Prehospital management and outcomes of traumatic cardiac arrest in older adults: a French nationwide registry study Authors: Axel Benhamed, Baptiste Morcel, Jean-Stephane David, Ewoud Ter Avest, Tobias Gauss, James van Oppen, Karim Tazarourte, Eric Mercier, Bages-Limoges Florence, Bages-Limoges Florence, Dhers Marion, Fuster Patrick, Bellow Tobola, Peguet Olivier, Besnier Sylvie, Jardel Benoit, Buleon Clement, Avondo Aurelie, Gillet Stephane, Monchard Franck, Guillaumee Frederic, Jacob Line, Kindle Carine, Marquet Jean-Luc, Watrelot Olivier, Cheminat Claude, Gonzalez Denis, Blottiaux Emmanuel, Branche Fabienne, Delemar Thibault, Ricard-Hibon Agnes, Curunet-Raoul Anne Dominique, Serre Patrice, du Besset Marc, Poincet Geraldine, Lefevre Nathalie, Puchois Aurelien, Leraitre Thierry, Wyckaert Geoffrey, Capel Olivier Published: 2026 ABSTRACT BackgroundTraumatic cardiac arrest (TCA) carries an overall poor prognosis, yet the impact of advanced age on prehospital management and outcomes remains poorly characterised. Whether older patients with TCA present distinct clinical profiles or respond differently to resuscitation compared with younger adults is largely unknown.MethodsWe conducted a multicenter cohort study using data from the French National Out-of-Hospital Cardiac Arrest Registry (RéAC). All adult patients (≥18 years) with out-of-hospital TCA were included. Patients were categorised as adults (18–64 years) or older adults (≥65 years). The primary outcome was return of spontaneous circulation (ROSC). Secondary outcomes included survival at hospital admission (D0), 30-day survival (D30), and favourable neurological outcome (Cerebral Performance Category 1–2 at D30). Univariate and multivariable logistic regression analyses were performed. A pre-specified subgroup analysis stratified older patients into three age categories (65–74, 75–84, ≥85 years).ResultsAmong 5595 patients (4249 younger adults; 1346 older adults [24.1%]), older patients more often achieved ROSC (22.5% vs 18.6%; OR 1.22, 95%CI 1.03–1.45) but were less likely to survive to 30 days (0.5% vs 1.2%; OR 0.34, 95%CI 0.15–0.81). Survival at hospital admission did not differ after adjustment (16.1% vs 13.3%; OR 1.14, 95%CI 0.95–1.38). Older patients more often sustained a fall (52.2% vs 28.5%) and less often sustained thoracic injuries (43.8% vs 59.6%). They received fewer trauma-specific prehospital interventions, most notably thoracostomy (15.2% vs 23.9%) and fluid resuscitation (28.3% vs 37.4%).ConclusionsIn this large French registry study, older patients with TCA achieved return of spontaneous circulation more often than younger adults but were less likely to survive to 30 days. Age alone should not preclude resuscitative efforts, but a concurrent or precipitating medical cause of arrest should be systematically considered in this population. Key messages What is already known on this subject Traumatic cardiac arrest carries a poor prognosis with overall survival rates below 5–8%. Most existing evidence on TCA is derived from predominantly young adult cohorts, and the few studies examining the effect of age have reported conflicting findings. What this study adds In a large French national registry, older TCA patients (≥65 years) had higher ROSC rates than younger adults but significantly lower 30-day survival. How this might change practice Age alone should not discourage resuscitative efforts. Clinicians managing older patients presenting as TCA should systematically consider concurrent or precipitating medical causes of arrest, particularly when injury severity appears low or the mechanism is unclear. Introduction Traumatic cardiac arrest (TCA) occurs when severe injury leads to cardiac standstill through mechanisms including exsanguination, hypoxia, tension pneumothorax, or cardiac tamponade.1 Although historically considered a near-uniformly fatal condition, TCA is no longer regarded as futile. Reported prognosis nonetheless remains heterogeneous across studies, depending on case-mix, the cause of arrest, and reporting methods, with survival rates ranging from 2% to 8% in large registry studies.2, 3, 4 Some factors associated with return of spontaneous circulation in TCA have been identified, including non-asystolic initial rhythm and cardiac motion on ultrasound,5, 6 while the impact of age is still debated. These outcomes have been attributed to advances in prehospital care algorithms that emphasise the rapid identification and correction of reversible causes, including external haemorrhage control, oxygenation, chest decompression, and early haemostatic resuscitation.7 However, despite increased adherence to guideline-recommended trauma-specific interventions, their impact on patient-centred outcomes remains uncertain.3, 8 The global population is ageing rapidly leading to a shift in trauma patient demographics being older with more severe trauma.9, 10, 11 Notably, older individuals are increasingly exposed to mechanisms of injury capable of causing TCA, particularly motor vehicle collisions and falls from height.9 However, the epidemiology and outcomes of TCA in older patients remain largely understudied. Analyses based on the Japan Trauma Data Bank reported a counter-intuitive finding: survival to discharge was higher in older adults (≥60 years) compared with non-older patients with TCA even after adjustment for injury severity.12 However, the authors only included patients with blunt trauma and did not report long-term outcome. More recently, some authors analysed the US National EMS Information System and similarly found that older adults with TCA had higher rates of ROSC compared with younger adults.13 However, they reported no data on prehospital management nor on long-term survival. Yet, these findings raise important questions about the clinical profile of TCA in older patients and whether the current paradigm of TCA management, developed predominantly from young adult cohorts, is equally applicable to older patients. In France, the physician-staffed prehospital emergency medical system (EMS) provides on-scene advanced resuscitation for TCA patients, offering a unique perspective on the prehospital phase of care.7 Our primary objective was to compare ROSC rates between younger and older adult patients with out-of-hospital TCA. Secondary objectives were to compare survival at hospital admission (D0), 30-day survival (D30), favourable neurological outcome (CPC 1–2 at D30), prehospital management, and to explore outcomes across older age subgroups. Methods Study design We conducted a retrospective multicenter cohort study using prospectively collected data from the RéAC registry. This manuscript was prepared in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement.14 Setting and data source The French prehospital emergency medical system (EMS) operates on a two-tiered model. In cases of out-of-hospital cardiac arrest (OHCA), both a fire brigade ambulance and a mobile medical team are dispatched simultaneously to the scene. A bystander may initiate cardiopulmonary resuscitation (CPR) and, if available, defibrillation using a public-access automated external defibrillator (AED) before EMS arrival, often under the guidance of the emergency dispatcher. The fire brigade typically arrives first, takes over from any bystander, and continues basic life support (BLS). Fire brigade personnel are also trained to perform trauma-specific interventions such as external haemorrhage control, spinal motion restriction, and pelvic stabilisation. The mobile medical team consists of an emergency physician, a nurse, and an ambulance driver, and can be dispatched via either ground or helicopter ambulance. The mobile medical team provides advanced life support (ALS), including advanced airway management, intravenous or intraosseous access, administration of vasopressors, fluid resuscitation, blood transfusion, and, when indicated, surgical procedures such as thoracostomy at the scene.15 Participants We included all adult patients (≥18 years) registered in the RéAC database with an out-of-hospital traumatic cardiac arrest as defined by Utstein criteria,16 including those who died before hospital admission. We excluded cases with spontaneous circulation at mobile medical team arrival, aetiology other than traumatic (medical, asphyxia, overdose, electrocution, drowning) or unknown, burns as primary or contributing cause, signs of irreversible death (rigour mortis, lividity), documented do-not-attempt-resuscitation orders or end-of-life status, and absence of resuscitation by the mobile medical team. We categorised patients into two groups: younger adults (18–64 years) and older adults (≥65 years). The older group was further stratified into three pre-specified subgroups: 65–74, 75–84, and ≥85 years. The threshold of 65 years was chosen because it is the conventional definition of the older adult in trauma epidemiology and in most national trauma and cardiac arrest registries, it corresponds to the age from which geriatric-specific trauma triage criteria are generally applied, and it allows comparison with the previous study of TCA in older adults.17 Variables Mechanism of trauma was classified as blunt or penetrating. Cause of trauma included motor vehicle accident, fall, firearm injury, and stab wound. The registry did not differentiate falls from standing height from falls from a greater height. Site of injury (head, thoracic, abdominal, limbs) was recorded on the basis of the on-scene clinical assessment by the prehospital physician, without systematic radiological or post-mortem confirmation. Location of arrest was categorised as public (e.g. street or other public space) or private (e.g. home or other indoor domestic setting). Witness status was categorised as unwitnessed, bystander-witnessed, or EMS-witnessed (fire brigade or mobile medical team). Bystander CPR and bystander AED use were documented. Response times for fire brigade and mobile medical team were recorded in minutes. Initial cardiac rhythm was classified as asystole, pulseless electrical activity (PEA), or ventricular fibrillation/ventricular tachycardia (VF/VT). Prehospital interventions recorded included epinephrine administration, external haemorrhage control, fluid resuscitation, blood transfusion, endotracheal intubation, and thoracostomy. Outcomes The primary outcome was return of spontaneous circulation (ROSC). Secondary outcomes were: (1) survival at hospital admission (D0); (2) 30-day survival (D30); and (3) favourable neurological outcome defined as Cerebral Performance Category (CPC) 1–2 at D30. Statistical analysis Categorical variables were expressed as numbers and percentages and continuous variables as medians with interquartile ranges (IQR). Comparisons between groups used chi-squared tests or Fisher’s exact test for categorical variables and Wilcoxon rank-sum tests for continuous variables. Missing covariate data were handled using multiple imputation by chained equations (10 imputed datasets, 5 iterations), with logistic regression for categorical variables and predictive mean matching for continuous variables. Separate logistic regression models were fitted for each outcome with age group (≥65 vs. <65 years) as the main exposure of interest. Univariate logistic regression was performed to estimate crude odds ratios (ORs) with 95% confidence intervals (CIs). Multivariablee logistic regression was used to estimate adjusted OR (and 95% CI) for potential confounders selected on the basis of prior literature and clinical relevance (sex, mechanism of trauma, site(s) of injury, witness status, bystander CPR and AED use, fire brigade and mobile medical team response times, signs of life and initial cardiac rhythm at mobile medical team arrival as well as prehospital interventions). Complete adjusted estimates for all covariates included in each multivariable model are presented in Supplementary Tables S1–S4. A pre-specified subgroup analysis repeated these analyses within the older age strata (65–74, 75–84, ≥85 years), using the younger adult group (18–64 years) as the reference. As a sensitivity analysis, multivariate logistic regression models were fitted separately on each of the 10 imputed datasets and on the original complete-case dataset. The resulting estimates were compared with the pooled estimates to assess the potential impact of missing data and the imputation procedure on the findings (Supplementary Table S5). A p-value < 0.05 was considered statistically significant. Ethics The registry received approval from the French national data protection commission (Commission Nationale de l’Informatique et des Libertés, CNIL) [#910946] and approval from the French advisory committee on information processing in material research in the field of health (Comité Consultatif sur le Traitement de l’Information en Matière de Recherche dans le Domaine de la Santé) [#10.326Ter]. In line with the French legislation on retrospective studies of registry data, patient consent was not required. Results Population characteristics A total of 5595 adult patients with out-of-hospital TCA were included (Fig. 1): 4249 (75.9%) in the younger adult group (18–64 years) and 1346 (24.1%) in the older adult group (≥65 years). Within the older adult group, 622 (46.2%) were aged 65–74 years, 489 (36.3%) were aged 75–84 years, and 235 (17.5%) were aged ≥85 years. The median age was 39 years (IQR 27–51) in the younger adult group and 75 years (IQR 70–82) in the older adult group. The proportion of males was lower in older adults (65.4% vs. 80.4%, p < 0.001). Blunt trauma was the most common mechanism of injury in both groups (86.7% vs. 78.3%, p < 0.001), with falls representing the leading cause in older adults (52.2% vs. 28.5%) and motor vehicle accidents in younger adults (38.5% vs. 50.1%) (p < 0.001). Head injuries were slightly more frequent in older adult patients (78.8% vs. 75.8%, p = 0.033), while thoracic injuries (43.8% vs. 59.6%, p < 0.001), abdominal injuries (26.1% vs. 36.0%, p < 0.001), and limb injuries (18.9% vs. 23.5%, p < 0.001) were more frequently observed in younger adults. Initial cardiac rhythm did not differ significantly between groups (p = 0.258), most patients exhibited non-shockable rhythms: asystole (87.6% vs. 88.7%) or pulseless electrical activity (9.5% vs. 9.9%). However, signs of life were more frequent in older adults (7.1% vs. 5.2%, p = 0.010).Fig. 1Flow chart. Flow chart. Regarding mobile medical team prehospital management, we found that epinephrine was administered less frequently to older adult patients (94.7% vs. 97.0%, p < 0.001), as were fluid resuscitation (28.3% vs. 37.4%, p < 0.001), blood transfusion (3.4% vs. 5.0%, p = 0.018), intubation (87.5% vs. 91.9%, p < 0.001), and thoracostomy (15.2% vs. 23.9%, p < 0.001) (Table 1). Subgroups characteristics are presented in Table 2.Table 1Prehospital characteristics of younger adult and older adult patients with out-of-hospital traumatic cardiac arrest.Younger adults (18–64 years)n = 4249Older adults (≥65 years)n = 1346p-valueAge, years39 [27–51]75 [70–82]<0.001Sex<0.001 Male3415 (80.4%)880 (65.4%) Female832 (19.6%)466 (34.6%)Mechanism of trauma<0.001 Blunt3327 (78.3%)1167 (86.7%) Penetrating922 (21.7%)179 (13.3%)Cause of trauma*<0.001 Motor vehicle accident1525 (50.1%)358 (38.5%) Fall866 (28.5%)485 (52.2%) Firearm328 (10.8%)69 (7.4%) Stab325 (10.7%)17 (1.8%)Site of injury† Head2903 (75.8%)949 (78.8%)0.033 Thoracic2282 (59.6%)528 (43.8%)<0.001 Abdominal1378 (36.0%)314 (26.1%)<0.001 Limbs899 (23.5%)228 (18.9%)<0.001Location<0.001 Public3025 (77.6%)665 (54.6%) Private873 (22.4%)553 (45.4%)Witness status<0.001 Unwitnessed914 (21.5%)385 (28.7%) Bystander2628 (61.9%)730 (54.4%) EMS702 (16.5%)228 (17.0%)Bystander CPR1429 (33.6%)428 (31.8%)0.515Bystander AED14 (0.3%)5 (0.4%)0.790EMS response times, min Fire brigade7 [2–10]7 [3–11]0.020 Mobile medical team15 [9–22]16 [10–24]0.002Signs of life221 (5.2%)95 (7.1%)0.010Initial cardiac rhythm0.258 Asystole3767 (88.7%)1179 (87.6%) PEA404 (9.5%)133 (9.9%) VF/VT78 (1.8%)34 (2.5%)Prehospital interventions Epinephrine4120 (97.0%)1275 (94.7%)<0.001 External haemorrhage control415 (9.8%)117 (8.7%)0.242 Fluid resuscitation1591 (37.4%)381 (28.3%)<0.001 Blood transfusion211 (5.0%)46 (3.4%)0.018 Intubation3903 (91.9%)1178 (87.5%)<0.001 Thoracostomy1015 (23.9%)205 (15.2%)<0.001Categories not mutually exclusive.AED = automated external defibrillator; CPR = cardiopulmonary resuscitation; EMS = emergency medical services (fire brigade or mobile medical team); IQR = interquartile range; PEA = pulseless electrical activity; VF/VT = ventricular fibrillation/ventricular tachycardia.Data are n (%) or median [IQR].*Missing: 1205 younger adults, 417 older adults.†Missing: 418 younger adults, 141 older adults.Table 2Prehospital characteristics of older adult patients by age subgroup.65–74 yearsn = 62275–84 yearsn = 489≥85 yearsn = 235p-valueAge, years69 [67–72]79 [77–82]87 [86–90]<0.001Sex0.081 Male426 (68.5%)305 (62.4%)149 (63.4%) Female196 (31.5%)184 (37.6%)86 (36.6%)Mechanism of trauma0.343 Blunt531 (85.4%)427 (87.3%)209 (88.9%) Penetrating91 (14.6%)62 (12.7%)26 (11.1%)Cause of trauma<0.001 Motor vehicle accident160 (37.6%)134 (40%)64 (37.2%) Fall223 (52.3%)165 (49.8%)97 (56%) Firearm37 (8.7%)22 (6.6%)10 (5.8%) Stab6 (1.4%)10 (3.0%)1 (0.6%)Site of injury Head426 (77.0%)346 (79.0%)177 (83.1%)0.136 Thoracic268 (48.5%)180 (41.1%)80 (37.6%)0.010 Abdominal159 (28.8%)113 (25.8%)42 (19.7%)0.042 Limbs108 (19.5%)82 (18.7%)38 (17.8%)0.872Location0.148 Public318 (57.2%)225 (51.0%)122 (55.2%) Private238 (42.8%)216 (49.0%)99 (44.8%)Witness status0.602 Unwitnessed190 (30.6%)129 (26.4%)66 (28.0%) Bystander326 (52.6%)277 (56.8%)127 (54.0%) EMS104 (16.8%)82 (16.8%)42 (18.0%)Bystander CPR205 (33.0%)150 (30.7%)73 (33%)0.556Bystander AED3 (0.6%)2 (0.5%)0 (0.0%)0.851EMS response times, min Fire brigade7 [3–12]8 [4–12]6 [3–10]0.156 Mobile medical team16 [10–24]17 [10–25]16 [9–22]0.343Signs of life37 (5.9%)37 (7.6%)21 (8.9%)0.269Initial cardiac rhythm0.445 Asystole554 (89.1%)426 (87.1%)199 (84.7%) PEA53 (8.5%)52 (10.6%)28 (12%) VF/VT15 (2.4%)11 (2.2%)8 (3.4%)Prehospital interventions Epinephrine597 (96.0%)463 (94.7%)215 (91.5%)0.052 External haemorrhage control47 (7.6%)48 (9.8%)22 (9.4%)0.383 Fluid resuscitation184 (29.6%)133 (27.2%)64 (27.2%)0.629 Blood transfusion27 (4.3%)15 (3.1%)4 (1.7%)0.143 Intubation544 (87.5%)442 (90.4%)192 (81.7%)0.002 Thoracostomy94 (15.1%)81 (16.6%)30 (12.8%)0.409Data are n (%) or median [IQR].AED = automated external defibrillator; CPR = cardiopulmonary resuscitation; EMS = emergency medical services; IQR = interquartile range; AED = automated external defibrillator; CPR = cardiopulmonary resuscitation; EMS = emergency medical services (fire brigade or mobile medical team); FB = fire brigade; IQR = interquartile range; PEA = pulseless electrical activity; VF/VT = ventricular fibrillation/ventricular tachycardia.; PEA = pulseless electrical activity; VF/VT = ventricular fibrillation/ventricular tachycardia. Table 1Prehospital characteristics of younger adult and older adult patients with out-of-hospital traumatic cardiac arrest.Younger adults (18–64 years)n = 4249Older adults (≥65 years)n = 1346p-valueAge, years39 [27–51]75 [70–82]<0.001Sex<0.001 Male3415 (80.4%)880 (65.4%) Female832 (19.6%)466 (34.6%)Mechanism of trauma<0.001 Blunt3327 (78.3%)1167 (86.7%) Penetrating922 (21.7%)179 (13.3%)Cause of trauma*<0.001 Motor vehicle accident1525 (50.1%)358 (38.5%) Fall866 (28.5%)485 (52.2%) Firearm328 (10.8%)69 (7.4%) Stab325 (10.7%)17 (1.8%)Site of injury† Head2903 (75.8%)949 (78.8%)0.033 Thoracic2282 (59.6%)528 (43.8%)<0.001 Abdominal1378 (36.0%)314 (26.1%)<0.001 Limbs899 (23.5%)228 (18.9%)<0.001Location<0.001 Public3025 (77.6%)665 (54.6%) Private873 (22.4%)553 (45.4%)Witness status<0.001 Unwitnessed914 (21.5%)385 (28.7%) Bystander2628 (61.9%)730 (54.4%) EMS702 (16.5%)228 (17.0%)Bystander CPR1429 (33.6%)428 (31.8%)0.515Bystander AED14 (0.3%)5 (0.4%)0.790EMS response times, min Fire brigade7 [2–10]7 [3–11]0.020 Mobile medical team15 [9–22]16 [10–24]0.002Signs of life221 (5.2%)95 (7.1%)0.010Initial cardiac rhythm0.258 Asystole3767 (88.7%)1179 (87.6%) PEA404 (9.5%)133 (9.9%) VF/VT78 (1.8%)34 (2.5%)Prehospital interventions Epinephrine4120 (97.0%)1275 (94.7%)<0.001 External haemorrhage control415 (9.8%)117 (8.7%)0.242 Fluid resuscitation1591 (37.4%)381 (28.3%)<0.001 Blood transfusion211 (5.0%)46 (3.4%)0.018 Intubation3903 (91.9%)1178 (87.5%)<0.001 Thoracostomy1015 (23.9%)205 (15.2%)<0.001Categories not mutually exclusive.AED = automated external defibrillator; CPR = cardiopulmonary resuscitation; EMS = emergency medical services (fire brigade or mobile medical team); IQR = interquartile range; PEA = pulseless electrical activity; VF/VT = ventricular fibrillation/ventricular tachycardia.Data are n (%) or median [IQR].*Missing: 1205 younger adults, 417 older adults.†Missing: 418 younger adults, 141 older adults. Prehospital characteristics of younger adult and older adult patients with out-of-hospital traumatic cardiac arrest. Categories not mutually exclusive. AED = automated external defibrillator; CPR = cardiopulmonary resuscitation; EMS = emergency medical services (fire brigade or mobile medical team); IQR = interquartile range; PEA = pulseless electrical activity; VF/VT = ventricular fibrillation/ventricular tachycardia. Data are n (%) or median [IQR]. Missing: 1205 younger adults, 417 older adults. Missing: 418 younger adults, 141 older adults. Table 2Prehospital characteristics of older adult patients by age subgroup.65–74 yearsn = 62275–84 yearsn = 489≥85 yearsn = 235p-valueAge, years69 [67–72]79 [77–82]87 [86–90]<0.001Sex0.081 Male426 (68.5%)305 (62.4%)149 (63.4%) Female196 (31.5%)184 (37.6%)86 (36.6%)Mechanism of trauma0.343 Blunt531 (85.4%)427 (87.3%)209 (88.9%) Penetrating91 (14.6%)62 (12.7%)26 (11.1%)Cause of trauma<0.001 Motor vehicle accident160 (37.6%)134 (40%)64 (37.2%) Fall223 (52.3%)165 (49.8%)97 (56%) Firearm37 (8.7%)22 (6.6%)10 (5.8%) Stab6 (1.4%)10 (3.0%)1 (0.6%)Site of injury Head426 (77.0%)346 (79.0%)177 (83.1%)0.136 Thoracic268 (48.5%)180 (41.1%)80 (37.6%)0.010 Abdominal159 (28.8%)113 (25.8%)42 (19.7%)0.042 Limbs108 (19.5%)82 (18.7%)38 (17.8%)0.872Location0.148 Public318 (57.2%)225 (51.0%)122 (55.2%) Private238 (42.8%)216 (49.0%)99 (44.8%)Witness status0.602 Unwitnessed190 (30.6%)129 (26.4%)66 (28.0%) Bystander326 (52.6%)277 (56.8%)127 (54.0%) EMS104 (16.8%)82 (16.8%)42 (18.0%)Bystander CPR205 (33.0%)150 (30.7%)73 (33%)0.556Bystander AED3 (0.6%)2 (0.5%)0 (0.0%)0.851EMS response times, min Fire brigade7 [3–12]8 [4–12]6 [3–10]0.156 Mobile medical team16 [10–24]17 [10–25]16 [9–22]0.343Signs of life37 (5.9%)37 (7.6%)21 (8.9%)0.269Initial cardiac rhythm0.445 Asystole554 (89.1%)426 (87.1%)199 (84.7%) PEA53 (8.5%)52 (10.6%)28 (12%) VF/VT15 (2.4%)11 (2.2%)8 (3.4%)Prehospital interventions Epinephrine597 (96.0%)463 (94.7%)215 (91.5%)0.052 External haemorrhage control47 (7.6%)48 (9.8%)22 (9.4%)0.383 Fluid resuscitation184 (29.6%)133 (27.2%)64 (27.2%)0.629 Blood transfusion27 (4.3%)15 (3.1%)4 (1.7%)0.143 Intubation544 (87.5%)442 (90.4%)192 (81.7%)0.002 Thoracostomy94 (15.1%)81 (16.6%)30 (12.8%)0.409Data are n (%) or median [IQR].AED = automated external defibrillator; CPR = cardiopulmonary resuscitation; EMS = emergency medical services; IQR = interquartile range; AED = automated external defibrillator; CPR = cardiopulmonary resuscitation; EMS = emergency medical services (fire brigade or mobile medical team); FB = fire brigade; IQR = interquartile range; PEA = pulseless electrical activity; VF/VT = ventricular fibrillation/ventricular tachycardia.; PEA = pulseless electrical activity; VF/VT = ventricular fibrillation/ventricular tachycardia. Prehospital characteristics of older adult patients by age subgroup. Data are n (%) or median [IQR]. AED = automated external defibrillator; CPR = cardiopulmonary resuscitation; EMS = emergency medical services; IQR = interquartile range; AED = automated external defibrillator; CPR = cardiopulmonary resuscitation; EMS = emergency medical services (fire brigade or mobile medical team); FB = fire brigade; IQR = interquartile range; PEA = pulseless electrical activity; VF/VT = ventricular fibrillation/ventricular tachycardia.; PEA = pulseless electrical activity; VF/VT = ventricular fibrillation/ventricular tachycardia. Comparison of patient outcomes ROSC was achieved in 22.5% (303/1346) of older patients compared with 18.6% (791/4248) of younger patients. Survival at hospital admission (D0) and at 30-days (D30) were respectively 16.1% (217/1345) and 0.5% (7/1346) in the older adult group and 13.3% (565/4249) and 1.2% (51/4249) in the younger adult group. After adjustment, older age was independently associated with a higher rate of ROSC (OR 1.22, 95% CI 1.03–1.45, p = 0.019), a similar rate of D0-survival (OR 1.14, 95% CI 0.95–1.38, p = 0.161) and a lower rate of D30-survival (OR 0.34, 95% CI 0.15–0.81, p = 0.015). Favourable neurological outcome (CPC 1–2 at D30) did not differ significantly between groups, being observed in 5 of 1344 older patients (0.4%) and 23 of 4236 younger adults (0.5%) (OR 0.58, 95% CI 0.19–1.76, p = 0.336) (Fig. 2).Fig. 2Forest plot of outcomes in older adults vs younger adults with traumatic cardiac arrest.Reference: younger adults; Adjustment variables: sex, mechanism of trauma, site(s) of injury, witness status, bystander CPR, bystander AED use, FB response time, MMT response time, signs of life, initial cardiac rhythm, epinephrine, external haemorrhage control, fluid resuscitation, blood transfusion, intubation, thoracostomy.CI = confidence interval; CPC = cerebral performance category; D0 = day 0; D30 = day 30; ROSC = return to spontaneous circulation. Forest plot of outcomes in older adults vs younger adults with traumatic cardiac arrest. Reference: younger adults; Adjustment variables: sex, mechanism of trauma, site(s) of injury, witness status, bystander CPR, bystander AED use, FB response time, MMT response time, signs of life, initial cardiac rhythm, epinephrine, external haemorrhage control, fluid resuscitation, blood transfusion, intubation, thoracostomy. CI = confidence interval; CPC = cerebral performance category; D0 = day 0; D30 = day 30; ROSC = return to spontaneous circulation. Among older subgroups, the 75–84 year age group showed a higher rate of ROSC (OR 1.41, 95% CI 1.10–1.80, p = 0.006) and D0 survival (OR 1.42, 95% CI 1.09–1.86, p = 0.010) compared with younger adults. Multivariable models for CPC 1–2 at D30 in the subgroup analysis failed to converge due to small event numbers. Discussion This study, based on a large French national registry, examined the impact of age on prehospital management and outcomes of out-of-hospital TCA. We found that older patients received fewer prehospital invasive interventions and displayed a dissociation between short-term and longer-term outcomes: older age was significantly associated with a higher probability of ROSC but a lower 30-day survival. As detailed below, several clinical and pathophysiological mechanisms, together with differences in care, can plausibly account for this dissociation. Our findings are consistent with the two existing studies specifically examining age in TCA. Norii et al. reported higher survival to discharge in older adult blunt TCA patients from Japan (5.3% vs. 3.4%; OR 1.58, 95% CI 1.27–1.96), with the older survivor group having notably lower Injury Severity Scores (median 18 vs. 34 in non-survivors).12 Our study extends this observation by demonstrating that this apparent advantage at the time of ROSC does not translate into 30-day survival, a finding that could not be assessed in the Japanese study where only survival to discharge was available. Shekhar et al., using the US NEMSIS database, similarly found higher ROSC (11.3% vs. 7.5%) in older adults.13 Our study provides the first evidence from a physician-staffed EMS system, where on-scene diagnostic capability is presumably higher than in paramedic-based systems, yet the same dissociation persists. Several factors may contribute to this apparent paradox. First, the clinical profile of older TCA patients differed substantially from that of younger adults. Older patients more frequently sustained cardiac arrest after lower-energy mechanisms, predominantly falls, and less frequently had thoracic and abdominal injuries, whereas in younger patients out-of-hospital TCA more often followed high-energy or penetrating mechanisms leading to rapidly fatal thoracoabdominal haemorrhage or severe multisystem injury. Importantly, the cause of arrest, rather than the mechanism alone, is a major determinant of the timing of death: exsanguination tends to be rapidly fatal in the prehospital phase, whereas isolated traumatic brain injury more often leads to delayed in-hospital death.18 A lower contribution of exsanguination to the cause of arrest in older patients, who less frequently sustain high-energy or penetrating trauma, may therefore favour a higher probability of transient ROSC while shifting mortality to a later, in-hospital phase, a sequence compatible with the lower 30-day survival we observed. In addition, older adults presented more frequently with signs of life and a non-significant trend toward more shockable rhythms, which may preserve some short-term reversibility and further increase the probability of transient ROSC. Finally, manual detection of a pulse or of a low-output state is unreliable in the absence of ultrasound or invasive arterial monitoring19; some older patients categorised as being in cardiac arrest may in fact have had a profound low-flow state rather than true mechanical arrest, which could facilitate the recovery or maintenance of a spontaneous circulation while leaving cerebral perfusion critically impaired and thereby contribute to poor longer-term outcomes. However, ROSC in this context does not necessarily translate into meaningful survival, as older patients are more vulnerable to post-resuscitation shock, hypoxic-ischemic brain injury, coagulopathy, reduced physiological reserve, and treatment limitation decisions.20 Thus, age may operate less as a determinant of immediate resuscitation success than as a determinant of post-ROSC trajectory and longer-term survival. This interpretation is consistent with the observed dissociation between ROSC and 30-day survival and suggests that, in older TCA patients, ROSC should not be interpreted as a sufficient marker of prognosis without considering injury pattern, pre-arrest physiology, frailty, and early post-resuscitation evolution. However, the hypothesis raised by Norii et al. in their analysis of the Japanese Trauma Data Bank, suggesting that the observed higher survival in older adult TCA may reflect the inclusion of medical cardiac arrests misclassified as traumatic cannot be excluded.12 This misclassification may operate in both directions: a medical event, such as syncope, an arrhythmia, hypoglycaemia, or sepsis, may precipitate a fall and a subsequent arrest that is then labelled traumatic, while conversely the extent of injuries is frequently under-recognised in older patients, in whom major trauma is notoriously underestimated. As falls are often multifactorial, the initial response to resuscitation may in part reflect treatment of the precipitating medical condition rather than of the injury itself. We cannot exclude a similar misclassification in our database, and the absence of comorbidity data prevents its direct assessment. Differences in prehospital management were also observed: older patients received fewer invasive interventions, including less fluid resuscitation, fewer blood transfusions, lower intubation rates, and substantially fewer thoracostomies. Epinephrine, by contrast, was administered to the large majority of patients in both age groups, in keeping with its routine use during advanced life support irrespective of the cause of arrest, so that the age-related differences concerned trauma-specific procedures rather than standard resuscitation drugs. These differences may reflect a genuine reduction in the intensity of care, including explicit or implicit limitation of resuscitative effort in patients perceived as frail or unlikely to survive, which could itself contribute to the lower longer-term survival observed. Consistent with this, Spering et al. reported in a large trauma cohort that severely injured older patients are managed with a more “wait and see” approach, resulting in higher mortality rates.21 Alternatively, and not mutually exclusively, the lower-energy injury pattern and the lower prevalence of penetrating or thoracoabdominal trauma in older patients may require fewer time-critical invasive procedures, so that part of the observed difference may reflect appropriate tailoring of care rather than under-treatment. From a clinical perspective, this study provides evidence that older age alone should not discourage the initiation of resuscitative efforts in TCA. Although 30-day survival was very low in older patients, it was not futile, and favourable neurological outcomes were observed among survivors as reported elsewhere.21 However, a favourable 30-day outcome was achieved in only 7 of more than 1300 older patients, so such outcomes remain exceptional; the potential for non-beneficial or overly intensive treatment must therefore be weighed against the small probability of meaningful recovery, and a frailty-attuned approach together with time-limited trials of treatment may help to balance these considerations. Second, clinicians should maintain a high index of suspicion for a precipitating or concurrent medical cause of cardiac arrest when managing older trauma patients, particularly when the mechanism is a fall, apparent injuries are minor, or the initial rhythm is shockable, and should adopt a comprehensive, frailty-attuned assessment rather than settling on a single diagnosis. In such situations, integrating standard ACLS measures (including defibrillation when indicated) alongside the trauma-specific resuscitation algorithm may be appropriate. Third, the finding that fewer invasive prehospital interventions were performed in older patients warrants further investigation to determine whether it represents appropriate tailoring of care or potentially modifiable under-treatment. Several avenues for future investigation emerge from these findings. First, a prospective study incorporating systematic recording of initial cardiac rhythm, point-of-care echocardiography, and post-mortem computed tomography would help quantify the proportion of medical cardiac arrests misclassified as TCA in older patients. Second, linkage of prehospital TCA registry data with hospital databases (including anatomic trauma severity scoring systems, comorbidities, and in-hospital management) would enable more robust adjustment for potential confounders and assessment of the in-hospital phase of care. Third, a multicenter study comparing TCA outcomes in older patients across physician-staffed and paramedic-based EMS systems would clarify whether the prehospital care model modifies the age-outcome relationship. Fourth, the development and validation of a clinical decision rule to guide the choice between trauma-specific and medical cardiac arrest algorithms in older TCA patients could have direct clinical utility. Fifth, qualitative research exploring clinician decision-making regarding the intensity of resuscitative interventions in older TCA patients could identify potential sources of age-based treatment variation. Limitations Several limitations must be acknowledged. First, as a registry-based study, the analysis is constrained by the variables collected. Notably, we lacked data on injury severity scoring (e.g. ISS, AIS), comorbidities, anticoagulant or antiplatelet use, and prehospital ultrasound findings. These unmeasured confounders may influence the observed associations. Second, patients in whom no resuscitation was attempted by the mobile medical team were excluded; if resuscitation was less frequently attempted in apparently frail older patients judged unlikely to survive, this exclusion could introduce a selection bias whose direction is difficult to anticipate, and a formal analysis of the missingness and of the characteristics of these excluded cases would help to quantify its potential impact.22 Third, we did not have data on in-hospital management, which may independently affect outcomes. Notably, decisions regarding trauma team activation, damage-control interventions, intensive care admission, neurosurgical procedures, transfusion strategies, and limitation or withdrawal of life-sustaining therapies may differ according to age, frailty, perceived prognosis, comorbidities, and patient preferences.23, 24, 25, 26 Because these variables were not captured in the registry, we could not determine whether the poorer longer-term outcomes observed among older patients reflected irreversible injury severity, reduced physiological reserve, differential in-hospital treatment intensity, or earlier transition to comfort-oriented care. Finally, the adjusted analysis at 30-day should be interpreted with caution because of the limited number of survival events, particularly among older patients (7 survivors and 5 with CPC 1–2) which may have affected model stability and increased the risk of overfitting. However, covariates were selected a priori based on clinical relevance and previous literature to limit data-driven model selection. Conclusion In this large French national registry study, older patients with out-of-hospital traumatic cardiac arrest achieved return of spontaneous circulation more often than younger adults but were significantly less likely to survive to 30 days. This dissociation is compatible with several mechanisms, including a different injury profile, greater vulnerability to post-resuscitation injury, differences in the intensity of prehospital and in-hospital care, and the possible inclusion of medical cardiac arrests misclassified as traumatic. Age alone should not preclude resuscitative efforts, but clinicians should systematically consider a concurrent or precipitating medical cause of arrest, particularly in older patients with low-energy mechanisms, minor apparent injuries, or a shockable rhythm. Declaration of generative AI and AI-assisted technologies in the manuscript preparation process During the preparation of this work the authors used Claude (Anthropic) in order to assist with language editing. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the published article. CRediT authorship contribution statement Axel Benhamed: Writing – original draft, Validation, Methodology, Investigation, Formal analysis, Conceptualization, Supervision. Baptiste Morcel: Writing – review & editing, Formal analysis, Data curation, Conceptualization. Jean-Stephane David: Writing – review & editing, Methodology, Resources. Ewoud Ter Avest: Writing – review & editing, Methodology. Tobias Gauss: Writing – review & editing, Supervision. James van Oppen: Writing – review & editing, Methodology. Karim Tazarourte: Writing – review & editing, Investigation, Resources. Eric Mercier: Writing – review & editing, Validation. Funding The RéAC registry is supported by the Société Française de Médecine d’Urgence (SFMU), the Fédération Française de Cardiologie (FFC), the Mutuelle Générale de l’Éducation Nationale (MGEN), the University of Lille, and the Institute of Health Engineering of Lille. The authors declare that the sources of funding had no role in the conduct, analysis, interpretation, or writing of this manuscript. Data availability Data were obtained from the RéAC registry and are available upon reasonable request and approval from the RéAC steering committee. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.