Abstract
Circulating triglyceride (TG) and leukocytes, the main components of the vascular system, may impact each other and co-fuel atherosclerosis. While the causal relationship between plasma TG levels and leukocyte counts remains unclear. Bidirectional Mendelian randomization (MR) analysis was conducted to investigate the potential causal relationship between TG levels and the counts of leukocytes and their subtypes. A cross-lagged panel model (CLPM) using longitudinal healthy screening data (13,389 adults with a follow-up of 4 years) was fitted to examine the temporal relationship between them. Genetically predicted plasma TG levels were positively associated with total leukocyte counts (TLC) [β(se) = 0.195(0.01)], lymphocyte counts (LC) [β(se) = 0.196(0.019)], and neutrophil counts (NC) [β(se) = 0.086(0.01)], which remained significant after adjusting for several confounders. Inversely, the genetically predicted TLC [β(se) = 0.033(0.008)], LC [β(se) = 0.053(0.008)], and NC [β(se) = 0.034(0.008)] were positively associated with plasma TG levels. However, when all three of them were put into the MR model adjusted for each other, only LC was significantly associated with TG levels. There was no association between genetically predicted TG levels and monocyte counts (MC), basophil counts, and eosinophil counts. The results of CLPM showed that the temporal effect of elevated TLC, MC, LC, and NC on plasma TG levels was stronger than the inverse effect. Our findings suggest causal associations of plasma TG levels with TLC, LC, and NC. In turn, LC was positively associated with plasma TG levels. Additionally, elevated circulating LC may precede high plasma TG levels.