Sex differences in survival after out-of-hospital cardiac arrest: a meta-analysis

Out-of-hospital cardiac arrest (OHCA) refers to the sudden loss of heart function outside of the hospital, which leads to complete cessation of systemic circulation. Cardiogenic factors, such as coronary heart disease and cardiomyopathy, are the main cause of OHCA. At present, OHCA is a leading cause of sudden cardiac death worldwide. It is reported that for every 100,000 people in Europe, 1753 people have experienced OHCA that was subsequently treated and registered by the emergency medical service (EMS) [1]. In the USA, approximately 55 out of every 100,000 people have OHCA [2]. Moreover, a considerable number of patients suffering from OHCA do not receive timely treatment from EMS, so they are not recorded. The mortality rate of OHCA remains high. Globally, the discharge survival rate of OHCA is low: approximately 3.0% in Asia, 6.8% in North America, 7.9% in the UK, 7.6% in Europe, and 9.7% in Australia [3]. It is estimated that 275,000 people receive EMS treatment each year in Europe, and only 29,000 people are discharged alive [4]. Previous studies have shown that the prognosis of OHCA is related to factors such as age, primary disease, bystander cardiopulmonary resuscitation, and initial heart rhythm.

Recently, an increasing number of studies have focused on the differences in the prognosis of various diseases between males and females. Studies have shown that there are significant differences in pathophysiology between men and women, and there may be different clinical symptoms and prognoses [5, 6]. Compounding this, research has identified differences in the medical services received by males and females after admission [7, 8]. The medical services include post-resuscitation care and nursing measures. Cardiovascular disease is the most well-studied topic related to studies on sex differences. Various studies have shown that there is a significant correlation between sex and the pathophysiology and prognosis of cardiovascular disease. For example, for coronary heart disease, females are more likely to have non-obstructive microvascular disease and endothelial dysfunction, while males are mainly characterized by the formation of obstructive macrovascular plaques [9]. Females, especially young females, also fare worse than men after acute myocardial infarction [10].

However, it is not clear whether sex differences in heart-related diseases affect the prognosis of OHCA. Multiple studies have addressed possible sex differences associated with OHCA prognosis, but the findings are inconsistent: some studies report no difference between males and females in survival after OHCA [7, 11]; others have reported better survival for males or better survival for females [12, 13]. A meta-analysis published in 2015 showed that the discharge survival rate of female OHCA patients was 10% higher than that of male, but the results of the latest large clinical studies show conflicting trends [14].

Therefore, to address these issues, we conducted a systematic review and meta-analysis to evaluate the relationship between prognosis and sex after OHCA from a multidimensional perspective. Prognostic factors are divided into three categories: baseline characteristics, pre-hospital care, and hospital care. Our findings provide guidance related to narrowing the difference in prognosis between males and females after OHCA.

Among the 1042 studies obtained in our initial search, 33 qualified articles (Fig. 1) were included in this meta-analysis. All 33 articles were full-text publications, from 34 countries and regions, with a total of 1,268,664 patients. All of the studies were retrospective cohort studies, and most of the data were obtained from local and national official databases. Agreement between the two reviewers was 94% for study selection and 91% for quality evaluation. No secondary analysis or additional publications were found related to the above-mentioned articles.

In our baseline statistical analysis of OHCA patients, we found that females are older, which may be associated with coronary artery disease occurring 10–15 years later in females than in males; it may also be related to the longer life expectancy of females than males [5]. At the same time, females are more likely to have cardiac arrest at home than in public, which may be because females spend more time at home than males. As a result, females are less likely to be witnessed by bystanders during OHCA. Previous studies have shown that even when both males and females experience cardiac arrest with bystander witnesses, females are less likely to receive CPR because of conservative attitudes and privacy concerns [15]. However, our findings were not in agreement; we found no statistically significant difference in the proportion of females and males who received bystander CPR. This might be related to regional and temporal differences of the studies included in this meta-analysis. In two clinical studies in Japan, for example, data from 1998 to 2007 showed that the proportion of females and males receiving bystander CPR after OHCA was 28% and 34% [16], while data from 2013 to 2016 showed that the proportion of bystander CPR increased significantly, and the proportion of females receiving bystander CPR was even higher than males (56.4% vs 49.6%). We speculate that with the development of society, the comprehensive popularization of CPR-related knowledge, and the change of stereotypes, an increasing number of bystanders can perform CPR, which not only increase the proportion of bystander CPR, but also make up for the corresponding sex differences.

Previous meta-analyses have shown that factors, such as age, bystander witnesses, bystander CPR, and shockable rhythm, are all associated with the outcome of OHCA [17]. Our baseline statistics also showed some differences between females and males, but our analysis showed that there was no significant difference in survival between females and males after OHCA until admission. Even though the treatment measures were different from OHCA to admission, they had little effect on the survival at admission. Therefore, we speculate that some pre-hospital baseline factors (age, bystander witness, bystander CPR, and shockable rhythm) do not have an obvious immediate impact on the differences in survival from OHCA to admission between the sexes. Moreover, there may be corresponding long-term effects, such as affecting survival after discharge and nervous system prognosis of patients.

Our meta-analysis showed that after cardiac arrest, the survival of females from admission to discharge was significantly lower than that of males even considering no significant sex difference in survival at admission. The overall discharge survival rate for females remains much lower than that for males, and the neurological prognosis for females is also poor. Our fundamental analysis showed that females generally received less post-admission interventions, such as PCI, CAG, and TTM, than males. Previous studies have shown that women are significantly less likely to receive PCI after OHCA than men, although the prognosis after PCI is not significantly related to sex [7]. Therefore, the survival of females after OHCA is lower than males, which may be related to the differences in medical services received after admission.

In contrast, some studies have suggested that female hormones are the dominant factor in OHCA because estrogen can resist apoptosis and inflammation. Estrogen has a protective effect on heart and nerves, but this has not been widely confirmed in clinical practice [18]. We believe that although females have estrogen as a favorable prognostic factor, there are also corresponding unfavorable factors, such as age, pathophysiology, acceptance of medical services, and willingness to express opinion. First, females are older at onset and more likely to have various complications, such as chronic obstructive pulmonary disease (COPD), tumor, and psychiatric history. Secondly, female pathophysiological differences make them more prone to complications such as bleeding [19] and acute renal failure [20]. Females have a lower acceptance of corresponding medical services, drug treatment, and invasive procedures. Additionally, females generally receive less corresponding nursing measures. For example, during OHCA, females were less likely than males to receive intravenous access and intraosseous access [8]. This may be related to the fact that nurses perceived intravenous access as more difficult in females than in males, although there was no difference in success rate. Finally, females are more willing to express their intentions and participate in the discussion about their death [21]. To a certain extent, these factors will affect the choice of resuscitation measures.

Combined with factors, such as education level, religious belief, and economic level, females are more likely to issue the instruction of "Do not attempt resuscitation" at the early stage of OHCA, thus adopting more conservative treatment [12]. Overall, this leads to a poor prognosis of OHCA in females.

Our meta-analysis also had some limitations. We have only searched three major databases, and there may be studies that are not included. Specifically, it was limited by the type of study; therefore, we could only analyze the correlation of various factors, but could not directly analyze the causal relationship. Moreover, the types of studies included were all observational studies, so it was difficult to control the corresponding confounding factors. For example, the time and location of studies were uncontrollable confounding factors. As mentioned above, the studies included were conducted between 1980 and 2018. However, post-resuscitation care guidelines have changed over the years these studies were conducted, especially in terms of TTM. On the other hand, the included studies were distributed in different emergency medical systems, and these emergency medical systems have different medical levels and different termination of resuscitation rule, which may affect the survival rate of OHCA. Based on the above factors, the heterogeneity of our meta-analysis is relatively high. However, we conducted meta-regression and sensitivity analysis to find the main source of heterogeneity and to ensure the reliability of the results. Finally, post-resuscitation care research is all affected by immortal time bias. Immortal time bias likely results in apparent lower mortality rates and thus overestimates of the potential benefits on survival.

Our results showed that after OHCA, the overall discharge survival rate of females was much lower than males and the prognosis of the nervous system was poor. We speculate that this may be related to different medical services and nursing measures; however, these findings should be interpreted with caution due to the presence of several confounding factors. Many of these adverse conditions for females can be changed with education, so our results also provide new insights and directions for how to narrow the difference in prognosis between males and females after OHCA.