyperinsulinemia, hypertriglyceridemia and hypertension have been described. Nevertheless, the existence of an independent association between hypertriglyceridemia (HTG) and cardiovascular diseases (CVD) remains controversial. This may be due to the insufficiency of longitudinal study data that include severe elevations of TG. The purpose of this study was to characterize the associations between systolic blood pressure and various cardiovascular features under different degrees of naturally-occurring HTG severity and to determine the independent effect of severe HTG.
Methods: We characterized TG in a large cohort of longitudinally studied nonhuman primates (n=227; 164 males) maintained for their lifetimes under constant environmental and dietary conditions. The TG ranged from normal TG levels <100 mg/dl, to mild HTG 100-200 mg/dl, high HTG 200-500 mg/dl, very high 500-1000 mg/dl and severe >1000 mg/dl. The Kruskal Wallis H test was applied, as the data samples were not normally distributed.
Results: Severe hypertriglyceridemia was statistically significantly related to systolic blood pressure and was significantly higher (p<0.0001) at the severe HTG levels of 500-1000 mg/dl and ≥1000 mg compared to all other TG levels. Diastolic blood pressure and mean arterial blood pressure were not significantly different among the normal to severe HTG groups. HDL cholesterol was significantly lower (p<0.0001) at all levels of HTG (TG 100-200, 200-500, 500-1000 and ≥1000mg/dl) compared to normal TG <100 mg/dl, and was not related to the severity of the HTG, a finding similar to the relationship with body weight. By contrast, LDL cholesterol was significantly higher (p<0.05) in monkeys with severe HTG >500-1000 mg/dl, as was systolic blood pressure.
Conclusions: In conclusion, much of the association between triglyceride levels and cardiovascular features, including blood pressure, may be principally determined by the severely elevated triglyceride levels, possibly highlighting the importance of longitudinal within subject evaluation of such associations for break points in the interactions of these features.