Interleukin-10 as a potential marker of immunosuppression in sepsis-associated myocardial injury: a propensity-matched analysis of 53 pairs

Samenvatting

Background Sepsis-associated myocardial injury (SAMI), defined by elevated high-sensitivity cardiac troponin T (hsTnT), occurs frequently in septic patients. However, its independent association with mortality after adjusting for confounders remains controversial. Furthermore, the role of immunomodulatory cytokines in SAMI-related outcomes is poorly characterized. Methods We performed a retrospective, propensity score-matched (PSM) cohort study in a single intensive care unit (ICU). Consecutive patients with suspected infection and procalcitonin (PCT) >= 0.5 ng/mL within 24 h of ICU admission were included as a pragmatic surrogate for Sepsis-3 criteria. Patients were classified as SAMI+ (hsTnT > 0.014 ng/mL) or SAMI-. The primary outcome was in-hospital mortality. PSM (1:1, caliper = 0.25) was used to control for baseline confounders. Results Among 375 eligible patients, 317 (84.5%) were SAMI+. PSM yielded 53 well-balanced pairs ( n = 106), although a residual imbalance in hypertension prevalence persisted (SMD = 0.442). Post-PSM, in-hospital mortality did not differ significantly between SAMI+ and SAMI− groups (11.3% vs. 18.9%, P = 0.416). Notably, SAMI+ patients exhibited elevated levels of the anti-inflammatory cytokine interleukin-10 (IL-10) compared to SAMI− patients (median: 21.71 vs. 7.06 pg/mL, P = 0.040), although this finding should be considered hypothesis-generating given the small subset analysis. As expected, hsTnT and NT-proBNP remained significantly higher in the SAMI+ group. Conclusions After propensity matching, we were unable to detect a statistically significant association between SAMI and increased in-hospital mortality. The higher IL-10 levels in SAMI+ patients suggest a possible association with a compensatory anti-inflammatory response, though this finding requires validation. These results highlight the need to explore immune-phenotypes rather than cardiac injury alone in determining the prognosis of septic patients. The interpretation of mortality outcomes is limited by residual confounding, possible survivor bias, and the restricted generalizability of the matched cohort.