Evaluation of Three Trauma Mortality Scores in Adults Admitted to Intensive Care in Tunisia (TICU-SCORES)

External Validation and Local Recalibration of GAP, MGAP, and the New Trauma Score for Triage (T-NTS) for All-Cause In-Hospital Mortality in Adult Trauma Patients Admitted to Intensive Care in Tunisia: A Retrospective Cohort Study

The goal of this retrospective observational study is to learn how well three simple trauma scores predict death during the hospital stay in adults admitted to intensive care after an injury in Tunisia. The scores are the Glasgow Coma Scale, Age, and Systolic Blood Pressure score (GAP); the Mechanism, Glasgow Coma Scale, Age, and Systolic Blood Pressure score (MGAP); and the New Trauma Score for Triage (T-NTS). Researchers will review existing medical records for admissions from 1 January 2018 through 31 December 2025. They will not assign treatment or contact participants. The main question is how accurately the three scores distinguish participants who died in the hospital from those discharged alive. Secondary analyses will examine agreement between score-based risk and observed mortality and performance for mortality by day 7 after ICU admission.

Study Overview

• Status: Not yet recruiting
• Conditions: Wounds and Injuries; Critical Illness; Mortality
• Intervention / Treatment: Other: Trauma mortality prediction scores: GAP, MGAP, and T-NTS

Detailed Description

This single-centre retrospective cohort study will externally validate GAP, MGAP, and T-NTS in an ICU-selected adult trauma population at a tertiary hospital in Tunisia. All consecutive eligible adults physically admitted to the participating ICU during the fixed eight-year clinical window will be screened. Predictor values will be taken from the earliest qualified assessment after injury and before sedation, neuromuscular blockade, and tracheal intubation. The study therefore evaluates prognosis among trauma patients who reached intensive care and does not claim validation for triage of all emergency-department trauma patients.

The primary endpoint is death from any cause before live discharge from the continuous index admission to the study hospital. The primary comparison will use one common complete-case cohort with known outcome, age, mechanism, qualified Glasgow Coma Scale, systolic blood pressure, and peripheral oxygen saturation. Missing predictor values will not be imputed and will not change cohort eligibility; missingness and the flow from screened to eligible and analyzable participants will be reported.

Score discrimination will be summarized by area under the receiver operating characteristic curve with 95% confidence intervals and paired comparisons using DeLong's method. Observed mortality will be described within original score risk groups. Separate one-predictor logistic models will estimate local score-to-risk relationships, with 1,000 bootstrap resamples for internal validation. Mortality by calendar day 7 after index ICU admission will be a prespecified secondary endpoint if retained in the final signed protocol. No treatment, diagnostic procedure, or exposure is assigned by the investigators.

• Study Type: Observational
• Enrollment (Estimated): 1000

Contacts and Locations

• Study Contact: Name: sameh ben ahmed, MD; Phone Number: +216 72 100 532; Email: sameh.benahmed@fmt.utm.tn

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 18 years and older (Adult, Older Adult)
• Accepts Healthy Volunteers: No
• Genders Eligible for Study: All

• Sampling Method: Probability Sample

Study Population

Adults aged 18 years or older who were physically admitted to the intensive care unit of Mohamed Tahar Maâmouri Hospital, Nabeul, Tunisia, because acute blunt or penetrating trauma, or an immediate consequence of that trauma, was the principal indication for critical care between 1 January 2018 and 31 December 2025. All consecutive potentially eligible admissions will be identified from the ICU registers and medical records. Only the first qualifying trauma episode and ICU admission per patient will be included. Eligible patients with missing score variables will remain in the eligible cohort but may not enter the common complete-case analysis

Description

Inclusion Criteria:

1. Age 18 years or older at the earliest qualified assessment.
2. Acute physical injury caused by a blunt or penetrating mechanism.
3. Trauma or an immediate consequence of trauma was the principal reason for the first qualifying intensive care unit admission.
4. The first qualifying intensive care unit admission occurred from 1 January 2018 through 31 December 2025, inclusive.
5. The intensive care unit admission belonged to the continuous acute hospitalization following the index injury.

Exclusion Criteria:

1. Age younger than 18 years.
2. No physical admission to the participating intensive care unit, including death before ICU admission.
3. Isolated burn, isolated inhalation injury, drowning, hanging, asphyxia, poisoning, or isolated thermal, electrical, or chemical injury without qualifying blunt or penetrating trauma.
4. ICU admission for a medical condition when trauma was incidental and not the principal indication for critical care.
5. Elective, reconstructive, rehabilitation, or late readmission after discharge from the acute trauma hospitalization.
6. Iatrogenic injury or procedural complication without qualifying external blunt or penetrating trauma.
7. Duplicate record or ICU readmission within the same trauma hospitalization.
8. A later eligible trauma episode in a patient already included during the study period.

Study Plan

How is the study designed?

Design Details

• Observational Models: COHORT
• Time Perspectives: RETROSPECTIVE

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort

Intervention / Treatment

Adult Trauma ICU Cohort; All consecutive eligible adults admitted to the participating intensive care unit because of acute blunt or penetrating trauma between 1 January 2018 and 31 December 2025. GAP, MGAP, and the New Trauma Score for Triage (T-NTS) will be calculated retrospectively from routinely recorded clinical data. No treatment or diagnostic procedure is assigned by the investigators.

Other: Trauma mortality prediction scores: GAP, MGAP, and T-NTS; The Glasgow Coma Scale, Age, and Systolic Blood Pressure score (GAP); the Mechanism, Glasgow Coma Scale, Age, and Systolic Blood Pressure score (MGAP); and the New Trauma Score for Triage (T-NTS) will be calculated retrospectively using existing clinical information recorded before sedation, neuromuscular blockade, or tracheal intubation whenever available. The scores are evaluated as prognostic tools; they are not used to assign treatment or alter patient care.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Discrimination of GAP, MGAP, and T-NTS for All-Cause In-Hospital Mortality	The clinical endpoint is death from any cause before live discharge from the continuous index admission to the study hospital. For each score, discrimination will be quantified by the area under the receiver operating characteristic curve (AUROC) with a 95% confidence interval in the common complete-case cohort. AUROC ranges from 0.5 (no discrimination) to 1.0 (perfect discrimination). Correlated AUROCs will be compared using DeLong's method.	From index hospital presentation to in-hospital death or live discharge during the continuous index hospitalization (expected up to 12 months)

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Calibration and Local Recalibration of GAP, MGAP, and T-NTS for All-Cause In-Hospital Mortality	Observed all-cause in-hospital mortality with 95% confidence intervals will be reported within the original risk groups of each score. For each score, a one-predictor logistic model relating the original total score to mortality will estimate a local score-to-risk relationship. Internal validation of the recalibration procedure will use 1,000 bootstrap resamples.	From index hospital presentation to in-hospital death or live discharge during the continuous index hospitalization (expected up to 12 months)
Discrimination of GAP, MGAP, and T-NTS for All-Cause In-Hospital Mortality by Day 7	The clinical endpoint is death from any cause in the study hospital on or before calendar day 7 after the first qualifying ICU admission; the ICU admission date is day 0. The endpoint is derived from dates and is not an exact 168-hour measure. Discrimination of GAP, MGAP, and T-NTS will be summarized by AUROC with 95% confidence intervals; paired DeLong comparisons will be reported if the number of day-7 deaths is adequate.	From index ICU admission through calendar day 7 (ICU admission date = day 0)

Collaborators and Investigators

• Sponsor: University Tunis El Manar
• Investigators: Principal Investigator: sameh ben ahmed, MD, Anesthesiology and Intensive Care Unit, Mohamed Taher Maamouri Teaching Hospital Faculty of Medicine of Tunis, University Tunis El Manar Mrezga, 8000 Nabeul, Tunisia

Study record dates

Study Major Dates

• Study Start (Estimated): October 1, 2026
• Primary Completion (Estimated): December 15, 2026
• Study Completion (Estimated): December 15, 2026

Study Registration Dates

• First Submitted: September 23, 2026
• First Submitted That Met QC Criteria: September 23, 2026
• First Posted (Actual): September 30, 2026

Study Record Updates

• Last Update Posted (Actual): October 2, 2026
• Last Update Submitted That Met QC Criteria: September 29, 2026
• Last Verified: September 1, 2026

More Information

Terms related to this study

• Keywords: trauma; intensive care; external validation; Tunisia; prognostic model; hospital mortality; MGAP; GAP; New Trauma Score of Triage
• Additional Relevant MeSH Terms: Pathologic Processes; Disease Attributes; Pathological Conditions, Signs and Symptoms; Critical Illness; Wounds and Injuries
• Other Study ID Numbers: HTMM-TICU-SCORES-01; 59-2026 (Other Identifier: Comité d'éthique de l'Hôpital Mohamed Tahar Maâmouri)

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

IPD Plan Description

Individual participant-level data extracted from confidential medical records will not be made publicly available. The data dictionary and statistical code may be shared with publication subject to ethics and institutional approval

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No