What is the Impact of the European Resuscitation Council Covid-19 Guidelines on Resuscitation Quality – A Manakin Study

In an observational manikin study paramedics and emergency physicians performed Advanced-Cardiac-Life-Support in three settings: ERC guidelines 2015 (Control), Covid-19-guidelines as suggested with minimum staff (Covid-19-minimal-personnel); Covid-19-guidelines with paramedics and an emergency physician (Covid-19-advanced-airway-manager). Main outcome measures were no-ow-time, quality metrics as dened by ERC and time intervals to rst chest compression, oxygen supply, intubation and rst rhythm analysis. Data were presented as mean (standard 30 resuscitation scenarios were completed. No-ow-time was markly prolonged in Covid-19-minimal-personnel [113±37 sec] compared to Control [55±9 sec; p<0.001] and Covid-19-advanced-airway-manager [76±38 sec; p<0.001]. In both Covid-19-groups chest compressions started later [Control: 21±5 sec, Covid-19-minimal-personnel: 32±6 sec; Covid-19-advanced-airway-manager: 37±7 sec; each p<0.001 vs. control], but oxygen supply 77±19 sec; Covid-19-minimal-personnel: 29±5 sec; Covid-19-advanced-airway-manager: 34±7 sec; each p<0.001 vs. control] and rst intubation attempt [Control: 178±44 sec; Covid-19-minimal-personnel: 111±14 sec; Covid-19-advanced-airway-manager: 131±20 sec; each p<0.001 vs. control] were earlier than in the control group. succesful 198±48 sec; 181±42 sec; Covid-19-advanced-airway-manager: 130±25 a sec] compared to Covid-19-advanced-airway-manager (p=0.002) and control [19±6 sec; p<0.001]. Time to rst rhythm analysis was more than doubled in Covid-19-minimal-personnel [138±96 sec] compared to control [50±12 sec, p<0.001]. Covid-19-guidelines led to earlier attempts at intubation, delay in starting chest compressions, longer interruption in chest compression and markedly worsen the quality of resuscitation. These negative effects are attenuated by increasing the number of staff and addition of an experienced airway manager. Specic indications for Covid-19-guidelines are urgently required to carefully balance the risk of infection with SARS-CoV-2 for the staff vs. the potentially worse outcome for the patients. signicant in the 72-hour-survival-rate for initial airway a laryngeal co-workers., the initial use of a supraglottic airway device to endotracheal intubation after days of implementation of early debrillation no- low-ow-time


Introduction
The rapid spread of the Covid-19 pandemic around the world poses particular challenges, even for highly developed health systems 1 : In addition to managing the pandemic, maintaining a high quality of medical care for the general population is of fundamental importance 2 .
Before the Covid-19 pandemic guidelines of the European Resuscitation Council (ERC) 2015 provided the best evidence, for carrying out high-quality cardiopulmonary resuscitation in Europe. Based on current evidence the main focus was applied to minimizing no-ow-time by early and correct chest compressions as well as an early rhythm analysis and de brillation 3 .
Currently, there is no valid evidence on the risk of SARS-CoV-2 infection for resuscitation providers [4][5] . However, data from severe acute respiratory syndrome in 2003 and our existing knowledge on the transmission of SARS-CoV-2 indicate that aerosol generating measures during resuscitation such as chest compressions, mouth-to-mouth-and bag-mask-ventilation may be associated with a relevant risk of infection and subsequent morbidity and mortality for medical staff performing resuscitation [6][7][8] . Prioritising rescuer safety given the uncertainty of risk of COVID-19 infection to rescuers, ERC guidelines were adapted during the Covid-19 pandemic with the primary focus on the best possible protection of medical staff, protection of colleagues and bystanders, acknowledging that there might be delays in some interventions 9 . Table 1 summarizes the main differences in the recommendations for resuscitation between Covid-19 guidelines and ERC recommendations 2015 [4][5]9 . However, due to the rapid progress of the present pandemic, these recommendations were published without evidence-based knowledge on their impact on quality of resuscitation and the subsequent potential effects on outcome of the patients 9 . Furthermore, a valid evaluation based on data from registries will be possible only in the remote future. Nevertheless, this information is necessary to optimise the recommendations and prevent or minimise a deterioration of the outcome.
Against this background, the present simulated manikin study evaluates the effects of the Covid-19 ERC guidelines on established quality indicators of cardiopulmonary resuscitation as well as circulatory and respiratory interventions in comparison to the previous recommendations of the ERC 2015 and provides the rst evidence available 1,3-4,9 .

Methods
Ethical approval for this study was provided by the Ethical Committee of the University Hospital of Muenster, Muenster, Germany (Ethical Committee No 2020-504-f-S). The study was performed at the 'Studieninstitut fuer kommunale Verwaltung Westfalen-Lippe, Fachbereich Medizin und Rettungswesen', Bielefeld, Germany. After providing written informed consent, 10 teams of professional paramedics and emergency physicians trained in Advanced-Cardiac-Life-Support (ACLS) were randomly selected and asked to perform resuscitation according to the guidelines of the ERC 2015 with the crew of an ambulance (two paramedics) and one additional paramedic and an emergency physician experienced in advanced airway management according to the emergency physician-based rescue service system in Germany (control group) 3 . After completing these scenarios, the participants were educated and trained in the speci cs of the ERC Covid-19 guidelines [3][4] . The training focused on changes of detection of cardiac arrest, covering the face with an oxygen mask with a reservoir and maximum oxygen ow before starting chest compressions, early use of a de brillator, as well as early videolaryngoscopic intubation with intentional renunciation of chest compressions to minimize aerosol release during intubation 1, 4 . If intubation was not possible, participants were able to perform resuscitation using the oxygen mask or bag-mask-ventilation (Covid-19 pandemic-adapted two-helper method). In a special training course, the participants were intensively trained in videolaryngoscopic intubation under medical supervision.
according to the recommended restriction in staff (Covid-19-minimal-personnel) and then the ERC Covid-19 guidelines in the same team composition as in the control group (Covid-19-advanced-airway-manager) or vice versa.
All teams were given the same scenario in all simulations, which was previously not known to the participants. To enable better comparability, the rhythm in the scenarios was persistent asystole without return-of-spontaneous-circulation. The duration of each scenario was limited to 10 minutes. During the Covid-19 scenarios, participants wore personal protective equipment (PPE) as recommended in the Covid-19 guidelines, but importantly, putting on the equipment was not part of the scenarios.
The primary goal was to achieve the highest possible resuscitation quality taking into account the protection speci cities due to the Covid-19 pandemic 1, 3, 4 . The time required for airway management (intervals to application of oxygen, rst intubation attempt, successful intubation, duration of intubation attempts, number of intubation attempts, unsuccessful intubation attempts, false intubations) as well as indicators of resuscitation quality according to current ERC guidelines (no-ow-time, total number of compressions, depth of compressions, compression frequency, proportion of compressions with correct handposition and su cient compression depth, time to rst rhythm analysis and to rst epinephrine injection, overall dose of epinephrine) were measured 3  The primary outcome was no-ow time. Secondary outcomes were the established quality indicators of resuscitation as mentioned above. The differences between the two interventions and control were assessed using a multilevel model to account for repeated measurement obtained on the same team. A sample size of 9 teams was required for a power of 80% signi cance level of 0.05 under the assumption that the correlation between the three settings within a team was 0.8 to obtain an effect of 0.33. In order to test if there are any differences between the two interventions, a correction to the signi cance level (0.025) was used. Resuscitation data were presented as mean and standard deviation. Differences between intervention groups and control were obtained and tested for signi cance using random intercept multilevel linear models to adjust for repeated measures by the same teams. A Bonferroni correction was applied to the signi cance level to allow a further comparison between the two intervention groups.

Results
Ten different teams of paramedics (male=80%; female=20%, average work experience 12±3 years) and emergency physicians (male=66%; female=33%, average work experience 12±4 years) completed a total of 30 resuscitation scenarios apportioned equally to the three study groups. The results of the quality parameters for chest compressions and circulatory interventions are listed in tables 2 and 3.   Covering the face and oxygen supply in the Covid-19-groups were faster by passive oxygenation than in the control group by bag-mask-ventilation (each p<0.001). Although the rst intubation attempt in the Covid-19-groups was earlier than in the control group (each p<0.001), time to successful intubation did not differ between groups. This was mainly due to the longer intubation time in the Covid-19-minimal-personnel group compared to the Covid-19-advancedairway-manager group (p=0.002) and the control group (p<0.001). Unsuccessful intubation attempts or false intubations were not observed, suggesting a su cient training.
Comparison of the two Covid-19-groups revealed that the prolonged time intervals to rst rhythm analysis, establishment of a vascular access and rst epinephrine application were reduced by increasing the number of staff members and the addition of an advanced airway manager (p<0.001 each vs. Covid-19-minimal-personnel). While in the control group an average of 3.7 (0.7) mg cumulative epinephrine was applied throughout the scenario, lower cumulative epinephrine doses were applied in Covid-19-minimal-personnel and Covid-19-advanced-airway-manager (p<0.001 each).

Discussion
The present manikin study examined the effects of the adaptation of the ERC resuscitation guidelines to the Covid-19 pandemic on established quality indicators of resuscitation. Compared to the ERC guidelines of 2015, no-ow-time was doubled and time intervals to the start of chest compressions, rhythm analysis, vascular access and rst dose of epinephrine were prolonged in the recommended minimal team setting. Although oxygen supply and the rst intubation attempt were performed earlier, earlier successful intubation was not achieved, because time for airway management was prolonged. Notably, increasing the number of team members and the involvement of an advanced airway manager reduced no-ow-time, time for intubation and time intervals to rst rhythm analysis, vascular access and rst dose of epinephrine for the Covid-19 algorithm.
In recent years, no-ow-time (interval between collapse and start of chest compression and chest-compression-free time) and low-ow-time (time between start of chest compressions and return of spontaneous circulation) have been identi ed as central prognostic factors on long-term-survival with good neurological outcome after cardiac arrest [10][11][12][13][14] : Established evidence suggests that even short interruptions in chest compressions worsen patients´ outcome 3, 10, 12, 14 . Therefore, the primary focus of efforts to optimize resuscitation has been to minimize no-ow-time, e.g. by starting early, minimizing interruptions and increasing the quality of chest compressions 3, 10-14 . The present data indicate that during a 10 min scenario, resuscitation according to the Covid-19 algorithm results in a doubling and therefore signi cant extension of no-ow-time.Due to a later start as well as a relevant extension of compression-free phases in the context of airway management, which most likely has a negative impact on the neurological outcome. In addition, the interval to rst rhythm analysis was markedly prolonged in the Covid-19-minimal-personnel group. This is of potentially high relevance too, because rapid assessment and early de brillation in case of initial shockable rhythms can signi cantly improve both, survival rate and neurological outcome in cardiac arrests with shockable rhythms 3, 10, 13 . Accordingly, the use of a de brillator is recommended early in the Covid-19 and ERC-2015 algorithms, assuming that it does not lead to increased aerosol release but signi cantly improves the outcome of the affected patients 3,4 . Potential reasons for the present results could be an increased focus on airway coverage and preparation/performance of intubation as well as the reduced number of staff members. The later hypothesis is supported by the signi cant attenuation of the delay in rst rhythm analysis in the Covid-19-advanced-airway-manager group. However, in the later setting there is an increased number of staff members at risk of infection. Through improved training and increased awareness, it might be possible to improve the time to the rst rhythm analysis even with a reduced number of providers.
Taken together, the protection of the resuscitation providers must be balanced against the potential impairment of outcome for the patients. In view of the large locoregional variability of Covid-19 between and within the affected countries 4 , applying the Covid-19 algorithm in general would potentially expose Covid-19-negative patients to the risk of a worse outcome. Facing the current pandemic situation, it appears necessary to de ne speci c indicators to differentiate between the performance of two resuscitation guidelines with respect to patient care and the safety of the medical personnel. In addition to evidence or symptoms of Covid-19, the contact status of the patient and the daily incidence of Covid-19 on site are of particular importance 15 . Professional resuscitation providers are therefore faced with special challenges to guarantee optimal care for the patients concerned: in addition to the consistent use of appropriate personal protective equipment associated with speci c risks of lowering resuscitation quality (e.g. early fatigue, deterioration of intubation and access conditions), they not only need to be trained in both resuscitation algorithms as well as in airway management, but also in decision pathways to choose the appropriate algorithm.
The implementation of optimal airway management during resuscitation has been the aim of various large-scale studies in recent years and has been a subject of controversial discussion.
While studies within ACLS for professional resuscitation assistants emphasized the importance of bag-mask-ventilation 3, 16, 17 , Wang and co-workers were able to demonstrate a signi cant improvement in the 72-hour-survival-rate for initial airway protection using a laryngeal tube in adult patients with out-ofhospital cardiac arrest 18 . Benger and co-workers., on the other hand, could not demonstrate a favorable effect on the functional outcome in the Airways-2study for the initial use of a supraglottic airway device compared to endotracheal intubation after 30 days 19 . These data support that the importance of advanced airway management was considered secondary to the implementation of early de brillation and minimal no-or low-ow-time in the 2015 ERC guidelines 3 . In view of the transmission path of the SARS-CoV-2 pathogen, Covid-19 guidelines focus on minimization of aerosol release to protect the providers from infection 1,4,5 . Based on the assumption that a su ciently blocked endotracheal tube using a suitable virus lter can effectively prevent aerosol release and thus reduce the exposure of providers, early endotracheal intubation is recommended in Covid-19 guidelines with the interruption of chest compressions, for which there are indications in recent manikin studies 1,4,5,20 . Since the use of videolaryngoscopy increases the rst-pass-success rate even for inexperienced assistants and also allows the airway-manager a greater distance to the mouth, nose and throat, it is currently explicitly recommended 4,21 . In the present study, an earlier start of the rst intubation attempt could be observed for resuscitation according to the Covid-19 algorithm, but the time until successful intubation did not differ between the three groups despite the use of a videolaryngoscope. Notably, speci cally in the Covid-19-minimalpersonnel group there was a signi cant extension of the intubation time and thus a relevant extension of no-ow-time. Deterioration of intubation conditions could additionally be caused by using eye and face protection. However, both Covid-19 groups wore PPE. Although the paramedics in Covid-19-minimalpersonnel had intensive training in performing videolaryngoscopic intubation before the scenarios were carried out, there was a high variability of intubation duration in the Covid-19-minimal-personnel group. While this observation might be reduced by regular training in the future, given the dynamics of the current pandemic situation, it seems unrealistic to train every preclinical paramedic su ciently in the technique of videolaryngoscopic intubation and to equip every rescue vehicle with a videolaryngoscope at short-term.
The present data are in agreement with previous study results showing that prolonged intubation attempts are associated with an increased no-ow-time, which may lead to a delay of return of spontanoues circulation and worsen the outcome 3,17 . Therefore, the presence of an experienced airway manager in these emergency situations may be bene cial. This assumption is supported by the results of the Covid-19-advanced-airway-manager group, that revealed a marked reduction of the time interval to successful intubation as well as the duration of intubation. These results might foster a discussion about the heterogeneity of pre-clinical systems in various countries with the existence of non-emergency physician-based as well as inter-professional character of emergency physician-based systems such as in Germany against the background of the current Covid-19 pandemic 22 .
While covering the face with an aid (e.g. a jacket or similar) is recommended for the lay helpers, covering the airways with an oxygen mask with reservoir is suitable and recommended in connection with professional emergency care in the Covid-19 algorithm 4 . Using an oxygen mask allows the covering of the mouth, nose and throat area as well as an initial passive oxygenation. In a recently published manikin study, which explored strategies to reduce aerosolspread during chest compressions, an oxygen mask on the patient´s face de ected the spread. Therefore it might be suitable to reduce the exposure of providers 20 . In studies of passive oxygenation, an improvement in the survival rate and neurological outcome was demonstrated for the rst minutes of resuscitation even without airway intervention [23][24][25] . The present study shows that passive oxygenation using the Covid-19 algorithm can lead to an earlier application of oxygen than in the previous ERC 2015 algorithm. Therefore, this could represent a reasonable compromise between necessary provider protection on the one hand with a de ection of aerosol-spread and optimal patient care on the other hand, ensuring early oxygen application for the patients, while bag-mask-ventilation or airway interventions are prepared 20, 23-25 .
The current study has some limitations that need to be acknowledged. The installation of personal protective equipment was not included in the individual time intervals. Therefore, a further extension of no-ow-time and additional time intervals might be possible as observed in studies of SARS 26,27 . The scenarios consisted only of non-shockable rhythms. However, it is unlikely that the no-ow time in a corresponding scenario with a shockable rhythm shall deviate similar. Performing resuscitation under investigation during the study may have in uenced the participants and thus may have altered the results (Hawthorne effect). Furthermore, manikin-based investigations provide only a limited reproduction of reality and patients´outcome could not be evaluated. However, the effects are highly signi cant and in view of the current situation with the spread of the Covid-19 pandemic and the expected deterioration in the quality of resuscitation, there are only limited possibilities for the urgently required clinical evaluation of the corresponding guideline changes. In addition, results for the Covid-19 guidelines may improve following more intensi ed training of the providers. On the one hand training success was high, as suggested by the lack of false intubations, for example. On the other hand, due to the pandemic even training of the medical staff is very di cult to organize.

Conclusions
Our manikin study showed that COVID-19 guidelines for a non-shockable cardiac arrest led to earlier attempts at intubation, delay in starting chest compressions and longer interruption in chest compression, which markedly worsen the quality of cardiopulmonary resuscitation. These negative effects are attenuated by increasing the number of team members and the involvement of an experienced airway manager. Speci c indications for the Covid-19 ERC guidelines are urgently required to carefully balance the risk of infection with Sars-CoV-2 for providers vs. the potentially worse outcome for the patients.

List Of Abbreviations
ACLS

Funding
This work was solely supported by institutional and departmental sources.
Authors' contributions Airway and respiratory interventions