4. Discussion
The study provides evidence that glucocorticoids are associated with a reduction of myocardial injury in patients admitted with CAP. Furthermore, corticosteroids were associated with a lower incidence of MACE in a long-term follow-up. The effect of corticosteroids seems to be mediated by an antioxidant mechanism related to Nox2 down-regulation.
Previous studies showed that myocardial injury, as assessed by cardiac troponins elevation, is commonly detected in pneumonia. In a cohort of hospitalized CAP, we previously showed that elevated levels of hs-cTnT (> 0.014 µg/L; i.e. over the 99thpercentile) were found in more than 50% of patients and associated to in hospital MI in about 10% of patients.5 The increase of hs-cTnT was mostly detected within 24 -72 h from hospital admission, indicating that myocardial damage, as well as the risk of MI, were maximally evident in the first days of hospitalization.
In a cohort of 295 CAP patients in the Netherlands, Vestjens SMT al confirmed that hs-cTnT elevation is common at hospital admission and showed that hs-cTnT elevation is an independent predictor of short- and long- term mortality.22 Recently, in a cohort of 730 CAP patients followed-up for one year, Menéndez R found that hs-cTnT elevation at admission independently predicted early but not late cardiovascular events.23 In the present study, we extended these previous reports, underlying the importance of measuring hs-cTnT to identify patients at risk for cardiovascular events. In our study, hs-cTnT were evaluated not only at admission, but also every 12 hours up to 72 hours. Thus, we showed that maximum levels of in hospital hs-cTnT (i.e. at 24-72 hours) were a strong predictor of MACE in a dose- dependent fashion, during long-term follow-up.
We recently reported a potential role for systemic corticosteroids in reducing MI risk in patients with CAP, indicating that this drug category could encompass a vascular protection effect in this setting.11 However, no data exist about the potential effect corticosteroid on in hospital myocardial injury or long-term cardiovascular events. In the present study we found that systemic corticosteroids reduced troponin increase in patients with troponin elevation during hospitalization, while no changes were detected in patients with normal troponin. Moreover, in patients disclosing troponin increase, in-hospital corticosteroid use was independently associated with a reduction in MACE incidence at a long-term follow-up. Interestingly, no difference in MACE incidence was found in patients who did not disclose troponin increase during hospitalization.
Troponin elevation has been associated to Nox2 up-regulation in patients with CAP suggesting that Nox2- dependent oxidative stress is implicated in myocardial injury.24 Nox2 is among the most important enzymes generating reactive oxidant species (ROS) in the vasculature. ROS production has been proved to have several detrimental effects on the myocardium, such as apoptotic cell death, hypertrophy and dysfunction.25 Numerous studies indicated that Nox2 localizes in cardiomyocytes where it could play an important role in redox balance,26,27 suggesting that enhanced Nox2-derived oxidative stress may be implicated in myocardial injury. Overactivation of Nox2 in CAP patients may be explained by previous report indicating that single stranded RNA viruses irrespective of their classification including influenza A virus and DNA viruses infect cells via Toll-like 7-mediated Nox2 activation and that the virus pathogenicity is abolished in Nox2 knock-out cells.28
Then, we investigated if corticosteroid use could reduce Nox2 activation in this setting. At admission, patients disclosed elevated concentrations of sNox2-dp, a marker of Nox2 activation, that correlated to hs-cTnT levels and were reduced at hospital dismission. Patients treated with systemic corticosteroid showed a stronger reduction in serum sNox2-dp than not-treated ones; such a phenomenon was particularly evident in patients disclosing troponin elevation. To explore the biological plausibility of this finding, we performed in vitro experiments to assess if corticosteroids modulate Nox2 activation and ROS production. These experiments showed that corticosteroids, at concentrations detected in human blood in after administration, decreased Nox2 activation and H2O2production in LPS stimulated cells, through a reduction in AKT and p47phox phosphorylation.
The present study has limitation and implication. The study supports and extends previous reports indicating that patients with troponin elevation are at higher risk of long-term MACE. Patients with troponin elevation may benefit from corticosteroid treatment to improve vascular outcomes. However, the results of our study could be biased by its retrospective nature and by the fact that patients were not randomized to corticosteroid treatment. Corticosteroids were administered at the discretion of the managing physician for perceived need. However, corticosteroid-treated patients were older and with more commorbidities like COPD; thus this fact should reinforce the results of the present report. Even if the herewith reported antioxidant effect and previous reports showing glucocorticoids’ antiplatelet effects18 could explain the cardiovascular protection, the reasons for MACE reduction during the long-term follow-up need to be further investigated.
The beneficial effect of corticosteroid treatment in term of myocardial injury reduction and MACE risk were present only in patients disclosing hs-cTnT elevation. This finding could have important implications as it suggests the usefulness of glucocorticoids only in a subset of CAP patients. This may be in accordance with previous reports showing that corticosteroids reduce morbidity and mortality only in case of severe CAP.29 Thus, we suggest the need of planning RCT to assess if CAP patients with elevated troponin could have beneficial effect by corticosteroid treatment in terms of myocardial injury reduction and ultimately MACE risk reduction.
Nox2 overactivation in CAP patients might have important implications not only in CAP patients but also in other infections mediated by RNA viruses such as COVID-19. Thus, COVID-19 patients often present elevated troponin levels, which are associated with poor outcomes.30 Due to the role of Nox2 in the pathogenesis of RNA viruses, our findings suggest to evaluate if Nox2 activation is implicated in the pathogenicity of COVID-19; in fact, Nox2 oxidase activation down-regulates antibody production, that is required for virus clearance.28