This was a single-center, observational, retrospective study. It took place in Cova da Beira University Hospital Center in Covilhã, which serves an area with a population of 86000.

The first part of our study consisted of analysis of patients’ clinical records, focusing on survival time, independence, and admission to long-term care facilities. For this, we included all adults admitted to the intensive care unit (ICU) who suffered CA and had ROSC between January 2015 and December 2019. Ninety-seven patients met all criteria, and nine responded to the questionnaire (Figure 1).

Figure 1.

Flowchart of the patient selection process. CPR: cardiopulmonary resuscitation; ICU: intensive care unit; ROSC: return of spontaneous circulation.

(0,3MB).

This study was approved by the hospital's ethics committee and all patients who participated in the questionnaire gave their informed consent.

Clinical information was collected mainly from the SClínico and Centricity Critical Care Clinisoft software platforms.

Due to restrictions imposed by the COVID-19 pandemic, six of the patients responded to the questionnaire via phone call, and three responded in person. The survey included the EuroQoL group's EQ-5D-3L questionnaire, a question to compare quality of life before and after CA, and the Eastern Cooperative Oncology Group (ECOG) performance status scale. The EQ-5D-3L is a generic tool with five questions, each concerning one of five domains (mobility, self-care, usual activities, pain/discomfort and anxiety/depression), and a visual analog scale (EQ-VAS). Based on the value set derived for Portugal and the responses, an index that measures HR-QoL was obtained.14–16

For performance status, we used the ECOG scale. Unlike CPC, it conveys not just neurologic performance, but also the ability to perform activities of daily living (ADL).4,9,17

IBM SPSS version 27 was used for the statistical analysis. All statistical tests were interpreted at a 0.05 significance level, although we also considered p-values <0.10 as a tendency for correlation. First, we used descriptive statistics and applied normality tests. According to the normality of distribution, we used the Mann–Whitney U test, the Student's t test, Spearman's test and the one-sample t test. Survival function estimates were calculated using the Kaplan–Meier method and compared by the log-rank test. For loss of independence and institutionalization, we performed the McNemar test and two logistic regression models. The chi-square test or Fisher's exact test was used for categorical data, as appropriate. Cramer's V served as a measure of the relative strength of association with the following criteria for interpretation: <0.1 very weak, 0.1–<0.3 weak; 0.3–0.5 moderate; and ≥0.5 strong.18

Survival to discharge in our study (32.0%) was similar to other studies: 37% in Portuguese studies,19,20 39% in Chen et al.21 and 26.4% in EuReCa TWO.3

Similarly to other studies, age had a significant impact on survival to hospital discharge.8,22 Some point out the possible role of comorbidity in this association; however, Terman et al.23 and Hirlekar et al.8 propose age as an independent factor for these outcomes. Non-shockable rhythm and non-cardiac causes were also associated with a lower survival rate, which is supported by other studies.1,8,19,20 Severity scores have demonstrated a link to survival in other studies as well, but they have also shown some limitations in forecasting patient outcomes.24–26 By contrast, the relation between hypoxic–ischemic brain injury on CE-CT and survival highlighted in our study is supported by resuscitation guidelines, which recommend neuroimaging as a prognostication tool.4 Regarding our finding of a higher survival rate for OHCA when this occurs in other healthcare facilities, a plausible explanation is the greater probability of immediate assistance.

Studies on this topic tend to disagree, and this heterogeneity is highlighted in a 2018 meta-analysis on IHCA,27 in which survival to discharge and overall one-year survival indicate a one-year survival of 76% for those discharged alive. Some studies on OHCA report one-year survival of up to 90%.10,28 However, their samples were younger (mean age 61–69 years) and had a higher proportion of cardiac causes and shockable rhythms, which are factors associated with better outcomes.

Our results appear to contradict the high rates of independence (72.7–91%) and low rates of institutionalization (1–10%) observed in other studies.1,2,4,7 However, these studies refer to six months or more after CA, contrasting with our study, which also included patients with a shorter survival.1,2,4 If these studies are compared to our patients surviving one year or more, the difference begins to fade.

In our sample, diabetes was a major factor in loss of independence. Diabetes has previously been associated with worse functional recovery, worse neurologic outcomes, and lower survival rates.10,11 These links can be attributed to various mechanisms, including the association of diabetes with a larger infarct area, metabolic disturbances, and a higher risk of previous atherosclerosis and thus worse brain perfusion during CA.11 Female gender was also demonstrated to be associated with higher rates of admission to long-term care facilities in our analysis, which is also supported by a previous study.28 The correlation of length of hospital stay with these outcomes has not been explored in other studies, presumably due to the complexity of such analysis.

First and foremost, it should be pointed out that most studies indicate that CA survivors’ quality of life is good or acceptable and similar to the general population, but survivors do report a deterioration of quality of life after CA.2,4,6,12,13,19 As can be seen in the Results section, our study is in agreement with this statement. The quality of life index in our sample differs significantly from that of the sample used for the Portuguese population in Ferreira et al.15 only if we ignore the 20-year gap between the two. When younger age groups are excluded from Ferreira et al.’s sample,15 the difference in the mean EQ-5D scores is not statistically significant. Moreover, when our sample is compared to the overall population of Central Portugal, there is only a tendency for statistical significance, even though it also includes younger people. Overall, it is fair to believe that quality of life may not be significantly different between long-term CA survivors and the age-matched Portuguese population.

Secondly, many studies agree on the high prevalence of problems related to anxiety and depression.1,2,4,6,7 However, there is considerable variability in these studies’ samples and time since CA, making comparisons difficult. An Australian study29 that applied the EQ-5D-3L to 697 patients 12 months after CA had similar results to our study. Examples of very different results are the studies by Granja et al.19 and Bohm et al.13 Specifically in our sample, the most problematic domains overlapped with those of the general population, according to Ferreira et al.15

Unfortunately, our sample was too small to analyze which variables influenced quality of life and performance status.

In previous studies, the cut-off for a good functional capacity/performance status has tended to include those who are capable of some work activity and are independent in ADL.1,9 Taking this into account, we can say that 78% of our respondents had a good performance status, since they maintained their ability to do at least sedentary work (ECOG ≥1). Furthermore, all respondents were capable of self-care and were out of bed for the majority of the day. All in all, similarly to other studies, performance status was satisfactory.2,9

Nevertheless, the patients we were unable to reach should not be forgotten. Consequently, the proportion of good performance status among those alive at the time of this study may go down to 64%.

Due to its retrospective nature, our study was unable to provide detailed information about the post-CA period. Another limitation was the sample size, especially for the second section, which impeded analysis of the correlation between quality of life/performance status and other variables (e.g. site of CA and initial rhythm). Furthermore, there was considerable heterogeneity in medical records. For instance, it was not possible to include time until initiation of CPR or time until ROSC in the analysis due to the inconsistency in the way it was reported (some reports were in minutes, while others only had the number of CPR cycles or the number of shocks). Taking all this into account, it would be beneficial to have a prospective study that extended to other medical centers.

In clinical practice, we need to know our patients’ needs in order to guide rehabilitation plans, so it is paramount to continue investigation and to stress the importance of close follow-up of these patients. The International Liaison Committee on Resuscitation recommends recording of survival and modified Rankin Scale at discharge and at 30 days, as well as quality of life at six and 12 months. Data on other specific problems should also be collected, such as fatigue, anxiety, and participation in society.5 Another useful tool that can be used for functional recovery is the Extended Glasgow Outcome Scale.