Stronger Association of Albuminuria with the Risk of Vascular Complications than Estimated Glomerular Filtration Rate in Type 2 Diabetes

Abstract

Introduction: Albuminuria is a risk factor for macro- and microvascular complications of type 2 diabetes (T2D).With an increasing trend of normoalbuminuria, however, of the 2 predictors – estimated glomerular filtration rate (eGFR) and urine albumin-to-creatinine ratio (UACR) – which one is a better predictor of vascular complications of T2D is not clear. Objective: This study aimed to compare the impacts of albuminuria and eGFR on patients with T2D associated with micro- and macrovascular complications. Methods: This retrospective study recruited 4,715 patients with T2D and grouped them based on the values of UACR (high UACR: ≥30 mg/g, low UACR: <30 mg/g) and eGFR (mL/[min × 1.73 m²]) (G1: eGFR ≥ 90; G2: eGFR = 60–89; G3–5: eGFR < 60) from April 2008 to November 2018. Logistic regression analysis was carried out for risk factors in patients with diabetic retinopathy (DR), diabetic peripheral neuropathy (DPN), peripheral arterial disease (PAD), left ventricular remodeling, diastolic disorders, and carotid atherosclerotic plaque in 6 different groups: low UACR + G1 (control group), low UACR + G2, low UACR + G3–5, high UACR + G1, high UACR + G2, and high UACR + G3–5. Patients were grouped according to the change in the UACR value (UACR-decreased group: ≤−30%, UACR-stable group: −30 to 30%, and UACR-increased group ≥30%), eGFR value (eGFR-decreased group: >3%, and eGFR-stable group: ≤3%) and followed up. Results: Compared with the control group, patients with higher albuminuria and lower eGFR had higher adjusted odds ratio (OR) trends of complications, especially in the high UACR + G3–5 group. The OR of 2.010, 3.444, 1.633, 2.742, and 3.014 were obtained for DR, DPN, PAD, left ventricular remodeling, and diastolic disorders, respectively. No statistically significant difference was found in the risk of complications within each one of 2 phenotypes, regardless of the change in the eGFR. After grouping by eGFR, the regression analysis of the urinary protein level in each stage revealed that a majority of complications had a statistically significant difference, except for DR and PAD in the high UACR + G3–5 group. DR in the follow-up study had a higher risk in the UACR-stable/increased group than the UACR-decreased group (UACR stable: OR = 2.568; 95% confidence interval (CI): 1.128–5.849; p = 0.025; UACR increased: OR = 2.489; 95% CI: 1.140–5.433; p = 0.022). Conclusion: UACR is a more predictive risk factor for diabetic complications compared with a reduced eGFR.