Aspects of post-resuscitation care after out-of-hospital cardiac arrest

Abstract

Background: Cardiac disease is the most common cause of death in the western world and the majority of these deaths are due to out-of-hospital cardiac arrest (OHCA). In Sweden, approximately 10,000 persons suffer an OHCA annually and in 5000 cardiopulmonary resuscitation (CPR) is initiated. The successful return of spontaneous circulation (ROSC) and admission to hospital are just the first steps towards the goal of complete recovery from cardiac arrest. Aims: The aims of Papers I, II and IV were to evaluate different aspects of post-resuscitation care and their importance for survival after OHCA. These aspects included the use of implantable cardioverter defibrillators (ICD) and mild induced hypothermia (MIH). The aim of Paper III was to use variables and information available at intensive care unit admission to develop a risk score for poor outcome useful for comparing populations and defining patient risk when assessing effects and creating power calculations in interventional studies. Methods: Papers I, II and IV were retrospective observational studies of OHCA patients admitted to hospital in Gothenburg during different periods of time from 1980-2015 (n=1,609, n=390 and n=871). Paper III is a post-hoc analysis of the randomized multicenter Target Temperature Management trial (n=933). Results: In Paper I, we did not find any significant change in one-year survival between the two time periods (1980-2002 and 2003-2006) when all the patients were studied (27% vs. 32%; P = 0.14). Among patients found in ventricular fibrillation, an increase in one-year survival was found (37% vs. 57%; P=0.0001). The proportion of survivors to hospital discharge with low cerebral function (cerebral performance category score 3) decreased from 28% to 6% (P = 0.0006) among all patients. The use of ICDs increased (Paper II), but, in overall terms, only 58 of 390 survivors (15%) received an ICD. Among patients who received an ICD, the two-year mortality was 2%, versus 25% among those who did not (p < 0.0001). The long-term follow-up showed that the use of an ICD had a borderline association with lower risk of death (adjusted hazard ratio 0.49; 95% confidence interval (CI), 024-1.01; p = 0.052). In Paper III, we identified ten independent predictors of a poor outcome among patients who had ROSC on admission to hospital and created a risk score based on the impact of each of these variables. This score yielded a median area under the curve of 0.842 (range; 0.840-0.845) and good calibration. In Paper IV, we used a stratified propensity score analysis to adjust for factors potentially influencing choice of treatment with MIH. The odds ratio (OR) for 30-day survival was not significantly higher in patients treated with MIH compared with non-MIH-treated patients; OR 1.33 (95% CI 0.83-2.15; p=0.24). A good neurological outcome at hospital discharge was seen in 82% of patients who were discharged alive from hospital. Conclusions: We did not find any overall improvement in survival over time among patients who had ROSC on admission to hospital after OHCA, but we found signs of improved cerebral function among survivors to hospital discharge, following the introduction of more intensified post-resuscitation care. The use of ICDs was low but increased over time. Among survivors of OHCA caused by ventricular fibrillation or tachycardia who received an ICD during hospitalization, only 2% died during the following two years. Patients running a high risk of a poor outcome after OHCA could be identified at an early stage by using a simple, easy-to-use risk score, based on ten independent predictors of a poor outcome at six months. Treatment with mild induced hypothermia was not significantly associated with an increased chance of 30-day survival among patients who were still unconscious on admission to hospital after OHCA.