# Sepsis in patients with severe TBI: a retrospective CT scoring study

## TL;DR

This study shows that a CT scan score can predict mortality from sepsis in patients with severe traumatic brain injury.

## Contribution

The study introduces a rapid CT scoring method to predict sepsis mortality in severe TBI patients.

## Key findings

- Sepsis occurred in 60.4% of severe TBI patients, with high mortality.
- CT scores were significantly higher in sepsis patients and predicted mortality better than SOFA scores.
- Elevated CT scores showed 97% sensitivity and 100% specificity for predicting mortality.

## Abstract

Multiple organ failure (MOF) is a severe complication associated with high mortality in sepsis after severe TBI (sTBI).

To investigate the usefulness of a rapid computed tomography (CT) screening score for predicting of mortality and outcomes of sepsis after sTBI.

We retrospectively analyzed those data of patients who were admitted to the ICU. All sTBI patients with or without sepsis underwent rapid CT screening before ICU admission and were admitted to the ICU for > 24 h were included in this study. The main outcome was sepsis-related the mortality after sTBI. The secondary outcome was the GOSE score during the first 60 days.

Among a random sample of 412 adult patients with sTBI, we found 249 sepsis after sTBI (60.4%) and 163 (39.6%) non- sepsis after sTBI events. The main organ failure was early brain (94.8%) and lung injury(91.2%) caused by community-acquired pnumonia (CPA). The CT score was higher in the sTBI with sepsis group than in the sTBI without sepsis group(wean 3.5 score vs. 0.9 score, p < 0.001).The SOFA score was also higher in the sTBI with sepsis group than in the sTBI without sepsis group(wean 5.9 score vs. 3.6 score, p < 0.001). The risk of death for sepsis after sTBI was an elevated CT score (hazard ratio[HR], 4.6; 95% confidence interval[CI], 3.373–10.49; p < 0.001) and an elevated SOFA score (HR,3.0; 95% CI, 2.054–4.826; p < 0.001).The area under the ROC curve for mortality was significantly larger for the elevated CT score (0.90, 95%CI 0.86–0.97 ) than for the elevated score (0.85, 95%CI 0.81–0.89 ) (P < 0.001). The elevated CT score in the area under the ROC curve for mortality was with 97.0% of sensitivity and 100.0% of specificity. At 60 days follow-up, the risk of death for sepsis after sTBI was higher than those non- sepsis after sTBI (p < 0.001).

Elevated CT score is a well indicator of high incidence and mortality for sepsis after sTBI in the ICU, which suggests that this very current and practical event is involved to a global health care problem. But it could still need further verification in future investigation.

The online version contains supplementary material available at 10.1186/s12245-025-00911-6.