In this study, the association of visit-to-visit TG variability with eGFR decline and incidence of albuminuria in patients with type 2 DM were examined, respectively. The visit-to-visit variability of postprandial TG was a significant predictor of eGFR decline in patients with type 2 DM during long-term follow-up, even when adjusting for confounding factors. In addition, the visit-to-visit variability of postprandial TG was also significantly associated with incidence of microalbuminuria. Thus, we found that the visit-to-visit variability of postprandial TG is associated with DKD progression.
When considering “variability”, experimental models in vitro have shown that intermittent hyperglycemia is more detrimental for endothelial cells than continuous hyperglycemia (26). Glycemic variability is associated with the occurrence of various microvascular and macrovascular complications in DM because of excessive protein glycation end products and activation of oxidative stress in the causation of vascular complication (27) (28). In addition, blood pressure variability is also a significant prognostic factor in ESRD (29) (30) (31). Both plasma glucose level and blood pressure originally fluctuate to a certain extent and are components of metabolic syndrome, which are highly interrelated to the development and progression of DKD (5). Given these observation, it is likely that a similar association may be found for variability of TG, which is another factor of metabolic syndrome and fluctuates to a certain extent because it has already reported that both fasting and postprandial hypertriglyceridemia are associated with eGFR decline and incidence of albuminuria (14) (15) (16), variability of fasting TG is predictive of coronary events (20), and is also linked to incident microalbuminuria in patients with type 2 DM (19).
An appropriate device to measure the trend for 24-hour blood pressure and plasma glucose level is available, however, not for serum TG concentration. Therefore, we analyzed postprandial TG variability by SD, Adj-SD and MMD. Many previous papers about “variability” used various indices to evaluate their variability. SD has been used for evaluating glycemic, blood pressure and fasting TG variability (19) (20) (32) (33) (34) (35). Adj-SD has also been used for evaluating glycemic and fasting TG variability to adjust for the number of measurements (19) (21) (36) (22). MMD has been used for evaluating glycemic and blood pressure variability (37) (23) (24) (38). Our study suggested that SD and Adj-SD might be more reliable than MMD. However, MMD is calculated more easily. In this sense, this method could be suitable for clinical situations. In any method, postprandial TG variability is a significant risk factor for eGFR decline and incidence of microalbuminuria in patients with type 2 DM.
It is important to take the best timing for measurement of TG for consideration for the clinical setting. The lipid profile is conventionally measured in plasma or serum obtained after fasting for at least 8 hours, and therefore may not reflect the daily average plasma lipid (39). In patients with DM, remnant lipoprotein cholesterol levels remain high throughout the day except for a few hours before breakfast (40). There is no evidence that fasting is superior to postprandial when evaluating the lipid profile (41). The Danish Society for Clinical Biochemistry recommended that all laboratories in Denmark use random postprandial lipid profile measurements rather than fasting profiles (42). Traditionally, the Friedewald equation has been applied to a fasting lipid profile; however, calculated LDL-C which is determined with this equation at TG concentrations of 400 mg/dL or less, is similar to LDL-C measured directly on both fasting and postprandial lipid profiles (43) (44). In addition, numerous population-based studies and at least three major statin trials used random, postprandial blood sampling, providing a robust evidence base for a change in the conventional practice of using fasting samples (45) (46) (47). Thus, there are both advantages and disadvantages in taking postprandial lipid profile. Especially, in DM patients, both fasting and postprandial plasma glucose level are important information in controlling DM in clinical practice, but fasting test has a hypoglycemic risk, and HbA1c can be accurately evaluated in either fasting or non-fasting.
The association between renal dysfunction and dyslipidemia had been reported as a lipid nephrotoxicity hypothesis (48). There are pieces of evidence that renal lipid accumulation can cause structural and functional changes in mesangial cells, podocytes and proximal tubule cells, which all contribute to the nephron function. Thus, it is widely recognized that ectopic deposition of lipids causes harm to target cells and organs; ectopic lipid accumulation in the kidney promotes maladaptive responses of renal cells to the mechanical forces of hyperfiltration, leading to podocyte depletion, proteinuria, focal segmental glomerulosclerosis and interstitial fibrosis (49). Regarding lipid nephrotoxicity, our study adds a new significance of better lipid management to optimal DKD treatment, namely, only casual plasma TG concentration but also TG variability should be controlled.
In our study, no significant difference in baseline eGFR was found between the higher group and the lower group, divided by the median value of SD, Adj-SD and MMD, respectively. Although more patients with hyperfiltration might be included in the group with higher SD, higher Adj-SD and higher MMD, it is still important to note that the patients with shorter renal survival demonstrated higher variability of postprandial TG. It is also suggested that the patients with a higher peak of TG presented a higher risk of DKD progression. Furthermore, the possibility that pathophysiological mechanisms underlying postprandial TG variability also induce DKD progression, independent of TG levels, should be considered.
Several studies demonstrated that postprandial hypertriglyceridemia is involved in the production of proinflammatory cytokines, recruitment of neutrophils and generation of oxidative stress, resulting in endothelial dysfunction which is an initial process of atherogenesis and it might contribute to develop albuminuria in healthy subjects as well as hypertriglyceridemic patients and type 2 DM patients (50) (51) (52) (53). Our study showed that the patients with higher variability of postprandial TG developed microalbuminuria significantly earlier, although no significant differences were found in HbA1c, baseline UACR and baseline eGFR between the higher group and the lower group in SD, Adj-SD or MMD, respectively. Hence, it is important to consider deeply that microalbuminuria may be developed earlier in the type 2 DM patients with higher variability of postprandial TG.
Our study was not an interventional study, but an observational study. Therefore, further study is needed to clarify that lowering postprandial TG variability might be helpful for treating early stage of DKD to prevent its progression. So far, a clinical study, the Fenofibrate Intervention and Event Lowering in Diabetes study, showed that fenofibrate prevented progression from normo-albuminuria to micro-albuminuria in patients with type 2 DM (54) (55) (56). Fenofibrate in diabetic mice normalizes endothelial function by balancing vascular reactivity via increasing nitric oxide production and suppressing the vasoconstrictor prostaglandin, suggesting a mechanism of action of fenofibrate (57). It has also been reported that ezetimibe, alogliptin, bezafibrate, vildagliptin and omega-3 fatty acids improve postprandial hypertriglyceridemia and endothelial dysfunction, therefore, these medications have potential to be a treatment option for preventing DKD progression (58) (59) (60) (61) (62). Moreover, a higher variability of postprandial TG could be a marker of incomplete or intermittent compliance to lifestyle measures. In patients with type 2 DM, the risk of kidney events tended to decrease by multifactorial intensive treatment, including lipid control in addition to control of glucose level and blood pressure (63). Given these observations, it may be better to include care for visit-to-visit postprandial TG variability in the lipid control. Thus, further follow-up interventional study is required to clarify the efficacy of lowering postprandial TG variability on delaying progression of DKD to ESRD.