Complement activation is associated with neutrophil extracellular traps and all-cause mortality in ST-elevation myocardial infarction

Abstract

Abstract Background The complement system and neutrophil extracellular traps (NETs) are both parts of the innate immune system, and have been implicated in the ischemia-reperfusion injury in patients with ST-elevation myocardial infarction (STEMI). There is experimental evidence of reciprocal activation between the complement system and NETs. Any such link in patients with STEMI has not been investigated. Purpose To investigate a potential association between complement activation and clinical outcomes after STEMI, and assess any interplay between complement activation and NETs in this situation. Methods Patients with ST-elevation myocardial infarction were included at a median of 18 hours after percutaneous coronary intervention (n=864). The terminal complement complex (TCC) was measured by ELISA as a marker of complement activation. As markers of NETs were myeloperoxidase-deoxynucleic acid (MPO-DNA) and citrullinated histone 3 (CitH3) measured by ELISAs, while double stranded DNA (dsDNA) was measured by a nucleic acid stain. Patients were followed for a median of 4.6 years. The primary endpoint was a composite of new myocardial infarction, unscheduled revascularization, stroke, hospitalization for heart failure and death, whichever occurred first. All-cause mortality was also recorded. Results The composite endpoint occurred in 184 (21.3%) patients, while 70 (8.1%) died during follow-up. When dichotomizing at median TCC, the group with above-median TCC levels did not have an increased risk of reaching the composite endpoint (hazard ratio (HR): 1.069, 95% CI: [0.801, 1.428], p=0.651). However, this group exhibited an increased risk of all-cause mortality (HR: 1.650, 95% CI: [1.020, 2.671], p=0.041). This risk persisted when adjusting for age, sex, hypertension and LDL-cholesterol (HR: 1.673, 95% CI: [1.014, 2.761], p=0.044), but the significance was lost when adjusting for NT-proBNP (HR: 1.492, 95% CI: [0.885, 2.515], p=0.133). TCC was correlated to dsDNA (r=0.127, p<0.001) and CitH3 (r=0.102, p=0.003), but not MPO-DNA. The group with both TCC and dsDNA in the highest quartile exhibited a significantly higher incidence of all-cause mortality than the remaining population (17.6% vs, 7.2%, p=0.002). When examining the predictive value of TCC and dsDNA on all-cause mortality in ROC curve analysis, the area under the curve (AUC) for TCC was 0.549 (95% CI: [0.472, 0.625]), while the AUC for dsDNA was 0.653 (95% CI: [0.584, 0.722]). When combining TCC and dsDNA the predictive value was marginally higher than for TCC alone (AUC: 0.649, 95% CI: [0.579, 0.720]) Conclusion In this STEMI population, complement activation measured by TCC was not associated with the primary composite endpoint, but was associated with increased risk of death. TCC was weakly correlated with markers of NETs. Despite a high mortality rate in patients with high levels of TCC and dsDNA, combining these variables did not increase the prognostic value compared to TCC alone. Funding Acknowledgement Type of funding sources: Foundation. Main funding source(s): Stein Erik Hagen's Foundation for Clinical Heart Research Survival according to cox regression