Diagnostic utility of calprotectin combined with dynamic changes in cardiac troponin t in sepsis-induced myocardial injury

This study aimed to evaluate the diagnostic utility of calprotectin (CP) combined with dynamic changes in troponin T (cTnT) for the early identification of sepsis-induced myocardial injury (SIMI). This retrospective observational study included 207 patients with sepsis who were admitted to the intensive care unit or emergency intensive care unit of Hebei General Hospital between March 1, 2024, and March 31, 2025. Patients were initially classified according to the presence or absence of coronary heart disease and subsequently subgrouped according to a pragmatic functional definition of SIMI, defined as LVEF < 50% and/or diastolic dysfunction documented by formal echocardiography. Clinical characteristics and inflammatory markers were collected within 24 h of admission. Echocardiographic findings and cardiac injury biomarkers were collected during hospitalization, with cTnT measured at 24 and 48 h. Multivariable logistic regression and receiver operating characteristic (ROC) curve analyses were performed to evaluate the diagnostic performance of calprotectin (CP) and the absolute change in cTnT. CP levels were significantly higher in patients with SIMI than in those without SIMI (57.59 ± 15.91 vs. 47.89 ± 12.00 ng/mL; p < 0.001). In multivariable logistic regression analysis, CP (odds ratio [OR] = 1.058, 95% CI: 1.023–1.094; p = 0.001) and the absolute change in cTnT (OR = 1.021, 95% CI: 1.010–1.033; p < 0.001) were independently associated with SIMI. The area under the ROC curve (AUC) was 0.708 for CP alone and 0.829 for the change in cTnT alone. The combination of CP and cTnT change yielded an AUC of 0.832, representing only a modest numerical increase in discrimination compared with cTnT change alone. In the analysis differentiating SIMI from coronary heart disease-related heart failure, CP, cTnT change, age, and diabetes mellitus were independently associated with group classification in the multivariable model. The combined CP and cTnT ROC model yielded an AUC of 0.809. CP and dynamic changes in cTnT may provide complementary information for the identification of SIMI; however, the incremental improvement in discrimination achieved by adding CP to cTnT change alone was modest in this cohort. These findings should be considered exploratory and require validation in larger, independent cohorts before clinical implementation.