The value of admission Troponin I to predict outcomes in suspected infections in elderly patients admitted in Internal Medicine: results from the SOFA-T collaboration, a multi-center study

Elderly patients affected by suspected infection and declining clinical conditions can be admitted to stepdown units (SDU), but a risk stratification is necessary to optimize their management. Admission troponin I (aTnI) has a prognostic role, however, one of the most commonly used stratification tools, the Sequential Organ Failure Assessment score (SOFA), does not consider myocardial injury. With this paper, we aimed to evaluate the prognostic accuracy of a new score, named SOFA-T, considering both SOFA score and aTnI in a cohort of elderly patients admitted to the stepdown beds of two Internal Medicine departments. Patients aged > 65 years admitted in SDU of two different hospitals of the same region in a 12-months timeframe were retrospectively assessed obtaining age, sex, days of admission, in-hospital death, SOFA, aTnI and comorbidities. The best aTnI cutoff for in-hospital death was calculated with ROC curve analysis; dichotomous variables were compared with chi-squared test; continuous variables were compared with t test or Mann–Whitney test. We obtained a cohort of 390 patients. The best aTnI cutoff was 0.31 ng/ml: patients with increased aTnI had higher risk of in-hospital death (OR: 1.834; 95% CI 1.160–2.900; p = 0.009), and higher SOFA (6.81 ± 2.71 versus 5.97 ± 3.10; p = 0.010). Adding aTnI to SOFA increased significantly the area under the curve (AUCSOFA = 0.68; 95% CI 0.64–0.73; AUCSOFA-T = 0.71; 95% CI 0.65–0.76; p = 0.0001), with a slight improvement of the prognostic performance. In elderly patients admitted to SDU for suspected infection, sepsis or septic shock, aTnI slightly improves the accuracy of SOFA score of the in-hospital death prediction.