In-hospital cardiac arrest (IHCA) is a highly severe clinical event associated with high mortality and significant neurological disability among survivors. Data from large international registries report survival-to-hospital-discharge rates ranging between 17% and 25%.1

Most IHCA episodes are preceded by identifiable signs of clinical deterioration hours before the event. In response to this evidence, Rapid Response Teams (RRTs) have been implemented with the aim of initiating resuscitation maneuvers early in the event of cardiac arrest.2

In Spain, published data on IHCA are limited and mainly derived from multicenter registries or studies focused on specific settings. The availability of structured analyses based on Utstein criteria in individual hospitals may provide relevant information on prognostic factors in real-world clinical practice. To address this gap, we conducted a retrospective cohort study including 202 patients between January 2018 and June 2024 at the Hospital Clínico Universitario Lozano Blesa in Zaragoza. The aim of the present study is to analyze the clinical characteristics of IHCA in a tertiary hospital and to identify factors associated with survival and neurological outcomes following activation of the rapid response team.

Most patients were men (67.8%), with a mean age of 73.2 ± 11 years, while women had a mean age of 70.8 ± 13 years. The most frequent cardiovascular risk factors were hypertension (70.3%), dyslipidemia (50%), diabetes mellitus (41.1%), obesity (35.1%), and active smoking (18.8%).

The initial rhythm was documented in 170 of the 202 patients (84.3%); in the remaining 32 patients (15.7%) it was not recorded. Considering the total cohort (n = 202), asystole was observed in 42.2% of cases, pulseless electrical activity (PEA) in 21.8%, and ventricular tachycardia/ventricular fibrillation (VT/VF) in 20.3%. When the analysis was restricted to patients with a documented initial rhythm (n = 170), 75.8% presented non-shockable rhythms (asystole and PEA) and 24.2% shockable rhythms (VT/VF). At the arrival of the RRTs, chest compressions had already been initiated in all cases, along with ventilation using a bag-valve mask without intubation. No mechanical compression devices were used.

Duration of cardiopulmonary resuscitation (CPR) was recorded in 154 patients, with a median of 15 min (IQR 6–26); 42.2% lasted <10 min and 57.8% >10 min. In 48 patients (25%), duration was not documented, representing a limitation of the study. The etiology of cardiac arrest was cardiological in 33.8% of cases, respiratory in 26.9%, and undetermined in 39.3%.

Neurological outcomes assessed at hospital discharge using the Cerebral Performance Category (CPC) and Glasgow Outcome Scale (GOS) showed that 20% of patients achieved good recovery (CPC 1–2), 21% had severe disability (CPC 3), 7.4% were in a vegetative state (CPC 4), and 51% died (CPC 5). Findings according to GOS were consistent: good recovery in 14.9%, neurological sequelae in 23.6%, and death/vegetative state in 61.3%. Witnessed cardiac arrest and shockable rhythms were associated with better neurological outcomes (p = 0.024 and p = 0.001, respectively), and CPR duration <10 min was associated with recovery without sequelae in 65.4% of patients (p < 0.001).

Overall, 1-year mortality was 73.8%, with deaths occurring at the time of cardiac arrest (30.7%), within the first 24 h (16.8%), and after 24 h (26.2%). Limitation of therapeutic effort (LTE) was documented in 43 patients (21.2%).

No significant influence of day of the week or weekend occurrence on mortality was observed. However, witnessed cardiac arrest was associated with higher survival (HR 9.14, p = 0.002), and shockable rhythms (HR 18.99, p < 0.001, respectively) (Fig. 1). Multivariate analysis identified CPR duration >10 min (OR = 8.5; 95% CI 3.03–24.05) (Fig. 2) and cardiological etiology of the arrest (OR = 4.5; 95% CI 1.28–15.95) as independent predictors of mortality.

Figure 1.

Survival according to the shockable rhythm.

Figure 2.

Survival and resuscitation time.

Ninety percent of arrests were witnessed, 73.4% by healthcare personnel, figures comparable to the American National CPR Registry, where 73% of cardiac arrests were both witnessed and monitored.3

Regarding the initial rhythm, among patients with a documented rhythm 75.8% presented non-shockable rhythms (asystole and PEA), while 24.2% were shockable rhythms, figures consistent with series such as De la Chica4 and the AHA,5 which report 25–30% shockable rhythms. Rhythm classification and recording of clinical variables were performed according to the Utstein Style recommendations, ensuring comparability with other international studies and methodological consistency.6

Regarding the initial rhythm, 75.8% of cases presented with non-shockable rhythms (42.2% asystole and 21.8% pulseless electrical activity), while 24.2% were shockable, figures consistent with series such as De la Chica4 and the AHA,5 which report 25–30% shockable rhythms.

CPR duration was identified as a key determinant of survival. In the 154 recorded cases, the median duration was 15 min, with 42.2% lasting <10 min; these patients had better outcomes. Patients with PEA and asystole benefited more from prolonged CPR, whereas those with shockable rhythms depended mainly on immediate defibrillation, as previously described.7 These findings are also supported by modern registries using Utstein-style templates, where CPR duration remains a robust predictor of adverse outcomes.8

Neurological prognosis is a critical indicator of survival quality after IHCA. According to the CPC scale, 54.2% of patients were classified as death, 6.2% as vegetative state (CPC 4–5), 19.3% as moderate-to-severe disability (CPC 2–3), and 20.3% as having no neurological sequelae (CPC 1).

Witnessed arrests were associated with better neurological outcomes, regardless of whether they were detected by healthcare personnel, day of the week, time, or location. Recent registry studies emphasize that preservation of neurological function compared with baseline status is a crucial outcome to report, as many survivors maintain or only slightly worsen their CPC after the event, particularly when monitoring and post–cardiac arrest care is optimal.9

The type of arrest rhythm also influenced prognosis, with lower neurological deficit in shockable rhythms and in CPR < 10 min, findings consistent with multiple recent studies.3

An unexpected finding in our cohort was the association of cardiological etiology with worse prognosis (OR = 4.5; 95% CI 1.28–15.95), which contrasts with most published series, where a cardiac origin is typically linked to a higher proportion of shockable rhythms and, consequently, better survival and neurological outcomes.10 Several factors may explain this discrepancy. First, the relatively small sample size and the retrospective single-center design may have introduced selection bias. Second, a substantial proportion of arrests in our series were classified as undetermined etiology (39.3%), which may have diluted the expected prognostic advantage of cardiac causes, and finally because our cardiological group included a high proportion of elderly patients with significant comorbidities (hypertension, diabetes, dyslipidemia), whose baseline status may have negatively influenced outcomes regardless of the initial rhythm.

Regarding overall one-month mortality, it was 67.3%, with most deaths occurring during the cardiac arrest or within the first 24 h. Survival of 32.7% was higher than that reported by De la Chica et al.4 (23.1%), Santa Cruz et al.11 (25%), AHA 2020 (25.8%), and ERC 2021 (15–34%). However, some larger recent series using Utstein reporting show variability in survival and neurological outcomes, underscoring the importance of standardized data collection.12

In conclusion, survival and neurological prognosis were better in patients with witnessed arrests, shockable rhythms, and CPR duration <10 min. Independent predictors of mortality included CPR duration >10 min and cardiological etiology. These findings support that implementation of a standardized protocol with a rapid response team improves intervention times and resuscitation quality, translating into higher survival and better neurological outcomes.