Increased interleukin-6 levels associate with worse 1-year outcome after acute myocardial infarction and link infarct size, impaired iron status and heart failure
M A Matter, Y J Wang, V A Rossi, S Costantino, F Paneni, B E Staehli, L Raeber, B Gencer, N Rodondi, D Nanchen, J Gawinecka, A von Eckardstein, O Demler, F Ruschitzka, C M MatterAbstract
Background
Interleukin 6 (IL-6) has attracted attention as a key contributor to deleterious pro-inflammatory processes in patients after acute myocardial infarction (AMI). IL-6 may also worsen heart failure (HF) by disrupting iron metabolism. However, data about IL-6 in these settings remain incomplete.
Purpose
To assess the association of IL-6 with outcome, early markers of iron metabolism and HF in patients after AMI.
Methods
A case-control design of AMI patients within the prospective, multicenter SPUM-ACS cohort (n=4787) was used: 158 cases (all-cause death or reinfarction within 1 year) were matched 1:1 to controls (without events) by age, sex, AMI subtype, recruitment site and time from symptom onset to blood sampling (≤ 24h). All patients underwent successful primary percutaneous coronary intervention. Cytokines were analyzed by multiplex Mesoscale assays, iron biomarkers were measured as in routine practice. Linear mixed models (LMM) were used for the matched structure of cases and controls. Least absolute shrinkage and selection operator (LASSO) regression was performed, with all parameters entered as difference (∆) between cases and controls to identify the strongest predictors for ∆IL-6 levels. A stability selection approach was used: parameters with non-zero coefficients in ≥4 out of 5 independent LASSO runs were tested by linear regression for association with ∆IL-6 levels. “Iron status” was defined as a combined parameter of transferrin saturation (TSAT) and serum iron by principal component analysis, given their value as early markers of change in iron status in acute inflammation. Associations between infarct size, IL-6, iron status and HF were tested by serial LMM correcting for key confounders. The hypothesized path from high-sensitivity troponin-T (hsTnT) through IL-6 and iron status to N-terminal pro-B-type natriuretic peptide (NT-proBNP) was tested by serial mediation analyses.
Results
IL-6 levels were higher in cases than controls (25.7 ng/l vs. 20.0 ng/l, p = 0.035). Patients presenting with signs of acute HF (Killip Class ≥ 2) showed higher IL-6 levels. Implementing LASSO regression, hsTnT, IL-1b and tumor necrosis factor a were identified as robust determinants of IL-6 levels (Figure 1). Patients with increased IL-6 showed evidence of impaired iron status, which in turn associated with levels of NT-proBNP after adjustment for confounders. Further analyses confirmed significant partial mediation of the interaction between hsTnT and NT-proBNP by IL-6 and iron status (Figure 2).
Conclusions
IL-6 levels were increased in AMI patients who suffered death or re-infarction at one year. Our findings suggest a partially mediated connection between infarct size (hsTnT), inflammation (IL-6), impaired iron status (serum iron & TSAT) and HF (NT-proBNP). Future studies may investigate early IL-6 blockade and/or iron substitution in patients with large AMI and excessive IL-6 levels.