Variability of triglyceride levels and incidence of microalbuminuria in type 2 diabetes

BACKGROUND: Experimental and clinical studies have shown a strong association between hypertriglyceridemia and diabetic nephropathy. A variability of triglyceride (TG) levels has been reported in diabetes. OBJECTIVES: To investigate the relationship of TG variability with the incidence of microalbuminuria (albumin excretion rate > 20 μg/min), in patients with type 2 diabetes. METHODS: A longitudinal, retrospective, observational study was performed on a consecutive series of 457 normoalbuminuric outpatients, with measurements of HbA1c, lipids and microalbuminuria thrice per year with 6.8-year follow-up. TG variability, defined as standard deviation of TG (TG-SD) and TG-SD adjusted for the number of visits was calculated. A nested case-control sensitivity analysis was performed to validate the results of the primary cohort study. RESULTS: Incident microalbuminuria (N = 124, 27.1%) was associated with higher median TG-SD (33.6 mg/dL vs 29.0 mg/dL, P < .05) and TG-SD adjusted for the number of visits (31.4 mg/dL vs 26.7 mg/dL, P < .05). At multivariate (Cox) analysis, logTG-SD and adj-logTG-SD were significant predictors of incident microalbuminuria (hazard ratio 2.1 [1.1-4.2], P = .028 and 1.5 [1.1-3.3], P = .042, respectively). In the case-control analysis, time spent with TG > 150 mg/dL during the follow-up was significantly higher in cases vs controls (27.2 ± 19 vs 16.7 ± 12.5 months, P < .05) with hazard ratio 2.0 (1.1-5.1), P < .05, for adj-logTG-SD. CONCLUSIONS: A higher intraindividual TG variability is a predictor of incident microalbuminuria in type 2 diabetes. In addition, time of exposure to elevated TG levels (>150 mg/dL) predicts incident microalbuminuria.