Can I get a witness?

In the United States alone there are 450,000 cases of cardiopulmonary arrest annually. Eighty percent of these events occur at home with about 90% of the population dying and >50% of those that survive left with permanent brain damage [1]. In-hospital arrests have slightly better survival rates [2]. The desire to have family present during cardiopulmonary resuscitation (CPR) originates from 1987 when a family member insisted on being present during CPR [3]. There have since been several surveys designed to determine healthcare professionals’ preferences. Members of the healthcare team expressed reservations listing anxiety, family serving as barriers to provision of effective care and concerns that the family will suffer psychological ill effects. There has been increasing interest in the effect of family presence during CPR on both the family member and healthcare professionals with suggestions that it could be beneficial to the families as a grieving tool. In the pediatric world, family presence during CPR is the norm. However, resuscitation in pediatric patients differs from those in adults as survival rates are higher among children and the healthcare team is often prepared with the skills required to manage both the sick child and the anxious parent. Resuscitation guidelines advocate for the presence of families, including the European Resuscitation Council Guidelines for resuscitation and the American Heart Association guidelines for CPR and emergency cardiovascular care [4],[5]. Thus, Jabre and colleagues conducted the first randomized trial of its kind to understand the effects family presence during CPR has on both the family and the healthcare team.

This study was a well-designed prospective cluster randomized controlled trial and took place in France over 2 years. The primary end point was the presence of post-traumatic stress disorder (PTSD) in the relatives that witnessed CPR at 90 days. The secondary end point was the presence of anxiety and depression in the relatives of the decedents, the well-being of the healthcare team and occurrence of medicolegal claims. The intention-to-treat analysis showed that family members in the control group were more likely to exhibit symptoms of PTSD than those in the intervention group with an odds ratio (OR) of 1.7 (confidence interval (CI) 1.2 to 2.5). Family members who did not witness CPR were more likely to exhibit symptoms of PTSD compared with those that did with an OR of 1.6 (CI 1.1 to 2.5). The study also concluded that family presence did not interfere with medical resuscitation efforts, increase stress among members of the medical team, or result in additional medicolegal conflicts.

This study had several strengths. First, the inclusion and exclusion criteria were simple and could be rapidly applied in the field. Second, the sample size was adequately powered to observe the desired effect. Third, psychological effects of witnessing CPR or not was assessed using well-evaluated monitoring tools. The study had one significant weakness and a few minor ones. Importantly, there was significant crossover between the intervention and the control group with about 43% of the control group opting to witness CPR. They are listed as getting standard of care but it is unclear what standard care entails and more importantly how that differs from the intervention, especially as the controls were given communication guides too. This raised concern as to what the intervention really was: was it the presence of family during CPR or the presence of a communication guide for the families or both? Other limitations include that the observed variables are subject to cultural differences that may not allow direct extrapolation to North American populations. For example, does being a predominantly Catholic population in a largely paternalistic healthcare system inform patients/relatives responses to the healthcare team and end of life decisions? Randomization of the emergency medical service teams ab initio also allowed for the potential to introduce selection bias. Regardless, psychological benefits persisted for family members offered the possibility to witness the CPR of a relative in a follow-up study performed 1 year after the event [6].

The overwhelming number of patients that do not survive cardiac arrest or survive with significant neurologic sequelae makes cardiac arrest an end of life event requiring interventions designed to facilitate the grieving process. We recommend that in the appropriate circumstance, family should be offered the opportunity to be present during CPR. Specific protocols should be instituted to manage this event and a `communication guide’ as used in the trial should be with the family until CPR is terminated.