Abbreviations

F1000Research

2046-1402

F1000 Research Limited

London, UK

10.12688/f1000research.138364.1

Research Article

Articles

Predictors of mortality for major trauma patients in intensive care: A retrospective cohort study

[version 1; peer review: 2 approved with reservations]

Jennings

Michael

Data Curation Project Administration Writing – Original Draft Preparation Writing – Review & Editing https://orcid.org/0000-0001-7588-2827 a 1 2 Booker

James

Data Curation Formal Analysis Visualization Writing – Original Draft Preparation Writing – Review & Editing 3 Addison

Amy

Conceptualization Data Curation Formal Analysis Methodology Visualization Writing – Original Draft Preparation Writing – Review & Editing 4 Egglestone

Rebecca

Data Curation Writing – Original Draft Preparation Writing – Review & Editing 2 Dushianthan

Ahilanandan

Conceptualization Data Curation Formal Analysis Methodology Project Administration Supervision Visualization Writing – Original Draft Preparation Writing – Review & Editing https://orcid.org/0000-0002-0165-3359 2 4 5 1Department of Anaesthetics and Perioperative Medicine, University College London Hospitals NHS Foundation Trust, 235 Euston Road, London, NW1 2BU, UK 2General Intensive Care Unit, University Hospital Southampton NHS Foundation Trust, Tremona Road, Southampton, SO16 6YD, UK 3Department of Neurosurgery, Wessex Neurological Centre, University Hospital Southampton NHS Foundation Trust, Tremona Road, Southampton, SO16 6YD, UK 4Faculty of Medicine, University of Southampton, Tremona Road, Southampton, SO16 6YD, UK 5NIHR Southampton Clinical Research Facility and NIHR Southampton Biomedical Research Centre, University Hospital Southampton NHS Foundation Trust and University of Southampton, Tremona Road, Southampton, SO16 6YD, UK

a michael.jennings1995@gmail.com

No competing interests were disclosed.

14 8 2023

2023

974

24 7 2023

2023

This is an open access article distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Background: Major trauma places substantial demand on critical care services, is a leading cause of death in under 40-year-olds and causes significant morbidity and mortality across all age groups. Various factors influence patient outcome and predefining these could allow prognostication. The aim of this study was to identify predictors of mortality from major trauma in intensive care.

Methods: This was a retrospective study of adult trauma patients admitted to general intensive care between January 2018 and December 2019. We assessed the impact on mortality of patient demographics, patterns of injury, injury scores (Glasgow Coma Score (GCS), Charlson’s comorbidity index (CCI), Acute Physiology and Health Evaluation II (APACHE II), Injury Severity Score (ISS) and Probability of Survival Score (Ps19)), number of surgeries and mechanism of injury using logistic regression.

Results: A total of 414 patients were included with a median age of 54 years (IQR 34–72). Overall mortality was 18.6%. The most common mechanism of injury was traffic collision (46%). Non-survivors were older, had higher ISS scores with lower GCS on admission and probability of survival scores. Factors independently predictive of mortality were increasing age (OR 1.06, p <0.001) and GCS <15 on admission (OR 7.21, p <0.001). Ps19 was the best predictor of mortality (p <0.001 for each score category), with an AUROC of 0.90.

Conclusions: The significant mortality predictors were age, fall from <2 metres, injury of head or limbs, GCS <15 and Ps19. Contrary to previous studies CCI and APACHE II did not significantly predict mortality. Although Ps19 was found to be the best current prognostic score, trauma prognostication would benefit from a single validated scoring system incorporating both physiological variables and injury patterns.

intensive care critical care trauma mortality scoring systems

The author(s) declared that no grants were involved in supporting this work.

Abbreviations

AIS

Abbreviated Injury Severity

APACHE II

Acute Physiology and Chronic Health Evaluation II

AUROC

area under the receiver operator curve

CCI

Carlson Comorbidity Index

CKD

chronic kidney disease

GCS

Glasgow Coma Score

GICU

general intensive care unit

HLOS

Hospital Length of Stay

HRA

Health Research Authority

ICU

intensive care unit

ISS

injury severity score

IRAS

Integrated Research Application System

LOS

length of stay

MTCs

Multi Trauma Centres

NHS

National Health Service

NICU

Neuro Intensive Care unit

Ps19

Trauma Audit and Research Network Probability of Survival Score

RTCs

road traffic collisions

TARN

Trauma Research Audit Network

VAP

ventilator associated pneumonia

Introduction

Major trauma accounts for almost 10% of all deaths worldwide. ¹ The National Audit Office estimates that there are more than 20,000 major trauma cases each year in England, resulting in 5,400 deaths. ² Furthermore, the demographics and injury patterns of the major trauma population in the UK are changing. Data from the Trauma Audit and Research Network (TARN) show that there has been an increase in the mean age of trauma patients between 1990 and 2013 (36.1 years to 53.8 years), along with a change in the most common mechanism of injury from road traffic collision (RTC) (59.1%) in 1990, to low fall (39.1%) in 2013. Critically unwell trauma patients typically require admission to the Intensive Care Unit (ICU), frequently making up the most resource-intensive critical care patient group (46.9% of ICU patients in a multicentre US study), ³ with significant morbidity and mortality. The burden caused by trauma patients on healthcare systems and the shifts in trauma patient populations has increased the need for evaluating existing scoring systems for their prognostication potential in the ICU trauma population. ⁴

Predicting mortality in the critically unwell trauma patient poses a significant challenge due to the heterogeneity of the patient group and the multitude of patient specific factors that affect ICU outcomes, such as age, comorbidities, and injury patterns. ⁵ ^– ¹⁰ Many of these factors have been examined by several previous studies, but there is no consensus on the most useful prognostication scores. Both physiological and anatomical scoring systems have been purported to correlate best with mortality, ¹¹ ^, ¹² thus there is a current requirement for development of a new scoring tool for early mortality prediction in trauma ICU patients that incorporates a combination of physiological and anatomical scoring components. ¹³ ^, ¹⁴

The aim of this investigation was to determine which patient specific factors (present at the point of admission) and which injury severity scoring systems are the most accurate predictors of poor outcome in trauma patients admitted to ICU.

Methods Study design

This is a retrospective study of all critically ill trauma patients aged ≥18 years admitted to the General Intensive Care Unit (GICU) at Southampton General Hospital, between January 2018 and December 2019. ¹⁵ Only patients admitted to the GICU were included in this study and patients admitted to other areas including high dependency units and neuroscience intensive care unit were excluded. The sample size was determined by the number patients admitted during the defined time period and was comparable to studies of similar design. Authors did not have access to information that could identify individual participants during or after data collection. Ethical approval was obtained through Ethics and Research Governance Online (ERGO) by the Faculty of Medicine at Southampton University on 4 August 2020, Reference 56519. This study was part of the large CRIT-CO study (Outcomes of Patients Admitted with Critical-Illness to the General Intensive Care Unit – a Retrospective Observational Study) IRAS Reference 232922. This study used retrospective analysis of non-identifiable patient data, thus the need for individual informed consent was waived.

Baseline data and outcomes

The following variables were collected from all available Southampton General Hospital databases: age, sex, comorbidities, mechanism of injury, and injury distribution. ¹⁵ Admission scores including Glasgow Coma Score (GCS), Injury Severity Score (ISS), TARN Probability of Survival Score (Ps19), and the Acute Physiology and Health Evaluation II (APACHE II) score were calculated. We also quantified the prehospital comorbidity by Charlson’s comorbidity index (CCI). The outcomes evaluated were duration of mechanical ventilation, ICU, and hospital length of stay and 28-day all cause hospital mortality.

Statistical analysis

Continuous variables are expressed as median and interquartile range (IQR). Mann Whitney U was used as the statistical analysis for continuous variables and chi-square for categorical variables. The distribution of variables was assessed and if they had a non-normal distribution they were dichotomised into categorical variables with equal sized groups. Univariate analysis using logistic regression to investigate if variables that varied significantly between the survival and non-survival group, were also significant predictors of mortality. Prior to a multivariate analysis a correlation matrix was done to assess the collinearity between each of the significant predictors using Spearman’s test. This informed the subsequent multivariate analysis using a logistic regression to identify independent significant predictors of survival. Predictors were deemed significant if p < 0.05. Additionally, receiver operating characteristics (ROC) area under the curve (AUC) graphs were constructed to assess each variables performance in predicting mortality. Data analysis was done in SPSS Version 25 (RRID:SCR_016479) and RStudio Version 1.4.1103 (RRID:SCR_000432) using packages: dplyr, ggplot2, lme4 and pROC. ¹⁵

Results Demographics

A total of 414 critically injured trauma patients were admitted to the Intensive Care Unit between January 2018 and December 2019. Of these, 69.3% (n = 287) were male and 30.7% (n = 127) were female. The median age was 54 years (IQR 34–72). Of those admitted, 66.2% (n = 274) had at least one co-morbidity and the median CCI was 1 (IQR 0–3). The most common mechanism of injury was vehicle incident (46.1%), followed by fall <2 metres (23.9%). The most common body part injured was chest (29.2%), followed by other (20%), multiple injuries (19.1%), head (12.1%), abdominal (8.7%), spinal (6.3%), limbs (3.6%) and facial injuries (1%) ( Figure 1). The median GCS, ISS and APACHE II scores were 15, 22 and 11 respectively.

Figure 1. The mechanism of injury and proportion of injuries in different body regions.

Overall, the 28-day all cause hospital survival was 81.4% (n = 337), with survivors being on average younger than non-survivors (51 (32–68) vs 74-years-old (55–85). There were no survival differences between male and female patients. Survivors had fewer comorbidities than non survivors (CCI medians of 1 (0–4) and 4, ¹ ^– ⁴ respectively). Among non-survivors, fall of <2 metres was the most common mechanism of injury (39%) reflecting the older age of this group, followed by an RTC (32.5%). Among survivors, the most common mechanism of injury was RTC (49.3%), followed by fall from <2 metres (20.5%). Non-survivors had lower GCS at presentation than survivors (9 vs 15). They also had higher ISS (25 vs 20) and higher APACHE-II scores (13 vs 11) than survivors at presentation.

The Ps19 predictive model was significantly lower for non-survivors (59 vs 98). The type of body region injured also varied between survivors and non-survivors. The non-survivors had increased frequency of head injury and the survivors had more chest injuries. Abdominal injuries were more common in survivors than non-survivors (10.1% vs 2.6%), whereas limb injuries were more common in non-survivors than survivors (10.4% and 2.1% respectively) ( Table 1).

Table 1. Patient demographics and injury characteristics.

Variable	All patients (n = 414)	Survivors (n = 337)	Non-survivors (n = 77)
Age, years	54 (34–72)	51 (32–68)	74 (55–85)
Gender
Female	127 (30.7%)	103 (30.6%)	24 (31.2%)
Male	287 (69.3%)	234 (69.4%)	53 (68.8%)
Co-morbidity
Any	274 (66.2%)	218 (64.7%)	56 (72.72)
CCI median	1 (0–3)	1 (0–3)	4 (1–4)
Mechanism of injury
Vehicle incident (RTC)	191 (46.1%)	166 (49.3%)	25 (32.5%)
Fall > 2 m	58 (14.0%)	48 (14.2%)	10 (13.0%)
Fall < 2 m	99 (23.9%)	69 (20.5%)	30 (39.0%)
Other	66 (15.9%)	54 (16.0%)	12 (15.6%)
GCS	15 (14–15)	15 (14–15)	9 (3–15)
ISS score	22 (13–29)	20 (12–29)	25 (16–38)
Ps19	97 (86–99)	98 (92–99)	59 (25–85)
APACHE II	11 (7–14)	11 (7–14)	13 (9–18)
Number of surgeries (n)	1 (0–2)	1 (0–2)	0 (0–1)
Most severely injured body region:
Head	50 (12.1%)	24 (7.1%)	26 (33.8%)
Face	4 (1.0%)	4 (1.2%)	0
Chest	121 (29.2%)	112 (33.3%)	9 (11.7%)
Spine	26 (6.3%)	22 (6.5%)	4 (5.2%)
Abdomen	36 (8.7%)	34 (10.1%)	2 (2.6%)
Multiple	79 (19.1%)	64 (19.0%)	15 (19.5%)
Limbs	15 (3.6%)	7 (2.1%)	8 (10.4%)
Other	83 (20.0%)	70 (20.8%)	13 (16.9%)

Data presented as median and interquartile ranges or number and (%).

APACHE II: Acute physiology and chronic health evaluation; CCI: Charlson’s Comorbidity Index; GCS: Glasgow Coma Scale; ISS: Injury Severity Score; Ps19: Probability of survival; RTC: Road Traffic Collision.

We performed univariate logistic regression analysis to assess the association between common variables that demonstrated significant difference between survivor and non-survivor groups with 28-day hospital survival ( Table 2). In the univariate analysis, increasing age (OR 1.04, CI 1.03–1.06, p < 0.001), fall <2 metres (OR 3,17, CI 1.74–5.84, p < 0.001), GCS <15 (OR 3.79, CI 2.21–6.63, p < 0.001), ISS 41–60 (OR 3.10, CI 1.46–6.46, p = 0.00269), Ps19 < 81 (p = 0.001), number of surgeries (OR 0.627, CI 0.467–0.806, p < 0.001) and the most severely injured body region of the head (OR 11.1, CI 4.87–27.1, p < 0.001), multiple injuries (OR 2.60 CI 1.12–6.31, p = 0.0288) and other injuries (OR 12.7, 95% CI (3.84–44.0, p = 0.001) were all significant predictors of mortality ( Table 2). A multivariate analysis was conducted using these variables which demonstrated that increasing age (OR 1.06, CI 1.04–1.08, p < 0.001) and GCS < 15 (OR 7.21, CI 3.80–14.3), p < 0.001) were independent predictors of mortality ( Table 3).

Table 2. Univariate analysis of variables that significantly differed between survivor and non-survivor groups using logistic regression to predict mortality.

Predictor	Estimate	Standard error	OR (CI)	P value
Age	0.0437	0.00710	1.04 (1.03–1.06)	<0.001 ***
Charlson’s Comorbidity index (CCI)	0.0980	0.0618	1.10 (0.98–1.24)	0.122
Mechanism of injury
• RTC	0 ^a	0 ^a	0 ^a	0 ^a
• Fall > 2 m	0.371	0.411	1.45 (0.624–3.17)	0.366
• Fall < 2 m	1.15	0.308	3.17 (1.74–5.84)	<0.001 ***
• Other	0.436	0.387	1.55 (0.705–3.25)	0.260
Glasgow Coma Score (GCS)
• 15	0 ^a	0 ^a	0 ^a	0 ^a
• <15	1.33	0.279	3.79 (2.21–6.63)	<0.001 ***
Injury Severity Score (ISS)
• 1–20	0 ^a	0 ^a	0 ^a	0 ^a
• 21–40	0.541	0.285	1.72 (0.985–3.02)	0.0580
• 41–60	1.13	0.378	3.10 (1.46–6.46)	0.00269**
• 61–80	1.89	1.02	6.62 (0.767–57.1)	0.0645
Probability of survival (Ps19)
• 81–100	0 ^a	0 ^a	0 ^a	0 ^a
• 61–80	2.18	0.420	8.81 (3.82–20.1)	<0.001 ***
• 41–60	3.51	0.569	33.5 (11.6–112)	<0.001 ***
• 21–40	2.81	0.549	16.5 (5.68–50.3)	<0.001 ***
• 0–20	5.33	1.05	206 (39.4–3800)	<0.001 ***
APACHE II
• 0–10	0 ^a	0 ^a	0 ^a	0 ^a
• 11–20	0.648	0.406	1.91 (0.874–4.36)	0.110
• 21–30	0.638	0.701	1.89 (0.399–6.83)	0.363
Number of surgeries	-0.466	0.139	0.627 (0.467–0.806)	0.001 ***
Most severely injured body region
• Chest	0 ^a	0 ^a	0 ^a	0 ^a
• Head	2.41	0.435	11.1 (4.87–27.1)	<0.001 ***
• Face	-13.2	728	0 (0)	0.986
• Spine	0.702	0.636	2.02 (0.517–6.66)	0.270
• Abdomen	-0.426	0.799	0.653 (0.0973–2.63)	0.594
• Multiple	0.956	0.437	2.60 (1.12–6.31)	0.0288 *
• Limbs	0.723	0.448	2.06 (0.861–5.07)	0.106
• Other	2.54	0.614	12.7 (3.84–44.0)	0.001 ***

Predictor	Estimate	Standard error	OR (CI)	P value
Age	0.0437	0.00710	1.04 (1.03–1.06)	<0.001 ***
Charlson’s Comorbidity index (CCI)	0.0980	0.0618	1.10 (0.98–1.24)	0.122
Mechanism of injury
• RTC	0 ^a	0 ^a	0 ^a	0 ^a
• Fall > 2 m	0.371	0.411	1.45 (0.624–3.17)	0.366
• Fall < 2 m	1.15	0.308	3.17 (1.74–5.84)	<0.001 ***
• Other	0.436	0.387	1.55 (0.705–3.25)	0.260
Glasgow Coma Score (GCS)
• 15	0 ^a	0 ^a	0 ^a	0 ^a
• <15	1.33	0.279	3.79 (2.21–6.63)	<0.001 ***
Injury Severity Score (ISS)
• 1–20	0 ^a	0 ^a	0 ^a	0 ^a
• 21–40	0.541	0.285	1.72 (0.985–3.02)	0.0580
• 41–60	1.13	0.378	3.10 (1.46–6.46)	0.00269**
• 61–80	1.89	1.02	6.62 (0.767–57.1)	0.0645
Probability of survival (Ps19)
• 81–100	0 ^a	0 ^a	0 ^a	0 ^a
• 61–80	2.18	0.420	8.81 (3.82–20.1)	<0.001 ***
• 41–60	3.51	0.569	33.5 (11.6–112)	<0.001 ***
• 21–40	2.81	0.549	16.5 (5.68–50.3)	<0.001 ***
• 0–20	5.33	1.05	206 (39.4–3800)	<0.001 ***
APACHE II
• 0–10	0 ^a	0 ^a	0 ^a	0 ^a
• 11–20	0.648	0.406	1.91 (0.874–4.36)	0.110
• 21–30	0.638	0.701	1.89 (0.399–6.83)	0.363
Number of surgeries	-0.466	0.139	0.627 (0.467–0.806)	0.001 ***
Most severely injured body region
• Chest	0 ^a	0 ^a	0 ^a	0 ^a
• Head	2.41	0.435	11.1 (4.87–27.1)	<0.001 ***
• Face	-13.2	728	0 (0)	0.986
• Spine	0.702	0.636	2.02 (0.517–6.66)	0.270
• Abdomen	-0.426	0.799	0.653 (0.0973–2.63)	0.594
• Multiple	0.956	0.437	2.60 (1.12–6.31)	0.0288 *
• Limbs	0.723	0.448	2.06 (0.861–5.07)	0.106
• Other	2.54	0.614	12.7 (3.84–44.0)	0.001 ***

Predictor	Estimate	Standard error	OR (CI)	P value
Age	0.0437	0.00710	1.04 (1.03–1.06)	<0.001 ***
Charlson’s Comorbidity index (CCI)	0.0980	0.0618	1.10 (0.98–1.24)	0.122
Mechanism of injury
• RTC	0 ^a	0 ^a	0 ^a	0 ^a
• Fall > 2 m	0.371	0.411	1.45 (0.624–3.17)	0.366
• Fall < 2 m	1.15	0.308	3.17 (1.74–5.84)	<0.001 ***
• Other	0.436	0.387	1.55 (0.705–3.25)	0.260
Glasgow Coma Score (GCS)
• 15	0 ^a	0 ^a	0 ^a	0 ^a
• <15	1.33	0.279	3.79 (2.21–6.63)	<0.001 ***
Injury Severity Score (ISS)
• 1–20	0 ^a	0 ^a	0 ^a	0 ^a
• 21–40	0.541	0.285	1.72 (0.985–3.02)	0.0580
• 41–60	1.13	0.378	3.10 (1.46–6.46)	0.00269**
• 61–80	1.89	1.02	6.62 (0.767–57.1)	0.0645
Probability of survival (Ps19)
• 81–100	0 ^a	0 ^a	0 ^a	0 ^a
• 61–80	2.18	0.420	8.81 (3.82–20.1)	<0.001 ***
• 41–60	3.51	0.569	33.5 (11.6–112)	<0.001 ***
• 21–40	2.81	0.549	16.5 (5.68–50.3)	<0.001 ***
• 0–20	5.33	1.05	206 (39.4–3800)	<0.001 ***
APACHE II
• 0–10	0 ^a	0 ^a	0 ^a	0 ^a
• 11–20	0.648	0.406	1.91 (0.874–4.36)	0.110
• 21–30	0.638	0.701	1.89 (0.399–6.83)	0.363
Number of surgeries	-0.466	0.139	0.627 (0.467–0.806)	0.001 ***
Most severely injured body region
• Chest	0 ^a	0 ^a	0 ^a	0 ^a
• Head	2.41	0.435	11.1 (4.87–27.1)	<0.001 ***
• Face	-13.2	728	0 (0)	0.986
• Spine	0.702	0.636	2.02 (0.517–6.66)	0.270
• Abdomen	-0.426	0.799	0.653 (0.0973–2.63)	0.594
• Multiple	0.956	0.437	2.60 (1.12–6.31)	0.0288 *
• Limbs	0.723	0.448	2.06 (0.861–5.07)	0.106
• Other	2.54	0.614	12.7 (3.84–44.0)	0.001 ***

p < 0.05.

***

p < 0.001.

0 ^a, Reference category. APACHE II: Acute physiology and chronic health evaluation.

Table 3. Multivariate analysis of variables that were significant in predicting 28-day mortality in the univariate analysis.

Predictor	Estimate	Standard error	OR (CI)	P value
Age	0.0569	0.00959	1.06 (1.04–1.08)	<0.001 ***
Glasgow Coma Score (GCS)
• 15	0 ^a	0 ^a	0 ^a	0 ^a
• <15	1.98	0.337	7.21 (3.80–14.3)	<0.001 ***
Mechanism of injury
• RTC	0 ^a	0 ^a	0 ^a	0 ^a
• Fall > 2 m	-0.111	0.488	0.895 (0.330–2.27)	0.819
• Fall < 2 m	0.325	0.393	1.38 (0.639–3.01)	0.408
• Other	0.625	0.472	1.87 (0.727–4.69)	0.185
Number of surgeries	-0.0313	0.150	0.969 (0.708–1.28)	0.835

p < 0.05.

***

p < 0.001.

0 ^a, Reference category.

We performed a probability of survival analysis based on variables including age and the scoring systems Ps19, ISS, GCS, and APACHE-II ( Figure 2). For Ps19 ( Figure 2A), patients with a low Ps19 0–20 had an almost linear decrease in survival probability up until 14 days, had a lower survival probability and were more likely to die sooner. Patients with Ps19 scores between 21–60 had similar survival probabilities until day seven, at which point they diverge with the 41–60 group having the lowest survival probability at 28 days (28%), and the 21–40 group having a survival probability of 44%. Patients with Ps19 scores of 61 or higher had significantly higher probability of survival than the other groups, with the 61–80 group demonstrating more than 70% chance of survival at 28 days. The Ps19 score >80 group demonstrated a survival probability of over 90%.

Figure 2. Survival probability and trauma scoring systems.

Probability of survival against total number of days in hospital. Patients are categorised into groups based on their scores in scoring systems. A: Ps19 Survival probability. B: APACHE II Survival probability. APACHE II scores could not be calculated for 139 patients, so this is taken into consideration with a N/A line. One patient was excluded from the APACHE II graph due to being the only one who had a score >30. C: GCS Survival probability. D: ISS Survival probability.

APACHE II: Acute physiology and chronic health evaluation; GCS: Glasgow coma scale; ISS: Injury severity score; Ps19: Probability of survival.

For APACHE II score, likelihood of survival at 28 days decreased with increasing APACHE II scores ( Figure 2B). Until the seventh day there was a similar survival curve for all APACHE II scores groups, after which patients with a score of 0–10 clearly show a higher probability of survival compared to patients with an APACHE II score of 11–20 or 21–30, (92%, 89% and 87% at 28 days, respectively).

Patients with reduced GCS ( Figure 2C), had lower likelihood of survival compared to those admitted GCS 15 (73% and 94% at 28 days respectively).

For ISS scores ( Figure 2D), patients in the highest score range (61–80) had only a 50% chance of survival at 28 days. Patients with a score of 41–60 had a 70% survival probability, those scoring 21–40 had an 84% chance of survival whereas the group scoring 1–20 had a 90% probability of survival. Thus, a higher ISS score was associated with a lower probability of survival.

The difference between the predicted Ps19 and observed mortality for the cohort is shown in Figure 3. The Ps19 predicted score was similar to the expected mortality for most ages, except for the groups >80 years of age ( Figure 3).

Figure 3. ROC curves and table of comparison of AUROC and confidence intervals.

Calculated for APACHE II Score, ISS, GCS, Ps19 and Number of surgeries in mortality correlation prediction.

ROC: receiver operator characteristic, AUROC: area under the receiver operator curve, APACHE II: Acute physiological assessment and chronic health evaluation, ISS: Injury Severity Score, GCS: Glasgow Coma Scale, Ps19: Probability of Survival Score.

The area under the receiver operator curve (AUROC) was statistically significant for all variables ( Figure 4). Ps19 was the best predictor of mortality with an AUROC of 0.90 (95% CI 0.85–0.96) followed by GCS AUROC of 0.75 (95% CI 0.64–0.86) and age 0.73 (95% CI 0.62–0.85). ISS, APACHE II and number of surgeries were less predictive of mortality in comparison to these variables with ISS being the worst predictor with an AUROC of 0.66 (95% CI 0.50–0.76).

Figure 4. Observed vs Ps19 predicted survival in different age groups.

Overall, 190 patients (45.9%) required invasive mechanical ventilation and the proportion was higher in non-survivors, compared with survivors (68.8% vs 40.7%) and for both groups the duration of mechanical ventilation was three days. Median ICU and hospital length stay were 3 (IQR 1, 7) and 13 (IQR 7, 26) respectively. There was no statistically significant difference between the ICU length of stay (LOS) for survivors and non-survivors; however, survivors had a longer hospital LOS (15 vs 7 days p < 0.01) ( Table 4).

Table 4. Duration of mechanical ventilation, ICU and hospital length of stay.

Data presented as median and interquartile ranges.

Outcome	All patients	Survivors	Non-survivors	p-value
Mechanical ventilation days	3 (2, 6)	3 (2, 6)	3 (2, 6)	0.772
ICU length of stay (days)	3 (1, 7)	3 (1, 7)	3 (1, 6)	0.430
Hospital length of stay (days)	13 (7, 26)	15 (9, 27)	7 (2,18)	<0.001 ***

p < 0.05.

***

p < 0.001.

Discussion

This study evaluated different patient specific and injury specific factors that influence hospital mortality in critically ill trauma patients. Patient specific factors we investigated included age, gender, and pre-existing comorbidities. Increased age was unsurprisingly found to be associated with a higher mortality. This association is likely to be multifactorial due to an increased risk of frailty, higher likelihood of under triage, and altered physiological mechanisms in elderly patients. Presence of comorbidities in our cohort as determined by CCI was found to be poorly predictive of mortality, in contrast to previous studies which have found a direct impact on mortality. ¹³ Whilst some studies have given conflicting evidence as to the effect of comorbidities on hospital length of stay, ¹⁶ ^, ¹⁷ their importance is acknowledged by their inclusion in trauma scoring systems such as the Ps19 model. We found age and reduced GCS on admission were associated with increased hospital mortality in critically ill trauma patients.

Injury specific factors such as mechanism of injury, body region injured, and severity of injury were all found to affect mortality. In our cohort, most deaths were due to head injuries or polytrauma. Falls and RTC were the most common mechanism of injury, which is consistent with published data from both the UK and the USA. ¹⁸ ^, ¹⁹ The overall mortality in our cohort was 18.6%, with an increased mortality in patients with a fall from <2 metres. Of those injured in an RTC, 86.9% survived compared to 69.7% of those who fell <2 metres. The likely explanation for this finding is that patients who suffer injuries from falls <2 metres resulting in ICU admission are most likely to be an older, more frail population, which is consistent with national data from the TARN database. ²⁰ Moreover, older patients tend to have increased comorbidities with concurrent risk of polypharmacy including anticoagulant medication, which increases risk of adverse outcome from trauma.

Two recent single-centre retrospective observational studies found the following factors to be associated with increased mortality in ICU from trauma: age >60 years, comorbidities (CCI), severity of trauma (New Injury Severity Score (NISS) and Revised Injury Severity Classification (RISC)), patient severity (APACHE II), traumatic brain injury, the use of mechanical ventilation, renal dysfunction in the first 24 hours, and the use of vasoactive drugs and circulatory complications. ²¹ ^, ²² In contrast to our findings, the scores most highly predictive for mortality were APACHE II and NISS. An Australian meta-analysis of over 5000 patients across 25 centres demonstrated similar findings but also showed the Australian and New Zealand Risk of Death (ANZROD) mortality prediction model and APACHE III to be superior to the anatomical scoring systems for mortality prediction (e.g., ISS and NISS). ¹³

Although a higher APACHE II score and CCI were more commonly found in the non-survivor group in our study, these associations proved inadequate predictors of mortality in univariate analyses. However, our findings agreed with previous work that an increased ISS and decreased Ps19 both predict increased mortality. ²² ^– ²⁵ Ps19 was our best performing mortality predictor with an AUROC of 0.9 (95% Cl 0.85–0.96), outperforming ISS (the most used scoring tool).

Our study presents significant limitations. Firstly, the dataset was collected retrospectively from a single centre, notably excluding patients with primary head injury. Secondly, our study did not extensively examine indicators or patient morbidity following trauma; for example, renal dysfunction and ICU interventions such as other organ support measures including renal replacement therapy and the use of vasopressors. We also did not assess other important outcomes such as lasting neurological deficits, rehabilitation required following discharge, which may have provided further context for mortality prediction analyses. Finally, the outcome of our study was limited to 28-day mortality and does not report mortality data at longer timepoints. Whilst there are logistical challenges with data collection over extended timeframes in ICU patients following their discharge, the decision to limit the mortality window to 28 days limits the scope of conclusions to prognostication within a short pre-defined window. Nevertheless, our study complements the existing literature with noteworthy analysis of the mortality prediction capability of a range of scoring systems including ICU specific scoring systems and presents comparable sample sizes to similar recent single-centre studies. ²¹ ^, ²² It was also noted that neither of these studies included the Ps19 scoring system, which is currently used by the TARN (UK), which we found to be the most highly predictive of mortality from trauma in ICU.

Conclusions

This study shows that various internal and external factors determine the mortality of ICU patients within a single-centre general ICU. The most significant prognosticators were age, fall from a height of <2 metres, injury of head or limbs, and scoring systems: GCS, Ps19 and ISS. Ps19 was the best performing score for mortality prediction. Contrary to previous studies, we did not demonstrate an association between mortality and the CCI and APACHE II scoring systems. Trauma prognostication would benefit from an all-inclusive single validated scoring system incorporating physiological variables, injury patterns and ICU variables. This could mitigate extended ICU stays by ensuring that interventions patients receive are better tailored to their individual physiological profile and possibly reduce overall mortality in ICU trauma patients. Future studies would benefit from inclusion of morbidity indicators to provide context for the quality of life experienced by survivors of major trauma.

Data availability Underlying data

Zenodo: Underlying data for ‘Predictors of mortality for major trauma patients in intensive care: A retrospective cohort study’, https://doi.org/10.5281/zenodo.8032931. ¹⁵

This project contains the following underlying data: •

Data sheet 2018–2019 29.09.20.xlsx

•

Trauma analysis.pptx

•

Ps19 vs survival.tiff

•

ROCs.tiff

Reporting guidelines

STROBE checklist for ‘Predictors of mortality for major trauma patients in intensive care: A retrospective cohort study’, https://doi.org/10.5281/zenodo.8032931. ¹⁵

Data are available under the terms of the Creative Commons Zero “No rights reserved” data waiver (CC0 1.0 Public domain dedication).

Software availability

Source code available from: https://github.com/m-a-jennings/Predictors-of-Mortality-for-Major-Trauma-Patients-in-Intensive-Care-A-Retrospective-Cohort-Study/

Archived source code at time of publication: https://doi.org/10.5281/zenodo.8032931. ¹⁵

License: CC0

References 1

Krug

Sharma

Lozano

: The global burden of injuries. Am. J. Public Health. 2000 Apr;90(4):523–526. 10754963

10.2105/AJPH.90.4.523

Morse

: Major trauma care in England. National Audit Office. National Audit Office London;2010 [cited 2022 Aug 20]. Reference Source

Michetti

Fakhry

Brasel

: Trauma ICU Prevalence Project: The diversity of surgical critical care. Trauma Surg. Acute Care Open. 2019 Feb 1;4(1). 10.1136/tsaco-2018-000288

Kehoe

Smith

Edwards

: The changing face of major trauma in the UK. Emerg. Med. J. 2015 Dec 1;32(12):911–915. 26598629

10.1136/emermed-2015-205265

PMC4717354

Evans

Petersen

Meier

: Prehospital traumatic cardiac arrest: management and outcomes from the resuscitation outcomes consortium epistry-trauma and PROPHET registries. J. Trauma Acute Care Surg. 2016;81(2):285–293. 27070438

10.1097/TA.0000000000001070

PMC4961534

Eid

Barss

Adam

: Factors affecting anatomical region of injury, severity, and mortality for road trauma in a high-income developing country: lessons for prevention. Injury. 2009;40(7):703–707. 19117565

10.1016/j.injury.2008.07.012

Camilloni

Farchi

Rossi

: Mortality in elderly injured patients: the role of comorbidities. Int. J. Inj. Control Saf. Promot. 2008;15(1):25–31. 10.1080/17457300701800118

Egglestone

Sparkes

Dushianthan

: Prediction of mortality in critically-ill elderly trauma patients: a single centre retrospective observational study and comparison of the performance of trauma scores. Scand. J. Trauma Resusc. Emerg. Med. 2020;28(1):1–7.

Kisat

Latif

Zogg

: Survival outcomes after prolonged intensive care unit length of stay among trauma patients: The evidence for never giving up. Surgery. 2016;160(3):771–780. 27267552

10.1016/j.surg.2016.04.024

Saika

Bansal

Philip

: Prognostic value of FOUR and GCS scores in determining mortality in patients with traumatic brain injury. Acta Neurochir. 2015;157(8):1323–1328. 10.1007/s00701-015-2469-6

Hwang

Lee

: Comparison of the Sequential Organ Failure Assessment, Acute Physiology and Chronic Health Evaluation II scoring system, and Trauma and Injury Severity Score method for predicting the outcomes of intensive care unit trauma patients. Am. J. Emerg. Med. 2012;30(5):749–753. 21802884

10.1016/j.ajem.2011.05.022

Chou

Liu

: Performance of prognostic scoring systems in trauma patients in the intensive care unit of a trauma center. Int. J. Environ. Res. Public Health. 2020 Oct 1;17(19):1–12. 10.3390/ijerph17197226

Magee

Wilson

Bailey

: Comparison of Intensive Care and Trauma-specific Scoring Systems in Critically Ill Patients. Injury. 2021;52(9):2543–2550. 33827776

10.1016/j.injury.2021.03.049

Polita

Gomez

Friedman

: Comparison of APACHE II and three abbreviated APACHE II scores for predicting outcome among emergency trauma patients. Rev. Assoc. Med. Bras. 2014;60:381–386. 25211423

10.1590/1806-9282.60.04.018

Jennings

: m-a-jennings/ICUtraumapredictors: v1.0.1. 2023 Jun 13 [cited 2023 Jul 10]. 10.5281/zenodo.8032931

Gruskay

Bohl

: Factors affecting length of stay after elective posterior lumbar spine surgery: a multivariate analysis. Spine J. 2015;15(6):1188–1195. 24184639

10.1016/j.spinee.2013.10.022

Pugely

Martin

Gao

: Comorbidities in patients undergoing total knee arthroplasty: do they influence hospital costs and length of stay? Clin. Orthop. Relat. Res. 2014;472(12):3943–3950. 25190193

10.1007/s11999-014-3918-x

PMC4397785

DiMaggio

Ayoung-Chee

Shinseki

: Traumatic injury in the United States: in-patient epidemiology 2000–2011. Injury. 2016;47(7):1393–1403. 27157986

10.1016/j.injury.2016.04.002

PMC5269564

Christensen

Ridley

Lecky

: Outcomes and costs of blunt trauma in England and Wales. Crit. Care. 2008;12(1):1–13.

Banerjee

Baxter

Coats

: Major Trauma in Older People. 2017 [cited 2022 Mar 23]. Reference Source

Lentsck

Oliveira

de Corona

: Risk factors for death of trauma patients admitted to an Intensive Care Unit. Rev. Lat. Am. Enfermagem. 2020 Feb 28;28:e3236–e3236. 10.1590/1518-8345.3482.3236

Papadimitriou-Olivgeris

Panteli

Koutsileou

: Predictors of mortality of trauma patients admitted to the ICU: a retrospective observational study☆. Braz. J. Anesthesiol. 2021;71(1):23–30. 33712248

10.1016/j.bjane.2020.12.006

PMC9523669

Sainz Cabrejas

García Fuentes

García Juarranz

: Evaluation of quality of care in trauma patients using international scoring systems. Medicina Intensiva (English Edition). 2020;44(6):325–332. 10.1016/j.medine.2020.05.002

Xiao

Jin

Mei

: The predictive value of combination of anatomic scoring system and physiological scoring system in prediction of death in patients with severe trauma: a multicenter analysis of 614 cases. Zhonghua Wei Zhong Bing Ji Jiu Yi Xue. 2015;27(4):291–294. 25891460

10.3760/cma.j.issn.2095-4352.2015.04.013

Karataş

AÖ

Çam

: The effect of the use of trauma scoring systems on prognosis of patients with multiple traumas: A cross-sectional study. 2018.

10.5256/f1000research.151554.r266548

Reviewer response for version 1

Cohen

Neta

1 Referee https://orcid.org/0000-0001-7694-6407 1Tel Aviv Sourasky Medical Center, Tel Aviv University, Tel Aviv, Israel

Competing interests: No competing interests were disclosed.

21 5 2024

2024

This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

recommendation

approve-with-reservations

This is a retrospective study of all critically ill trauma patients aged ≥18 years admitted to the General Intensive Care Unit (GICU), aimed to determine which patient specific factors (present at the point of admission) and which injury severity scoring systems are the most accurate predictors of poor outcome in trauma patients admitted to ICU.

Such similar studies were performed earlier (Cohen, N., et.al., 2023 (Ref 1) However, considering the importance of the topic, and the appropriate design, I believe it merit publication.

Comments:

1. Why did you exclude patients from high dependency units and neuroscience intensive care unit?

2. CCI and AGE are co-factors, as age >60 is one of the factors including in the CCI. Try to examine each one separately in the logistic regression model. Please add it in addition to your limitation.

3, How do you explain falling from height< 2 meters to be more severe than fall > 2 meters? how did you decide about cut-off of 2 meters ( the usual cutoff is 3 meters).?

4. Please explain better what are the implication of your findings?

Is the work clearly and accurately presented and does it cite the current literature?

Yes

If applicable, is the statistical analysis and its interpretation appropriate?

Yes

Are all the source data underlying the results available to ensure full reproducibility?

Yes

Is the study design appropriate and is the work technically sound?

Yes

Are the conclusions drawn adequately supported by the results?

Partly

Are sufficient details of methods and analysis provided to allow replication by others?

Partly

Reviewer Expertise:

pediatric emergency medicine, emergency medicine, trauma

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

References 1

: Refining triage practices by predicting the need for emergent care following major trauma: the experience of a level 1 adult trauma center. Eur J Trauma Emerg Surg .2023;49(4) : 10.1007/s00068-022-02195-4 1717-1725

36522466

10.1007/s00068-022-02195-4

Jennings

Michael

Anaesthetics, UCLH, University College London, London, England, UK

Competing interests: None

13 6 2024

Thank you very much for your comments. The manuscript has been amended to address them and we have now submitted an updated version.

1) Exclusion of patients from the high dependency unit and the neuroscience intensive care unit has been further explained within the methods section.

2) To address your comments that CCI and age are cofactors, the univariate and multivariate analyses in tables 2 and 3 have been reanalysed with age as a categorical variable.

3) Falls of less than 2 meters were found to be a significant predictor of mortality in the univariate analysis, but not the multivariate analysis. The explanation of this finding has been added to the discussion section along with an explanation for why the cut-off of two meters was chosen.

4) This discussion and conclusion sections of the manuscript have been improved to explain the implications of our findings in more detail.

Many thanks

10.5256/f1000research.151554.r246878

Reviewer response for version 1

Özpek

Adnan

1 Referee https://orcid.org/0000-0002-2051-256X 1Umraniye Training and Research Hospital, University of Health Sciences, İstanbul, Turkey

Competing interests: No competing interests were disclosed.

1 3 2024

2024

recommendation

approve-with-reservations

I congratulate the authors for their study. My suggestions for the study are as follows:

1- It appears that all the patients in the study were blunt trauma patients, which should be noted in the title of this article. If patients with penetrating trauma are included in the study, it should be stated.

2- Calculated Revised Trauma Score (cRTS) also can be used in addition to the trauma scoring systems used in the study.

3- In the study, mortality rates for falls below 2 meters were significantly higher than for falls above 2 meters. This is not an expected result. It should be questioned whether the patients who fell below 2 meters fell due to an intracranial neurological reason.

Is the work clearly and accurately presented and does it cite the current literature?

Partly

If applicable, is the statistical analysis and its interpretation appropriate?

I cannot comment. A qualified statistician is required.

Are all the source data underlying the results available to ensure full reproducibility?

Partly

Is the study design appropriate and is the work technically sound?

Partly

Are the conclusions drawn adequately supported by the results?

Partly

Are sufficient details of methods and analysis provided to allow replication by others?

Partly

Reviewer Expertise:

General Surgery, Trauma and Emergency Surgery

Jennings

Michael

Anaesthetics, UCLH, University College London, London, England, UK

Competing interests: None

13 6 2024

Thank you very much for your comments. They have been addressed in the new updated manuscript.

1) The article has been amended to include stipulation that only blunt trauma patients were included. See changes to title, abstract and methods section.

2) Mention of cRTS has been added to the limitations section of the discussion as an example of useful scoring systems that were omitted.

3) Factors contributing to our finding that falls < 2 metres being associated with higher mortality have been discussed in more detail in the discussion section.