Soluble urokinase plasminogen activator receptor and interleukin-6 improves prediction of all-cause mortality and major adverse cardiovascular events in Type 1 diabetes

Abstract

Background: Type 1 diabetes (T1D) increases premature mortality risk, with cardiovascular disease being the leading cause. Chronic inflammation may play a role. Associations between inflammatory biomarkers and mortality are not well-known in T1D.

Methods: We evaluated a prospective clinical cohort with T1D without known cardiovascular disease. The inflammatory biomarkers soluble-urokinase-plasminogen-activator-receptor (suPAR) and interleukin-6 (IL-6) were measured. Patients were stratified by elevated/low suPAR or IL-6, or simultaneously elevated suPAR and IL-6. Primary and secondary endpoints were all-cause mortality and major adverse cardiovascular events (MACE), respectively. Cox models were adjusted for 10 Steno T1 Risk Engine variables and inflammatory biomarkers. Net reclassification improvement (NRI) and C-statistics were calculated.

Results: Among 962 participants (52% male, median age 50, median follow-up 13.1 years), mortality was higher in patients with elevated inflammation: 31% for elevated versus 9% for low suPAR; 30% for elevated versus 11% for low IL-6; and 50% for simultaneously elevated suPAR and IL-6 versus 5% for low suPAR and IL-6. In fully adjusted models, elevated inflammation was associated with mortality (hazard ratios [95% confidence intervals]: suPAR 2.0 [1.4-3.0, p < 0.001], IL-6 1.8 [1.3-2.6; p = 0.001], and combined 4.0 [2.3-7.2, p < 0.001]) and MACE (suPAR 1.9 [1.4-2.6, p < 0.001], IL-6 1.4 [1.0-1.8, p = 0.034], and combined 2.6 [1.7-4.1, p < 0.001]). Adding suPAR, IL-6, and their combination to the Steno T1 Risk Engine improved NRI for mortality by 61%, 53%, and 84%, respectively, whereas C-statistics improved from 0.808 to 0.829, 0.826, and 0.881, respectively.

Conclusions: suPAR, IL-6, and especially their combination independently predicts all-cause mortality and MACE in T1D without known cardiovascular disease.

Keywords: biomarkers; cardiovascular diseases; diabetes; inflammation.

Fig. 1

Kaplan–Meier curves for all‐cause mortality stratified by levels of suPAR and IL‐6. The dashed horizontal lines indicate the estimated event probability at the end of the follow‐up. Dashes in risk table (—) indicate that the number at risk is not displayed at that timepoint due to a low number of events between intervals. IL‐6, interleukin‐6; suPAR, soluble urokinase plasminogen activator receptor.

Fig. 2

suPAR and IL‐6 and all‐cause mortality in Type 1 diabetes. Model 1: crude, Model 2: adjusted for age, sex, systolic blood pressure, duration of diabetes, HbA1c, estimated glomerular filtration rate, use of statins, and albuminuria status. Model 3: additionally adjusted for smoking (current or prior) and physical activity levels. Model 4.1: additionally adjusted for IL‐6 and hsCRP. Model 4.2: additionally adjusted for suPAR and hsCRP. Model 4.3: additionally adjusted for hsCRP. CI, confidence interval; HR, hazard ratio; hsCRP, high‐sensitivity C‐reactive protein; IL‐6, interleukin‐6; suPAR, soluble urokinase‐type plasminogen activator receptor.

Fig. 3

suPAR and IL‐6 and MACE in Type 1 diabetes. Model 1: crude, Model 2: adjusted for age, sex, systolic blood pressure, duration of diabetes, HbA1c, estimated glomerular filtration rate, use of statins, and albuminuria status. Model 3: additionally adjusted for smoking (current or prior) and physical activity levels. Model 4.1: additionally adjusted for IL‐6 and hsCRP. Model 4.2: additionally adjusted for suPAR and hsCRP. Model 4.3: additionally adjusted for hsCRP. CI, confidence interval; HR, hazard ratio; hsCRP, high‐sensitivity C‐reactive protein; IL‐6, interleukin‐6; MACE, major adverse cardiovascular events; suPAR, soluble urokinase‐type plasminogen activator receptor.