Current studies have identified severe lipid metabolism diseases in diabetic microangiopathy patients, especially in diabetic kidney disease (DKD), diabetic retinopathy (DR) and diabetic neuropathy (DN), with unclear causal relationships. We employed a large-scale dataset containing 179 lipid species as the exposure and large-scale public summary-level datasets of DKD, DR and DN as the outcome. We applied Mendelian randomization (MR) approach to explore causal associations between circulating liposomes and diabetic microangiopathy. A sequence of sensitivity tests was conducted to verify the stability of the MR analysis. We manifest that diacylglycerol (18:1_18:3) (OR = 0.716, 95%CI = 0.559-0.917, P = 0.008), triacylglycerol (OR:0.741-0.763, P < 0.05) and phosphatidylcholine (OR:0.620-1.247, P < 0.05) have a potential association with DKD. And there is a nominal causal effect of phosphatidylinositol (16:0_18:2) (OR = 0.617, 95%CI = 0.401-0.948, P = 0.028), phosphatidylcholine (OR:0.499-0.672, P < 0.05) and sphingomyelin (OR:0.652-1.850, P < 0.05) to DR. In addition, phosphatidylethanolamine (18:1_0:0) (OR = 0.616, 95%CI = 0.405-0.935, P = 0.023), diacylglycerol (16:0_18:1) (OR = 0.675, 95%CI = 0.463-0.984, P = 0.041) and phosphatidylcholine (OR = 0.720-1.619, P < 0.05) nominally associate with DN. It is noteworthy that plasma lipidome of different structures show different effects. We establish a possible causal connection between certain plasma lipidome and major diabetic microangiopathies. Implementing intervention strategies targeting different lipid molecules may provide novel approaches for preventing and treating diabetic microangiopathies.
Read full abstract