The cardiac natriuretic peptides are involved in blood pressure regulation, and large cross-sectional studies have shown lower plasma levels of N-terminal pro-natriuretic peptide levels [N-terminal atrial natriuretic peptide (N-ANP) and N-terminal brain natriuretic peptide (N-BNP)] in patients with insulin resistance, obesity, and diabetes.

In this study, we prospectively tested whether plasma levels of mid-regional ANP (MR-ANP) and N-BNP predict new-onset diabetes and long-term glucose progression.

MR-ANP and N-BNP were measured in 1828 nondiabetic individuals of the Malmö Diet and Cancer cohort (mean age 60 yr; 61% women) who subsequently underwent a follow-up exam including an oral glucose tolerance test after a median follow-up time of 16 yr. Logistic regression was used to adjust for covariates.

During follow-up, 301 subjects developed new-onset diabetes. After full multivariate adjustment, MR-ANP was significantly inversely associated with incident diabetes (OR = 0.85; 95% CI = 0.73–0.99; P = 0.034) but not N-BNP (OR = 0.92; 95% CI = 0.80–1.06; P = 0.262). In fully adjusted linear regression models, the progression of fasting glucose during follow-up was significantly inversely related to baseline levels of MR-ANP (P = 0.004) but not N-BNP (P = 0.129). Quartile analyses revealed that the overall association was mainly accounted for by excess risk of incident diabetes in subjects belonging to the lowest quartile of MR-ANP. After full adjustment, the odds ratio for incident diabetes in the bottom compared with the top quartile of MR-ANP was 1.65 (OR = 1.08–2.51, P = 0.019) and 1.43 (OR = 1.04–1.96, P = 0.027) compared with all other subjects.

Low plasma levels of MR-ANP predict development of future diabetes and glucose progression over time, suggesting a causal role of ANP deficiency in diabetes development.