Impact of Different Waves of COVID-19 on Emergency Medical Services and Out-of-hospital Cardiopulmonary Arrest in Madrid, Spain

Background: COVID-19 has led to decreased survival of out-of-hospital cardiorespiratory arrest (OHCA). We analysed the impact of the rst COVID-19 pandemic year on emergency medical services and OHCA care compared with the previous year. Methods: Data for this observational study were collected for OHCAs attended by the SUMMA 112 emergency service during March 2019 to March 2021. We compared data covering 15 March 2020–14 March 2021 (pandemic year) to retrospective data covering 15 March 2019–14 March 2020 (non-pandemic year). Results: During the pandemic period, 1743 OHCA patients were attended, compared to 1781 during the non-pandemic year. Median patient age during the pandemic period was lower than in the non-pandemic period (71 vs 72, p=0·037). Emergency services response activation time increased during the pandemic year, to 3 minutes, 16 seconds from 2 minutes, 48 seconds in the non-pandemic period (p=0·001). Time to arrival at the scene also increased during the pandemic (12 minutes vs 11 minutes, 25 seconds before the pandemic; p=0·001). The percentage of OHCAs in which resuscitation was attempted was lower during the pandemic (59·4% vs 62·9%, p=0·034), as were survival on hospital arrival (30·3% vs 34·6%, p=0·04). Differences in response activation time (p=0·003) and scene arrival times (p=0·003) were greater during the rst pandemic wave compared with the later phases. Conclusions: The different phases of the pandemic variably affected OHCA care. The rst wave led to longer resource activation, increased home events and scene arrival times, as well as lower patient survival.


Background
A mere three months elapsed from the time the rst COVID-19 cases were reported in December 2019, in Wuhan, China, until the World Health Organisation (WHO) declared a pandemic 1 on 11 March 2020. In Spain, 31 cases had been con rmed by 28 February. By the time a state of alert was declared on 14 March, 2940 cases and 86 deaths had been reported in the Autonomous Community of Madrid, bringing the total for Spain to 5753 cases and 136 deaths. 2 Since then, the disease has affected different countries and regions with peaks and troughs in cumulative incidence gures, referred to as pandemic waves.
Out-of-hospital cardiac arrest (OHCA) is one of the three leading causes of death in industrialised countries, affecting more than 370 000 individuals in the United States and Europe. 3,4 Several studies have described the impact of the rst pandemic wave in terms of the number of cases and OHCA survival rates in various regions, [5][6] but few have addressed the relationship between the successive phases and how they affected OHCA. 7,8 Similarly, some publications have focused on how COVID-19 has affected the management of emergency medical service (EMS) calls and resources, [9][10] but we have found none describing variations in these factors during different waves of the pandemic.
In Spain, the region hardest hit by the pandemic is the Autonomous Community of Madrid, with 603 840 cases reported up to 15 March 2021. 11 The 14-day cumulative incidence peaked at more than 990 cases, with these rates remaining above 200 for an eight-month period between 15 March 2020 and 15 March 2021. Importance: these data will help EMS plan ahead for increases in calls, as well as offering insight into how OHCA and its epidemiology are affected by uctuations during the course of a pandemic such as COVID- 19 12 Patients were categorised into one of the above-described periods according to the date of their OHCA.
Quantitative data are described using medians and interquartile ranges (IQRs) and the qualitative data with n and percentage values.
Comparative analyses were performed using Pearson's chi-square test or Fisher's exact test for qualitative variables and the Wilcoxon test for quantitative variables. Differences were considered statistically signi cant when the p-value was less than 0.05. All statistical analyses were performed using R software v 4.1.

Impact of the pandemic on OHCA
During the pandemic year, 1743 patients with OHCA were attended, whereas in the non-pandemic study year, 1781 were attended. Table 1 shows a comparison of values between the two time periods. We found a decreased median (IQR) age from 72 (59-82) years pre- In terms of the proportion of patients who survived to hospital discharge compared to those who were admitted, there were no signi cant differences; however, there was a trend to lower survival in the pandemic year (128 (12.9%) vs. 91 (10.3%); p=0.07). One-month survival rates also did not differ signi cantly between the two periods but trended lower in the pandemic year (118 (12.5%) vs 83 (9.7%); p=0.06).

Effect of pandemic phases on OHCA
We also sought to assess the in uence of different phases of the pandemic (three waves and one plateau after the rst wave) on OHCA care (Table 2). During the plateau phase, when the 14-day cumulative incidence was less than 45, the median age decreased, but the differences were not statistically signi cant (p=0.22). Activation time was longer during the rst wave compared to later phases (p=0.003), and arrival time was longest during the rst and third waves (p=0.003). Sex ratios did not differ (p=0.24) among the different periods. The percentage of home arrests increased signi cantly (p=0.006) during the rst wave, although with no signi cant changes in whether the OHCA was witnessed (p=0.43) or if CPR was performed before EMS arrival (p=0.40). During the rst wave, the proportion of advanced resuscitation attempts was signi cantly lower compared with other phases, at 46.0% vs 65.3% during the plateau, 61.1% during the second phase, and 61.3% during the third wave (p<0.001). The reason for not resuscitating during the rst wave was the excessive time interval from collapse to the arrival of emergency services (50.0%), leading to a signi cant difference with the other phases (p=0.013) (Fig. 1). There was no signi cant difference when comparing the percentage of shockable rhythms (p=0.53) or whether the airway was managed (p=0.79). Signi cant differences were found when comparing the type of airway management, with supraglottic airway devices used more in the rst wave (33.3%) compared to the plateau phase (23.0%) and the second (27.2%) and third waves (15.5%; p<0.001) (Fig. 2).
The OHCA aetiology was mainly medical, with percentages ranging from 89.5% to 94.1% and no signi cant differences (p=0.31). Similarly, no differences were found in analyses of survival on hospital arrival (p=0.95) (Fig. 2). Survival to hospital discharge also did not differ across periods (p=0.88) and neither did rates of one-month post-discharge survival (p=0.74).

Impact of the pandemic on EMS activity
As can be seen in Figure 3, during the rst wave of the pandemic, the number of calls to the emergency coordinating centre rose, increasing to 57 498 calls per week, compared to 22 262 in the same period of the previous year (158% increase). During the plateau phase, the number of calls was the same as in the previous year, whereas calls increased over the pre-pandemic period in the second wave, with a peak of 11 709 calls in one week. The increase intensi ed during the third wave, with the greatest difference at 15 295 calls per week.
Despite the increase in calls to the emergency coordinating centre, the number of devices mobilised for total health emergencies did not increase during any of the pandemic phases compared to the same periods of the previous year.

Discussion
In this study, we gathered data on the OHCA care provided by the Emergency Services in the Community of Madrid, Spain, one of the rst European regions to suffer the effects of COVID-19. These data have enabled a detailed analysis of OHCAs during the successive waves of the pandemic, including a plateau phase after the rst wave in which strict initial mitigation measures allowed transmission to be controlled, and to compare these periods with the same months in the year prior to the pandemic. We found no increase in the number of OHCAs during the pandemic year compared to the non-pandemic year. As far as we know, no other studies have made this comparison between full years, but some groups have compared speci c periods during 2020 with their counterparts in the previous year. Although a few of these investigations showed differences in the number of OHCAs, 15 most reported an increase during the pandemic period. 5,6,16,17,18 Throughout the pandemic, response times have increased, as described above, 5,19,6,20,16,17,21,22 because both resource activation and arrival times increased, as Lim et al. also reported. 23 We did, however, observe response times at pre-pandemic levels during the pandemic plateau phase (Fig. 1), after which they increased again slightly.
In most studies, no signi cant differences were reported in terms of sex ratios, similar to our ndings, but some studies have reported signi cantly higher patient ages during the pandemic period. 16,18,24 This result contrasts with our ndings and those of Rosell et al., 25 in both cases showing lower ages in this patient population, although usually not signi cantly so. The COVID-19 pandemic has clearly in uenced all aspects of life, both because of its direct health effects and because of home lockdown-related measures implemented to varying degrees. As a result, during the pandemic year, the percentage of OHCAs taking place in homes increased, 3,17,18,14,15,19,21,22,23 especially during the rst wave, although this percentage later decreased, possibly because of easing lockdowns (Fig. 3).
We observed that resuscitation attempt rates decreased signi cantly during the rst pandemic wave, as reported in other studies, 5,20,16,17,22 but we also found that they recovered to non-pandemic levels during the remainder of the pandemic year (Fig. 1), which has not been previously assessed. The most common reason for non-resuscitation, especially in the rst and third waves, was the excessive time between the OHCA and the arrival of the advanced life support. These longer periods between call placement and EMS arrival in the rst and third waves could be the result of both a higher number of calls and greater awareness of the need to use personal protective equipment, leading to longer call-response times.
We found no signi cant differences in whether or not the airway was managed, but as other studies have indicated, 6,24,25 supraglottic airway management devices were more commonly used during the pandemic period, a pattern that changed over time (Fig. 2). Even with the observed changes, the use of orotracheal tubes for airway management was still more common, despite recommendations otherwise.
As other researchers have reported 5,16,17 during the pandemic year, the aetiology of OHCA was most frequently medical. Only during the plateau phase did this percentage drop, falling even to non-pandemic levels (Fig. 2).
As described in other studies, 5,19,6,20,16,17,21,23,24,25,26 we found that OHCA survival at hospital admission was signi cantly reduced. One prediction might have been that as the rst wave ended and COVID-19 incidence fell, the survival rate would have improved; however, it remained low, with no signi cant changes during the successive waves (Fig. 2). This pattern could be explained by the high probability of thromboembolism 27 among patients who had suffered COVID-19 and its sequelae.
When comparing survival at the time of hospital discharge and comparing one-month survival rates, we found no signi cant differences for either between the pre-pandemic and pandemic years, but there was a trend towards worse survival during the pandemic. We also found no signi cant differences in these outcomes among the different periods of the pandemic. Limitations: This study has the limitations inherent to a prospective study starting on 15 March 2020 and involving comparison with retrospective data acquired from the SUMMA 112 OHCA registry. Although the variables were considered in the same way in all cases, some patient data may be missing, such as survival to hospital discharge and one-month survival. In addition, not knowing the underlying reason for the increase in resource mobilisation despite the increased number of calls creates a degree of uncertainty that we cannot address. Possible explanations include additional call handlers or new management guidelines. Another limitation is that according to some other studies, 31 possibly only one of every ten COVID-19 positive cases was detected during the rst wave, so that the actual incidence during this period could have been much higher than what o cial records indicate.

Conclusion
In summary, the different phases of the pandemic variably affected OHCA care. Survival from OHCA has decreased since the onset of the pandemic and has not recovered since then. The number of calls to EMS increases as AI increases, although the number of mobilizations does not.

Declarations
Ethics approval and consent to participate: The OHSCAR registry has been approved by ethics committees in Northwestern Malaga, Aragon, Cáceres, Navarra, and the Basque Country. Informed consent was not required because the registry is anonymised. STROBE guidelines for data reporting were followed.

Consent for publication:
Not applicable.
Availability of data and materials: The anonymised data on the participants included in this study will be made available on request. Researchers should submit a proposal for data analysis in line with the overall objectives of the project. The data will be available from the date of publication of this article for Figures Figure 1 Response time in minutes and seconds and percentage of aCPR attempts and survival upon arrival at hospital.

Figure 2
Percentage of cases at home, use of supraglottic airway devices, and medical aetiology. Number of calls and mobilisations per 14-day cumulative incidence period during the pandemic.