# Development and internal-external validation of a risk prediction nomogram for secondary myocardial injury in traumatic brain injury

## TL;DR

This study creates a tool to predict heart injury in brain trauma patients using clinical data, improving early risk assessment.

## Contribution

A novel nomogram with clinical parameters for early prediction of secondary myocardial injury in TBI patients is developed and validated.

## Key findings

- The nomogram achieved good discrimination with AUCs of 0.772 in training, 0.785 in internal validation, and 0.848 in external validation.
- Five key predictors were identified: blood urea nitrogen, hemoglobin, SOFA score, serum potassium, and creatinine.
- The model enables early risk stratification before cardiac biomarker elevation in TBI patients.

## Abstract

Traumatic Brain Injury (TBI)-induced secondary myocardial injury (SMI) is a severe complication with poor prognosis, but reliable early predictive tools are lacking. This study aimed to develop and validate a nomogram for predicting this risk in TBI patients admitted to the intensive care unit (ICU).

We retrospectively analyzed 1,042 ICU-admitted TBI patients without pre-existing cardiac disease from the MIMIC-IV database, randomly divided into training (n = 729) and internal validation (n = 313) sets at a 7:3 ratio. An external validation cohort of 200 patients from Zhejiang Provincial People’s Hospital (2020–2025) was also included. Five key predictors were identified via univariate and multivariate logistic regression.

The final model included blood urea nitrogen, hemoglobin, Sequential Organ Failure Assessment (SOFA) score, serum potassium, and creatinine. It showed good discriminative ability: training set AUC = 0.772 (95%CI: 0.737–0.808), internal validation set AUC = 0.785 (95%CI: 0.733–0.837), and external validation set AUC = 0.848 (95%CI: 0.778–0.917).

This nomogram, based on easily accessible clinical parameters, enables early risk stratification of SMI in TBI patients before cardiac biomarker elevation, providing a practical tool for targeted clinical monitoring and intervention.