Abstract

Rationale: The elderly population is growing year after year; especially individuals aged 90 and over. Centenarians account for a small proportion of out-of-hospital cardiac arrest (OHCA) victims. The CPR utility in this population is subject to debate and raises an ethical standpoint. Aims and Objectives: Our study aim is to determine characteristics associated with receipt of advanced life support (ALS) and with a return of spontaneous circulation (ROSC) among a cohort of centenarians with OHCA documented in the French National OHCA Registry (RéAC). Method: We included centenarians with OHCA recorded in the RéAC between 2011 and 2021. Clinical and demographic characteristics, OHCA context, care received, and survival were analysed. Primary outcome was the survival rate 30 days after the OHCA or upon discharge from hospital. Results: Of the 132 centenarians included, 17 had received ALS. ROSC was observed in 5 of the 17 cases. Basic life support (BLS) had been initiated by bystanders in 35.6% of cases and by firefighters in 69.2%. On arrival of the emergency medical team, asystole was noted in 91.3% of cases. Individuals having received ALS were more likely to have suffered from asphyxiation, electrocution, drowning, or an overdose (35.3%, vs. 10.5% of the non-ALS cases). Within the ALS group, ROSC was associated with the provision of BLS by bystanders (60.0%, vs. 50.0% of the cases without ROSC) or by firefighters (100.0% vs. 75.0%, respectively) and a shorter no-flow time (6 min vs. 11 min., respectively). None of the 132 included centenarians were alive 30 days after the OHCA. Conclusion: After OHCA in a centenarian, age appears to be a barrier to the ALS initiation but not the BLS initiation. The fact that none of the victims were alive 30 days after the OHCA raises ethical questions about the resuscitation utility for very elderly victims.

Keywords: Out-of-hospital cardiac arrest, Registry, Centenarian, Ethics, Resuscitation

Introduction

Out-of-hospital cardiac arrest (OHCA) is a considerable public health issue in Europe and in the United States. We recently reported that the incidence of OHCA in France is 46,000 per year [1]. Even though improvements in post-OHCA care have been observed over the last decade, the victim’s prognosis remains poor; in France the day 30 survival rate ranged from 4.9% to 10.4% [1,2]. In routine clinical practice, age appears to influence the treatment of cardiac arrest; however, the decision to attempt cardiopulmonary resuscitation (CPR) or not should not be based on this variable alone [3]. Although some researchers have suggested that CPR is futile in very elderly people, we demonstrated previously that the lower survival rate observed among older OHCA victims was due (at least in part) to a shorter duration of CPR and less intensive care (relative to that provided to younger victims [4].

The elderly population is growing year after year; this is especially true for the very oldest age groups (e.g. individuals aged 90 and over). Attitudes concerning resuscitation of elderly OHCA victims typically depend on the perceived likelihood of a successful outcome. Centenarians account for a small proportion of OHCA victims - less than 1% in France[5]. Although post-OHCA survival rates in older adults vary widely in the literature, the utility of CPR in this population is subject to debate [6-11]. The resuscitation of elderly OHCA victims (and especially centenarians) must be considered from an ethical standpoint. The primary objective of the present study was to describe a cohort of centenarian OHCA victims documented in the French National OHCA Registry (RéAC). The study’s secondary objectives were to determine the characteristics of individuals who received advanced life support (ALS) and those with a return of spontaneous circulation (ROSC).

Methods

Study setting

The study data were extracted from the RéAC registry, which gathers data on OHCAs dealt with by French mobile intensive care units (MICUs) [12]. In France, the prehospital emergency system has two tiers. The first professional responders are usually firefighters, who arrive rapidly on the scene and provide basic life support (BLS). The MICU (comprising at least an ambulance driver, a nurse, and a senior emergency physician) then arrives and can provide advanced life support (ALS)[13]. In the hospitals that are part of the RéAC network, a member of the MICU fills out a specific OHCA intervention form; the latter follows the Utstein resuscitation template[14] and complies with the French emergency medical services’ requirements.

Participants and data

Data were collected between July 1st, 2011, and September 1st, 2021. We included all OHCA victims aged 100 or over and then extracted data on the OHCA, the provision of BLS by bystanders and/or by firefighters, the provision of ALS, and the survival rate 30 days after the OHCA or upon discharge from hospital (the primary endpoint). The other endpoints were the frequency of ROSC, the survival rate on admission to hospital on day 0, and the frequency of a good neurologic outcome in day 30 survivors. A good neurological outcome was defined as Cerebral Performance Category of 1 or 2[15].

Statistical analysis

We firstly described the population of people aged 100 years and over. Then, we compared patients who received ALS with those who did not (ALS+ vs ALS-). Finally, we compared patients who sustained a ROSC with patients who did not (ROSC+ vs. ROSC-).

For quantitative variables, the normality of the data distribution was assessed using the Kolmogorov-Smirnov test. Non-normally distributed quantitative variables were described as the median (1st quartile Q1; 3rd quartile Q3). Qualitative variables were described as the frequency (percentage). Depending on the variable’s distribution, our bivariate analyses involved Pearson’s khi-2 test, Fisher’s exact test, or a nonparametric Mann-Whitney U test. All tests were two-sided, and the threshold for statistical significance was set to p<0.05. All analyses were performed using SPSS software (version 25.0, IBM, Armonk, NY, USA).

Ethics

The study was approved by the French Advisory Committee on Information Processing in Health Research (Comité consultatif pour le traitement de l’information en matière de recherche dans le domaine de la santé, Paris, France) and the French National Data Protection Commission (Commission nationale de l’informatique et des libertés (Paris, France): authorization number: 910946). In line with the French legislation on retrospective studies of registry data, patient consent was neither required nor sought.

Results The study population

During the study period, the RéAC registry included 130,678 OHCA victims, of which 132 were aged 100 or over and were included in the present study. Seventeen of the 132 OHCA victims received ALS, which led to ROSC in five cases (Figure 1). The median age of the overall study population was 101 years, and 71.2% of the OHCA victims were men. OHCA mainly occurred at home (in 69.2% of cases) and was due to an underlying medical condition or disease (85.6%). A bystander was present at the time of the OHCA in 74.2% of cases and provided BLS in 35.6% of cases. A first professional responder (i.e. a firefighter) provided BLS in 68.2% of cases, and the MICU provided ALS in 12.9% of cases. The first cardiac rhythm recorded by the MICU was asystole in 91.3% of cases. The time interval between collapse and the arrival of the first professional responder was 9 minutes, and the time interval between collapse and the MICU’s arrival was 20 minutes. ROSC was observed in 6 cases (4.5%) and five patients were admitted alive to the hospital; hence, the day 0 survival rate was 3.8%. The day 30 survival rate was zero (i.e. no survivors).

Figure 1: Study flowchart.

Characteristics of the patients having received ALS

Patients having received ALS were more likely to have a history of cardiovascular disease (82.4%, vs. 45.2% in cases not given ALS, p=0.008). The OHCAs in the ALS group were more likely to have been caused by overdose, asphyxia, drowning or electrocution (35.3%, vs. 10.5% in cases not given ALS; p=0.013) and less likely to have been caused by a medical cause or by trauma (64.7% vs. 88.7%, respectively; p = 0.018). Patients having received ALS were less likely to be at the end of life (0.0%, vs. 11.3% in the non-ALS group) but the intergroup difference was not significant. “Do not resuscitate” orders were noted for one patient in the ALS group and 5 in the non-ALS group. The presence of a bystander was more frequent in the ALS group (88.2%, vs. 72.2% in the non-ALS group); this bystander was more likely to be a family member (52.9% vs. 40.9%, respectively) and was more likely to initiate BLS (52.9% vs. 33.0%, respectively). ALS always included intubation and usually included an epinephrine injection (in 70.6% of cases). The provision of ALS was associated with a shorter arrival time for the firefighters (8 vs. 10 min in the non-ALS group) and for the MICU (13 vs. 20 min, respectively), and a shorter no-flow time (7 vs. 12 min, respectively). The two groups did not differ significant with regard to the person’s location at the time of the OHCA; this was notably true for being in a nursing home (23.5% in the ALS group and 22.3% in the non-ALS group). The provision of ALS was associated with a significantly greater frequency of ROSC (29.4%, vs. 0.9% in the non-ALS group; p<0.001) and the D0 survival rate (29.4% vs. 0.0%, respectively; p<0.001). The day 30 survival rate was zero in both the ALS and non-ALS groups.

Table 1: Description of the study population as a whole and the ALS and non-ALS subgroups.

Characteristics of patients with ROSC

ROSC was noted in five of the 17 patients in the ALS group (Table 2). The small sample size prevented us from performing a statistical analysis. In the ROSC group, the OHCA was more likely to have occurred at home (80.0%, vs. 58.3% in the non-ROSC group) and less likely to have occurred in a nursing home (0.0% vs. 33.3%, respectively). The presence of a family member bystander was more likely in the ROSC group (80.0%) than in the non-ROSC group (41.7%). Patients with ROSC were more likely to suffer from an OHCA another cause of OHCA (other than medical or traumatic etiology, ie. overdose, asphyxia, drowning or electrocution) and more likely to receive bystander BLS (60.0%, vs. 40.0% in the non-ROSC group) or firefighter BLS (100.0% vs. 83.3%, respectively). In the ROSC group, the first cardiac rhythm recorded by MICU was less likely to be asystole (80.0%, vs. 90.9% in the non-ROSC group), and patients were more likely yo receive epinephrine was more often injected (80.0% vs. 66.7%, respectively). ROSC was associated with a shorter no-flow time (6 min, vs. 11 min in the non-ROSC group) and a longer low-flow time (22 vs. 13 min, respectively). All patients with a ROSC were admitted alive to hospital but none survived to D30.

Table 2: Characteristics of OHCA victims with or without ROSC.