Pena MJ, Lambers Heerspink HJ, Hellemons ME, Friedrich T, Dallmann G, Lajer M, Bakker SJL, Gansevoort RT, Rossing P, de Zeeuw D, Roscioni SS. Urine and plasma metabolites predict the development of diabetic nephropathy in individuals with Type 2 diabetes mellitus.
Diabet Med 2014;
31:1138-47. [PMID:
24661264 DOI:
10.1111/dme.12447]
[Citation(s) in RCA: 108] [Impact Index Per Article: 9.8] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 08/15/2013] [Revised: 01/09/2014] [Accepted: 03/17/2014] [Indexed: 12/26/2022]
Abstract
AIMS
Early detection of individuals with Type 2 diabetes mellitus or hypertension at risk for micro- or macroalbuminuria may facilitate prevention and treatment of renal disease. We aimed to discover plasma and urine metabolites that predict the development of micro- or macroalbuminuria.
METHODS
Patients with Type 2 diabetes (n = 90) and hypertension (n = 150) were selected from the community-cohort 'Prevention of REnal and Vascular End-stage Disease' (PREVEND) and the Steno Diabetes Center for this case-control study. Cases transitioned in albuminuria stage (from normo- to microalbuminuria or micro- to macroalbuminuria). Controls, matched for age, gender, and baseline albuminuria stage, remained in normo- or microalbuminuria stage during follow-up. Median follow-up was 2.9 years. Metabolomics were performed on plasma and urine. The predictive performance of a metabolite for albuminuria transition was assessed by the integrated discrimination index.
RESULTS
In patients with Type 2 diabetes with normoalbuminuria, no metabolites discriminated cases from controls. In patients with Type 2 diabetes with microalbuminuria, plasma histidine was lower (fold change = 0.87, P = 0.02) and butenoylcarnitine was higher (fold change = 1.17, P = 0.007) in cases vs. controls. In urine, hexose, glutamine and tyrosine were lower in cases vs. controls (fold change = 0.20, P < 0.001; 0.32, P < 0.001; 0.51, P = 0.006, respectively). Adding the metabolites to a model of baseline albuminuria and estimated glomerular filtration rate metabolites improved risk prediction for macroalbuminuria transition (plasma integrated discrimination index = 0.28, P < 0.001; urine integrated discrimination index = 0.43, P < 0.001). These metabolites did not differ between hypertensive cases and controls without Type 2 diabetes.
CONCLUSIONS
Type 2 diabetes-specific plasma and urine metabolites were discovered that predict the development of macroalbuminuria beyond established renal risk markers. These results should be confirmed in a large, prospective cohort.
Collapse