Effect of different factors on predicting patient mortality beyond the NACA classification: a multivariate analysis of more than 2,000 polytrauma patients Authors: Isabel Scala, Martin Riegger, Alessandro Bensa, Andrea Stefano Monteleone, Paolo Ivan Fiore, Christian Candrian, Giuseppe Filardo Published: 2025 ABSTRACT AbstractBackgroundThe National Advisory Committee for Aeronautics (NACA) score is widely used to assess polytrauma patients. NACA scores of 4 and 5 indicate severe and potentially life-threatening injuries. However, these two categories are rather broad, with a large variety of conditions within each level. Aim of this study was to investigate a large cohort of NACA 4 and NACA 5 trauma patients and quantify the impact of individual factors on the mortality risk.MethodsThe polytrauma registry of the Lugano Regional Hospital affiliated to the National Swiss Trauma Registry (STR) was retrospectively analysed to investigate all patients admitted between 2015 and 2023 with a NACA score of 4 or 5. Out of 2,152 patients, 1,684 were NACA 4 and 468 were classified as NACA 5. Extracted parameters included age, mechanism of injury, NACA score, prehospital variables and hospital variables, and patient diagnoses. Univariate and multivariate analyses were performed for all patients combined and separately for NACA 4 and NACA 5 subgroups.ResultsA total of 121 patients died during hospitalization. The multivariate analysis of the prehospital variables showed that NACA score (OR 2.96), age (OR 1.11), Glasgow Coma Scale (GCS) (OR 0.78), cardiopulmonary resuscitation (CPR) (OR 7.22), and administration of vasoactive drugs (OR 2.38) were statistically significant independent risk factors for all included patients (p < 0.05). In the NACA 4 subgroup analysis only age (OR 1.13) and GCS (OR 0.69) emerged as statistically significant predictors of mortality (p < 0.05). In the NACA 5 subgroup, age (OR 1.09), GCS (OR 0.28), and administration of vasoactive drugs (OR 2.40) emerged as statistically significant predictors of mortality (p < 0.05).ConclusionThis study demonstrated that different factors can influence the mortality risk of polytrauma patients beyond the mere NACA 4 and 5 classification, including CPR, administration of vasoactive drugs, GCS, and age. These findings highlight the association between specific clinical parameters and the mortality risk, whose early identification is paramount for a more targeted management of polytrauma patients.Level of Evidence (LoE)IIIKeywords: Polytrauma, Registry, NACA score, Emergency medicine, GCS, ISS Background Traumatic accidents are a leading cause of death, disability, and decreased quality of life worldwide, ranking among the top three causes of death in individuals up to 44 years old [1]. In particular, polytrauma continue to pose significant challenges for both treatment and prognosis [2, 3]. The lack of a standardized classification was a major challenge in guiding the therapeutic decisions when managing polytrauma patients, especially during the earliest phases of patient’s admission at the hospital [4]. An early and accurate classification system of polytraumatized patients is crucial for assessing injury severity and estimating mortality risk. Accordingly, over the years the definition of polytrauma evolved, shifting from a purely anatomical-based definition to a more specific framework that additionally integrated the presence of other abnormal clinical parameters [5–8]. In the prehospital setting, the National Advisory Committee for Aeronautics (NACA) score is now widely used by emergency medical services to assess patients. Originally created for 24 h post-admission evaluations, the NACA score was revised in 1980 by Tryba et al. to simplify and expand its use in the emergency setting for trauma and non-trauma patients [9]. Several studies demonstrated the validity of the NACA score in predicting mortality in trauma patients [10–12]. NACA scores of 4 and 5 are among the most frequently assigned classifications, indicating severe and potentially life-threatening injuries. However, these two categories are rather broad and may have overlapping clinical presentations on one side, and on the other hand a large variety of different conditions within each level. The ability to accurately assess the patient condition is paramount as it can directly influence triage decisions, prehospital interventions, and early hospital management. However, the factors having the greatest impact on patients’ overall outcome and on the mortality risk in these patients remain a subject of debate in the current literature, thus impairing optimal trauma care strategies and, in the end, management and survival of these complex patients. This study’s aim was to investigate a large cohort of NACA 4 and NACA 5 trauma patients and to quantify the impact of individual factors on the mortality risk. Materials and methods Study design and population This study was conducted at the Lugano Regional Hospital (Switzerland), a Level I Trauma Center, after the approval of the local Ethical Committee (prot 2024-00236 CE 4536). This research is based on data obtained from the local polytrauma registry affiliated to the National Swiss Trauma Registry (STR) [13], which consists in prospectively collected data from trauma patients including parameters ranging from pre- and hospital setting until patient’s discharge. Patients of all ages admitted to the emergency department (ED) at the Lugano Regional Hospital between January 2015 and December 2023 with a NACA score of 4 or 5 were included in this analysis. Extracted patient parameters included age, mechanism of injury, initial NACA score, prehospital variables (Glasgow Coma Scale (GCS), cardiopulmonary resuscitation (CPR), intubation, administration of vasoactive drugs and fluids), hospital variables (GCS ED, laboratory values ED, intensive care unit (ICU) stay, mechanical ventilation), calculated Injury Severity Score (ISS), and patient diagnoses. The values for prehospital fluids were categorized into four groups (0−500 ml, 501−1000 ml, 1001−1500 ml and > 1500 ml) for clearer interpretation. The survival at discharge was defined as outcome measurement. Patient assessment In the prehospital setting the NACA score and the GCS were the first tools used to describe and assess the status of polytraumatized patients by emergency medical staff. The NACA score classifies patients on a scale from 0−7 (Table 1) based on the severity of their injuries and their needs of medical support [10–12]. The GCS evaluates patients’ neurological responses according to the sum of 3 qualities (eye-opening, motor and verbal responses) to quantify their responsiveness on a scale of 3−15 and to determine a possible impaired neurological status [14]. Table 1National Advisory Committee on Aeronautics (NACA) score [9] and Abbreviated Injury Scale (AIS) [15]NACA scoreDescriptionAISDescriptionNACA 0No injury or diseaseAIS 0No injuryNACA 1Injuries/diseases without need for acute physician careAIS 1MinorNACA 2Injuries/diseases requiring examination and therapy by physician, but admission not indicatedAIS 2ModerateNACA 3Injuries/diseases without acute threat to life but hospital admission requiredAIS 3Serious, not life-threateningNACA 4Injuries/diseases that can lead to deterioration of vital signsAIS 4Serious, life- threateningNACA 5Injuries/diseases with acute threat to lifeAIS 5Critical, survival unknownNACA 6Injuries/diseases transported after successful resuscitationAIS 6Maximum, lethalNACA 7Lethal injuries or diseases (with or without resuscitation attempt)--The Abbreviated Injury Scale (AIS) is assigned to each patient with multiple injuries. For each of the six defined body regions (head/neck, face, chest, abdomen, extremities/pelvis, and external), the highest AIS score is recorded. The three highest AIS scores are then squared and summed to calculate the Injury Severity Score (ISS), which ranges from 0 to 75. If at least one injury is classified as AIS 6, the ISS is automatically set to 75. [15] National Advisory Committee on Aeronautics (NACA) score [9] and Abbreviated Injury Scale (AIS) [15] The Abbreviated Injury Scale (AIS) is assigned to each patient with multiple injuries. For each of the six defined body regions (head/neck, face, chest, abdomen, extremities/pelvis, and external), the highest AIS score is recorded. The three highest AIS scores are then squared and summed to calculate the Injury Severity Score (ISS), which ranges from 0 to 75. If at least one injury is classified as AIS 6, the ISS is automatically set to 75. [15] The severity of the summary of all sustained injuries in a trauma patient is estimated calculating the ISS, based on the Abbreviated Injury Scale (AIS) ranging from 0—6 (Table 1). A cut-off of ISS ≥ 16 is considered a polytrauma [6]. As an additional parameter for the description of trauma severity, the number of injured body regions was assessed based on the list of patients’ diagnoses. The following eleven different regions were defined: head, face, neck, cervical spine, thoracic spine, lumbar spine, thorax, abdomen, upper extremity, lower extremity and pelvis, external (skin) and thermal injuries. A cut-off of ≥ 2 involved regions regardless of the number of injuries within one region was set to detect more severely injured patients. Statistics Statistical analyses were performed using the statistical software STATA (version 17, StatCorp. LP, College Station, TX, USA). Descriptive statistics were used to describe differences in patient characteristics depending on if they were survivors or non-survivors. Continuous variables were reported as mean and standard deviation (SD). Qualitative variables were expressed as absolute and relative frequencies. Qualitative variables were analysed with the Pearson chi squared test or two-sided Fisher exact test. The parametric two-sample t-test (Student’s t-test) and the nonparametric two sample Wilcoxon rank-sum test (Mann Whitney test) were used to analyse continuous variables. Subsequently, univariate logistic regression was performed to determine which variables individually represented a risk factor for mortality. For the identification of which previously established risk factors for mortality acted as independent predictors, a multivariate logistic regression analysis was performed. To account for potential collinearity among independent variables, dedicated statistical checks were performed. Odds ratios (OR) > 1 indicate a higher risk of mortality with the presence or increase of a parameter, while an odds ratio (OR) < 1 expresses an increased risk of mortality for the absence or reduction of a parameter. Every step of the statistical analysis was conducted for all included patients together and for the two subgroups of patients with NACA 4 and NACA 5 separately. Missing values were left out of the analysis. A significance level of 5% (p < 0.05) was considered statistically significant. Results A total of 2,152 trauma patients were enrolled in the study, with 78.3% classified as NACA 4 and 21.7% as NACA 5. Among them, 2,031 patients survived (94.4%), while 121 patients (5.6%) died during hospitalization. The patient characteristics including all analysed parameters are shown in Table 2 divided into two subgroups according to the NACA severity. Table 2Patients’ characteristicsNACA 4 (N = 1,684, 78.3%)NACA 5 (N = 468, 21.7%)Survivors (N = 1,643, 97.6%)Non-survivors (N = 41, 2.4%)Survivors (N = 388, 82.9%)Non-survivors (N = 80, 17.1%)Age (mean ± SD, years)53.4 ± 22.7 (N = 1,630)86.0 ± 10.7 (N = 41)51.0 ± 20.7 (N = 385)72.8 ± 17.8 (N = 79)Injury mechanism (N, %)fall < 3 m: 526 (32.0%)fall > 3 m: 267 (16.3%)other fall: 34 (2.1%)pedestrian road accident: 50 (3.0%)road accident: 543 (33.0%)othera: 223 (13.6%)fall < 3 m: 28 (68.3%)fall > 3 m: 9 (22.0%)other fall: 0 (0%)pedestrian road accident: 0 (0%)road accident: 3 (7.3%)othera: 1 (2.4%)fall < 3 m: 78 (20.1%)fall > 3 m: 87 (22.4%)other fall: 6 (1.5%)pedestrian road accident: 20 (5.2%)road accident: 137 (35.3%)othera: 60 (15.3%)fall < 3 m: 40. (50.0%)fall > 3 m: 17 (21.3%)other fall: 4 (5.0%)pedestrian road accident: 2 (2.5%)road accident: 10 (12.5%)othera: 7 (8.8%)GCS prehospital (mean ± SD)14.3 ± 1.5 (N = 1,613)12.8 ± 2.7 (N = 41)10.9 ± 4.4 (N = 385)7.0 ± 4.1 (N = 79)CPR prehospital (N, %)5 (0.3%) (N = 1,642)1 (2.4%) (N = 41)5 (1.3%) (N = 388)3 (3.7%) (N = 80)Intubation prehospital (N, %)16 (1%) (N = 1,642)2 (4.9%) (N = 41)146 (37.6%) (N = 388)58 (72.5%) (N = 80)Vasoactive drugs prehospital (N, %)13 (0.9%) (N = 1,472)1 (2.7%) (N = 37)47 (13.6%) (N = 345)20 (29.8%) (N = 67)Fluids prehospital (N, %, mL)0−500: 1,295 (83.4%)501−1000: 215 (13.8%)1,001−1,500: 32 (2.1%)> 1,500: 11 (0.7%)0−500: 38 (92.7%)501−1,000: 3 (7.3%)1,001−1,500: 0 (0%)> 1,500: 0 (0%)0−500: 195 (52.7%)501−1,000: 110 (29.7%)1,001−1,500: 47 (12.7%)> 1,500: 18 (4.9%)0−500: 44 (58.7%)501−1,000: 22 (29.3%)1,001−1,500: 2 (2.7%)> 1,500: 7 (9.3%)GCS at ER (mean ± SD)14.3 ± 1.7 (N = 1,635)11.6 ± 3.4 (N = 41)9.4 ± 5.5 (N = 386)4.7 ± 3.7 (N = 80)Hemoglobin at ER (mean ± SD, g/dL)13.9 ± 1.8 (N = 1,609)12.2 ± 2.1 (N = 41)13.1 ± 2.2 (N = 386)11.9 ± 2.4 (N = 77)Thrombocytes at ER (mean ± SD, 109/L)243.9 ± 73.4 (N = 1,607)211.1 ± 71.3 (N = 41)239 ± 77.6 (N = 386)197.8 ± 74.2 (N = 75)INR at ER (mean ± SD)1.1 ± 0.3 (N = 1,564)1.3 ± 0.7 (N = 39)1.1 ± 0.3 (N = 384)1.4 ± 0.7 (N = 71)ICU stay418 (25.4%) (N = 1,643)18 (43.9%) (N = 41)279 (71.9%) (N = 388)55 (68.7%) (N = 80)Mechanical ventilation138 (8.9%) (N = 1,549)15 (36.6%) (N = 41)218 (57.5%) (N = 379)54 (67.5%) (N = 80)ISS ≥ 16 (N, %)312 (19.0%) (N = 1,643)22 (53.7%) (N = 41)188 (48.4%) (N = 388)71 (88.7%) (N = 80)≥ 2 injured regionsb (N, %)1,067 (64.9%) (N = 1,643)26 (63.4%) (N = 41)282 (72.7%) (N = 388)48 (60.0%) (N = 80)Abbreviations: NACA National Advisory Committee for Aeronautics score, GCS Glasgow Coma Scale, CPR cardiopulmonary resuscitation, ER Emergency Room, INR International Normalized Ratio, ICU intensive care unit, ISS injury Severity Score, N number of patients, SD standard deviation, m meters, mL milliliters, g/dL grams/deciliter, L literaOther includes: shooting, stabbing, hitting, explosion, avalanche and unknownbPossible injured regions: head, face, neck, cervical spine, thoracic spine, lumbar spine, thorax, abdomen, upper extremity, lower extremity and pelvis, external (skin) and thermal injuries Patients’ characteristics fall < 3 m: 526 (32.0%) fall > 3 m: 267 (16.3%) other fall: 34 (2.1%) pedestrian road accident: 50 (3.0%) road accident: 543 (33.0%) othera: 223 (13.6%) fall < 3 m: 28 (68.3%) fall > 3 m: 9 (22.0%) other fall: 0 (0%) pedestrian road accident: 0 (0%) road accident: 3 (7.3%) othera: 1 (2.4%) fall < 3 m: 78 (20.1%) fall > 3 m: 87 (22.4%) other fall: 6 (1.5%) pedestrian road accident: 20 (5.2%) road accident: 137 (35.3%) othera: 60 (15.3%) fall < 3 m: 40. (50.0%) fall > 3 m: 17 (21.3%) other fall: 4 (5.0%) pedestrian road accident: 2 (2.5%) road accident: 10 (12.5%) othera: 7 (8.8%) 0−500: 1,295 (83.4%) 501−1000: 215 (13.8%) 1,001−1,500: 32 (2.1%) > 1,500: 11 (0.7%) 0−500: 38 (92.7%) 501−1,000: 3 (7.3%) 1,001−1,500: 0 (0%) > 1,500: 0 (0%) 0−500: 195 (52.7%) 501−1,000: 110 (29.7%) 1,001−1,500: 47 (12.7%) > 1,500: 18 (4.9%) 0−500: 44 (58.7%) 501−1,000: 22 (29.3%) 1,001−1,500: 2 (2.7%) > 1,500: 7 (9.3%) Abbreviations: NACA National Advisory Committee for Aeronautics score, GCS Glasgow Coma Scale, CPR cardiopulmonary resuscitation, ER Emergency Room, INR International Normalized Ratio, ICU intensive care unit, ISS injury Severity Score, N number of patients, SD standard deviation, m meters, mL milliliters, g/dL grams/deciliter, L liter aOther includes: shooting, stabbing, hitting, explosion, avalanche and unknown bPossible injured regions: head, face, neck, cervical spine, thoracic spine, lumbar spine, thorax, abdomen, upper extremity, lower extremity and pelvis, external (skin) and thermal injuries Patients’ characteristics: survivors vs non-survivors The results of the comparative analysis between survivors and non-survivors, including all patients as well as the NACA 4 and NACA 5 subgroups evaluated separately, found significant differences (p < 0.001) in all prehospital and hospital variables, as well as within “age”. In both NACA 4 and NACA 5 subgroups, most parameters were statistically significant. Key variables included age, GCS (prehospital and hospital), prehospital intubation, hemoglobin, thrombocyte count, and INR (all p < 0.0001). In the NACA 4 subgroup, prehospital CPR (p = 0.024), ICU admission (p = 0.008), and mechanical ventilation (p < 0.0001) differed significantly between survivors and non-survivors. In the NACA 5 subgroup, prehospital use of vasoactive drugs (p = 0.001) and fluids (p = 0.040), as well as injuries in ≥ 2 regions (p = 0.024) differed significantly between survivors and non-survivors. A significant difference was identified for the injury severity defined by an ISS ≥ 16 between survivors and non-survivors both in the main analysis and in the two sub-analyses (p < 0.0001). Univariate analysis For the entire patient cohort, age and all analysed prehospital, and hospital collected variables were identified as significant risk factors influencing mortality (Table 3). Table 3Univariate logistic regression for predictors of mortalityORa All patientsp valuebORa NACA 4p valuebORa NACA 5p valuebAge1.07 (1.05–1.08)p < 0.0001*1.14 (1.10–1.18)p < 0.0001*1.06 (1.05–1.08)p < 0.0001*NACA score8.26 (5.58–12.23)p < 0.0001*----GCS prehospital0.76 (0.73–0.79)p < 0.0001*0.78 (0.71–0.86)p < 0.0001*0.83 (0.78–0.88)p < 0.0001*CPR prehospital6.91 (2.13–22.35)p = 0.001*8.19 (0.93–71.67)p = 0.0582.98 (0.70–12.75)p = 0.140Intubation prehospital11.34 (7.67–16.77)p < 0.0001*5.21 (1.16–23.45)p = 0.031*4.37 (2.57–7.44)p < 0.0001*Vasoactive drugs prehospital7.41 (4.30–12.76)p < 0.0001*3.12 (0.40–24.48)p = 0.2792.70 (1.47–4.95)p = 0.001*Fluids prehospitalc4.39 (1.87–10.31)p = 0.001*0.48 (0.15–1.55)p = 0.2190.19 (0.04–0.81)p = 0.024*GCS at ER0.77 (0.74–0.80)p < 0.0001*0.74 (0.69–0.80)p < 0.0001*0.82 (0.77–0.88)p < 0.0001*Hemoglobin at ER0.70 (0.64–0.76)p < 0.0001*0.68 (0.60–0.78)p < 0.0001*0.81 (0.73–0.89)p < 0.0001*Thrombocytes at ER0.99 (0.987–0.993)p < 0.0001*0.99 (0.987–1.0)p = 0.003*0.99 (0.987–0.995)p < 0.0001*INR at ER2.90 (2.13–3.93)p < 0.0001*2.40 (1.55–3.83)p < 0.0001*4.45 (2.27–8.74)p < 0.0001*ICU stay2.91 (2.00–4.24)p < 0.0001*2.29 (1.23–4.29)p = 0.009*0.86 (0.51–1.45)p = 0.570Mechanical ventilation5.86 (4.02–8.55)p < 0.0001*5.90 (3.05–11.4)p < 0.0001*1.53 (0.92–2.56)p = 0.100ISS ≥ 1610.17 (6.59–15.70)p < 0.0001*4.94 (2.64–9.24)p < 0.0001*8.39 (4.08–17.27)p < 0.0001* ≥ 2 injured regions0.80 (0.55–1.16)p = 0.2360.94 (0.49–1.78)p = 0.8400.56 (0.34–0.93)p = 0.025*Abbreviations: NACA National Advisory Committee for Aeronautics score, GCS Glasgow Coma Scale, CPR cardiopulmonary resuscitation, ER Emergency Room, INR International Normalized Ratio, ICU intensive care unit, ISS injury Severity Score, OR odds ratioaOR and 95% confidence interval in brackets, OR > 1: death more likely if variable is higher or present, OR < 1: death more likely if variable is lower or absentbsignificance level *p < 0.05cdifferent groups were analysed against each other, most significant association is indicated Univariate logistic regression for predictors of mortality Abbreviations: NACA National Advisory Committee for Aeronautics score, GCS Glasgow Coma Scale, CPR cardiopulmonary resuscitation, ER Emergency Room, INR International Normalized Ratio, ICU intensive care unit, ISS injury Severity Score, OR odds ratio aOR and 95% confidence interval in brackets, OR > 1: death more likely if variable is higher or present, OR < 1: death more likely if variable is lower or absent bsignificance level *p < 0.05 cdifferent groups were analysed against each other, most significant association is indicated Higher age, higher NACA score, prehospital CPR, intubation, administration of fluids or vasoactive drugs, need for ICU admission, mechanical ventilation, elevated INR levels were all associated with higher mortality risk. Lower prehospital and hospital GCS, as well as reduced hemoglobin and thrombocyte levels, were also significant predictors for increased mortality risk. Additionally, the severity of sustained injuries, as indicated by an ISS ≥ 16, was also identified as a significant predictor of mortality. Subgroup analysis confirmed significant associations with mortality for most variables for both NACA 4 and NACA 5 subgroups individually, particularly age, prehospital and hospital GCS, prehospital intubation, laboratory values – including hemoglobin and thrombocytes – and ISS ≥ 16. Some variables showed subgroup-specific effects: in the NACA 4 group, the need for ICU admission (OR 2.29, p = 0.009) and use of mechanical ventilation (OR 5.90, p < 0.0001) were significantly associated with mortality risk. In the NACA 5 subgroup, prehospital administration of vasoactive drugs (OR 2.70, p = 0.001), fluids (OR 0.19, p = 0.024), and sustaining injuries in ≥ 2 anatomical regions (OR 0.56, p = 0.025), were significant predictors of mortality. Multivariate analysis The multivariate analysis of the prehospital variables (Table 4) showed that age (OR 1.11, p < 0.0001), NACA score (OR 2.96, p = 0.002), GCS (OR 0.78, p < 0.0001), CPR (OR 7.22, p = 0.039), and administration of vasoactive drugs (OR 2.38, p = 0.037) were statistically significant independent risk factors for all included patients, while intubation and administration of liquids were not. Table 4Multivariate logistic regression for predictors of mortality with prehospital variablesORa All patients (N = 1,818)p valuebORa NACA 4(N = 1,427)p valuebORa NACA 5(N = 391)p valuebAge1.11 (1.08–1.13)p < 0.0001*1.13 (1.09–1.18)p < 0.0001*1.09 (1.06–1.12)p < 0.0001*NACA score2.96 (1.48–5.93)p = 0.002*----GCS prehospital0.78 (0.72–0.85)p < 0.0001*0.69 (0.58–0.81)p < 0.0001*0.82 (0.75–0.90)p < 0.0001*CPR prehospital7.22 (1.11–47.00)p = 0.039*14.02 (0.90–218.3)p = 0.0596.02 (0.38–94.57)p = 0.202Intubation prehospital1.56 (0.68–3.53)p = 0.2920.58 (0.03–10.46)p = 0.7111.81 (0.78–4.21)p = 0.167Vasoactive drugs prehospital2.38 (1.05–5.37)p = 0.037*0.47 (0.01–19.90)p = 0.6962.40 (1.05–5.50)p = 0.039*Fluids prehospital0.84 (0.55–1.28)p = 0.4220.70 (0.22–2.27)p = 0.5560.87 (0.56–1.36)p = 0.550Abbreviations: NACA National Advisory Committee for Aeronautics score, GCS Glasgow Coma Scale, CPR cardiopulmonary resuscitation, OR odds ratioaOR and 95% confidence interval in brackets, OR > 1: death more likely if variable is higher or present, OR < 1: death more likely if variable is lower or absentbsignificance level *p < 0.05 Multivariate logistic regression for predictors of mortality with prehospital variables Abbreviations: NACA National Advisory Committee for Aeronautics score, GCS Glasgow Coma Scale, CPR cardiopulmonary resuscitation, OR odds ratio aOR and 95% confidence interval in brackets, OR > 1: death more likely if variable is higher or present, OR < 1: death more likely if variable is lower or absent bsignificance level *p < 0.05 In the NACA 4 subgroup analysis only age (OR 1.13, p < 0.0001) and GCS (OR 0.69, p < 0.0001) emerged as statistically significant predictors of mortality. For the NACA 5 patients age (OR 1.09, p < 0.0001), GCS (OR 0.82, p < 0.0001), and administration of vasoactive drugs (OR 2.4, p = 0.039) emerged as statistically significant predictors of mortality. The dedicated statistical checks performed to account for potential collinearity among independent variables showed all pairwise correlation coefficients between predictors below 0.58, excluding strong correlations. Variance Inflation Factors (VIFs) were consistently < 2.3, indicating the absence of problematic multicollinearity. Condition index analysis revealed values > 30, suggesting some degree of multivariable dependency, most likely reflecting the expected clinical overlap between severity scores and pre-hospital interventions. However, these dependencies did not compromise model stability or overall fit, and exclusion of correlated variables did not improve model performance, thus confirming the study findings. Discussion The main finding of this study is that different factors can influence the mortality risk beyond the mere NACA 4 and 5 classification of polytrauma patients, including CPR, administration of vasoactive drugs, GCS, and age. Previous studies investigated the NACA score showing good correlations between the NACA score and survival as well as a strong ability to predict mortality for trauma patients [11, 12]. This study builds upon these findings, confirming such results for NACA 4 and NACA 5 patients. There is a significant difference in NACA score between survivors and non-survivors. This study demonstrated that the distinction between NACA 4 and NACA 5 was associated with an almost 300% increase in mortality risk. Hence, it can be regarded as a valid score for the early prehospital assessment of mortality in trauma patients. This study found significant differences between survivors and non-survivors in both prehospital and hospital factors. In addition to the NACA score, key mortality predictors included age, prehospital and hospital GCS, need for advanced interventions (e.g., resuscitation, intubation, vasoactive drugs, fluids), ICU admission, mechanical ventilation, specific laboratory values, and ISS ≥ 16. In the multivariate analysis, age, GCS, and NACA score emerged as the strongest independent predictors of mortality, highlighting their value in guiding early management before hospital admission. Older age and low GCS have been previously identified as independent mortality predictors in trauma patients and were therefore added to various definitions and scores related to trauma to increase their predictive power on survival [6, 7, 16–18]. This study confirmed both factors as significant predictors in multivariate analyses in all patient groups. Several other risk factors have been evaluated for their impact on mortality in polytrauma patients. Among these, traumatic cardiac arrest is recognized as both rare and highly fatal. Seewald et al. reported 4% of cardiac arrests stemmed from major trauma, with a 95% mortality rate [19]. The present study specifically analysed prehospital CPR in trauma patients as a risk factor for mortality and found significant results when including all NACA 4 and NACA 5 patients in the multivariate analysis, indicating that the necessity of prehospital resuscitation has the biggest impact on mortality. Prehospital intubation underlined controversial findings, not reaching statistical significance in the multivariate analysis. Previous investigations reported mixed findings regarding the impact of prehospital intubation and focused primarily on the comparison between prehospital and hospital intubation without examining prehospital intubation as an independent risk factor for mortality [20–22]. Additionally, they focused on patients with isolated traumatic brain injury (TBI) rather than on the general polytraumatized population. The findings of the present study build upon these previous results suggesting that prehospital intubation could play a role, albeit not being the most relevant factor influencing mortality in polytrauma patients. Fluids or vasoactive drugs are often used for prehospital hemodynamic support in polytrauma patients. Counterintuitively a decreased risk of death was found for the separate analysis of the NACA 5 subgroup. This apparent contradiction may reflect the heterogeneity between NACA 4 and 5 patients, with fluid use potentially indicating greater injury severity in the former, and timely intervention in the latter [18]. The administration of a greater amount of fluids may be a sign of earlier recognition of hemodynamic instability and earlier treatment leading to increased survival. The administration of vasoactive drugs was significantly associated with increased mortality both overall and within the NACA 5 subgroup, aligning with previous studies [23, 24]. Multivariate analysis showed that fluid administration was not an independent predictor of mortality, while vasoactive drug use significantly increased mortality in both the overall cohort and the NACA 5 subgroup. These findings support the NACA score’s ability to distinguish between patients at immediate (NACA 5) and potential (NACA 4) risk of life-threatening conditions. Concerning the parameters collected in the ED, parameters such as low GCS, hemoglobin, thrombocyte levels, and coagulopathy were confirmed as mortality risk predictors, in line with previous studies [12, 25–27]. Conversely, the impact of ICU admission and mechanical ventilation with its influence on mortality in polytrauma patients is less well studied, as most prior research focused solely on ICU-admitted populations [28–30]. The current study showed a significant association with mortality in the overall cohort and the NACA 4 subgroup, while the NACA 5 subgroup did not indicate an association between ICU admission and mechanical ventilation with patients’ survival. This may reflect a possible confounding factor, as more severely injured NACA 5 patients may have died before ICU admission or before mechanical ventilation could be initiated and registry documented. The assessment of injury severity measured by an ISS ≥ 16 is another widely acknowledged as a strong predictor of mortality. Current findings reinforce this association in both NACA 4 and NACA 5 subgroups. However, the injury distribution analysis showed that involvement of multiple regions was significantly associated with survival only in NACA 5. A lower mortality risk was found in patients with multiple injured regions, suggesting that certain isolated injuries, such as severe traumatic brain injury or major thoracic trauma, may carry a disproportionately higher mortality risk than multi-region injuries of moderate severity. Similarly, Paffrath et al. reported lower mortality (18.7%) in patients with multiple injuries versus 20.4% overall and 31% in monotrauma cases (AIS 4 or 5), highlighting the need to integrate physiological criteria into major trauma definitions to improve mortality risk stratification [7]. This study presents several limitations that must be noted. Its registry-based design may have introduced biases due to inconsistent or missing data, particularly regarding vital signs like arterial blood pressure and further laboratory results including lactate and base deficit. The absence of information on comorbidities and systemic inflammatory response (SIRS) criteria further limits the analysis. Additionally, being conducted in a single trauma center can also introduce a potential bias due to locally established treatment algorithms. While the multivariate analysis helped providing a control of confounding bias, residual confounding bias could not be excluded. A further limitation of the study is the potential multicollinearity between NACA score, GCS, and ISS, as these variables are conceptually and statistically related. Even though dedicated statistical checks were performed to account for it and the multivariate model identified certain parameters as independent predictors, correlations between them may attenuate or obscure their true individual effects. Future research using collinearity diagnostics or penalized regression techniques could help to further disentangle these associations. Despite these limitations, the study offers valuable insights into mortality risk factors in polytrauma patients. A major strength lies in the large and diverse cohort, enabling a broad evaluation of predictors across different trauma severities and care settings. The inclusion of extensive prehospital and hospital variables, along with both univariate and multivariate analyses, enhances the reliability and depth of the findings. Analysing NACA 4 and NACA 5 subgroups separately revealed meaningful differences in mortality predictors, highlighting the importance of tailored clinical approaches. The findings suggest that factors such as CPR, vasoactive drug use, GCS, and age significantly impact mortality, beyond the NACA classification alone. Recognizing these parameters early may support more targeted interventions. The NACA classification system has been explored throughout various studies, with some concerns on the repeatability of this system in trauma patients [9, 11, 31, 32]. This study suggested areas for improvement of this classification system. The prognostic accuracy of the NACA classification could be improved by complementing it with additional indicators such as age, GCS, need for reanimation, and use of vasoactive drugs. While the NACA score remains a valuable tool for rapid prehospital prognostic assessment, integrating these variables into an expanded or revised scoring system may provide a more precise estimation of in-hospital mortality risk, thereby supporting more informed triage and treatment decisions in the prehospital setting. Overall, this study contributes to the ongoing optimization of polytrauma care. Improvements in prehospital response, early injury recognition, and trauma system organization are key to enhancing the outcome and the survival rate of these patients. Future prospective studies with standardized clinical data and long-term outcomes are needed to better understand how injury severity, interventions, and individual patient factors interact in shaping prognosis and defining optimal management of these challenging patients. Acknowledgements The authors thank Prof. Dr. med. Alberto Pagnamenta for his work in the statistical analysis. Abbreviations Abbreviated Injury Scale Cardiopulmonary resuscitation Emergency department Emergency room Glasgow Coma Scale Intensive care unit International Normalized Ratio Injury Severity Score Number of patients National Advisory Committee for Aeronautics Odds ratio Standard deviation Systemic inflammatory response Swiss Trauma Registry Traumatic Brain Injury Variance Inflation Factor Authors’ contributions IS, MR and GF designed the study. IS, MR, AB and GF analysed the data. IS drafted the manuscript. MR, AB, ASM, PIF and GF reviewed the manuscript. All authors read and approved the final manuscript. Funding No funding. Data availability No datasets were generated or analysed during the current study. Declarations Ethics approval and consent to participate The study protocol was approved by the local cantonal ethics committee of the Canton of Tessin, Switzerland (prot 2024-00236 CE 4536). Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement No datasets were generated or analysed during the current study.