Abstract
Background
Resuscitation by laypersons is important in bridging the time between the occurrence of an out-of-hospital cardiac arrest (OHCA) and the arrival of emergency rescue service personnel. Depending on the reason for the cardiac arrest, however, the effectiveness of chest compressions is uncertain. The aim of this study was to explore the impact of lay resuscitation on survival following OHCA of different causes.
Methods
The data set for analysis comprised all cases of cardiac arrest before the arrival of emergency rescue service personnel that were fully documented in the German Resuscitation Registry in the period 2007–2019. The following endpoints related to resuscitation by bystanders were evaluated—separately for each cause—descriptively and by means of multivariate logistic regression analysis: return of spontaneous circulation (ROSC), 30 days’ survival/discharged alive from the hospital, and good neurological function at discharge.
Results
Altogether, 40 604 cases of cardiac arrest were included. Resuscitation by laypersons was carried out in 35.1% of these cases. The rate of ROSC was statistically significantly higher after lay resuscitation for OHCA caused by cardiac events, drowning, intoxication, or central nervous system disorders (overall 48.1% versus 41.0%). For all causes—with the exception of trauma/bleeding to death and sepsis—the endpoint 30 days’ survival/discharged alive was better with lay resuscitation (overall 17.0% versus 9.5%). In multivariate regression analysis, lay resuscitation was associated with improvement of the endpoint 30 days’ survival/discharged alive only for OHCA caused by cardiac events (odds ratio [OR] 1.16) or intoxication (OR 1.81). For all other causes—except hypoxia—lay resuscitation tended to yield better results. Neurological function at discharge was also significantly better (overall 11.5% versus 6.1%) after lay resuscitation for OHCA of all causes except trauma/bleeding to death, hypoxia, and sepsis.
Conclusion
Resuscitation by laypersons is associated with an improved result regarding the endpoint 30 days’ survival/discharged alive in cases of OHCA caused by cardiac events and intoxication. These two groups account for 81% of the resuscitation patients in the study. Because there was also a tendency towards higher survival rates following OHCA of other causes (except hypoxia), laypersons should continue to be encouraged to attempt resuscitation in all cases of OHCA, whatever the cause.
Out-of-hospital cardiac arrest (OHCA) plays an important role in the work of the emergency rescue services. According to recent studies, the annual incidence of OHCA in Europe is around 84.0–87.4 cases/100 000 inhabitants (1, 2). Gräsner et al. (1) evaluated OHCA and found that cardiopulmonary resuscitation (CPR) was carried out by laypersons or emergency rescue service personnel in approximately 49 cases/100 000 inhabitants in the European Union. The most recent annual report of the German Resuscitation Registry (GRR; Deutsches Reanimationsregister) states that the incidence of prehospital resuscitation was 62.6/100 000 inhabitants in Germany in 2019 (3). Despite intensive scientific efforts and public information campaigns in several countries, the survival rate following OHCA is still only 10–13% (1, 3, 4).
Crucial factors for survival after OHCA are early initiation and high-quality performance of basic resuscitation measures (5). Lay resuscitation seems to have an important role in this regard. Not only is it associated with better 30-day survival (6), but a positive effect can still be demonstrated a year later (7). In this respect, it appears relevant that lay resuscitation lengthens the time window for successful defibrillation (8, 9). A further essential component of basic resuscitation measures is the use of automatic external defibrillators (AED), which in recent years have been installed in many public buildings and other places where people congregate. The use of AED by bystanders in OHCA observed to occur in public spaces is associated with significant improvement of both the hospital discharge rate and neurological function following discharge. (Please note that throughout this article, “significant” means statistically significant.)
Although the resuscitation guidelines of the European Resuscitation Council (ERC) recommend chest compressions in cardiac arrest (CA) regardless of the cause, a uniform effect, e.g., in CA associated with trauma, is viewed as unlikely (11). The value of lay resuscitation with chest compressions in CA of non-cardiac origin is therefore uncertain. To the best of our knowledge, no systematic studies on the influence of lay resuscitation depending on the cause of CA have been published.
Our study group has shown that survival and the quality of neurological function following OHCA vary widely depending on the cause of the CA (4). This indicates that the measures comprising CPR are not uniformly successful in CA of different origins. We therefore designed this study to explore, on the basis of data from a large national resuscitation registry, the association between lay resuscitation and survival in OHCA of different causes.
Method
Prospectively collected data were retrospectively analyzed. The GRR data were provided in anonymized form, so that no patient or emergency rescue service center could be identified. The study was approved by the ethics committee of Ulm University (no. 138/19) and conducted in accordance with the tenets of the Declaration of Helsinki 2013.
This study was based on data from all prehospital resuscitations in the period 2007–2019 documented in the GRR with full detail of the primary care module and—in the event of hospital admission—complete information on the further clinical treatment module with data on patient outcome. To increase the data quality, analysis was restricted to the reference centers of the GRR (for detail, see the eMethods) (3). Cases in which OHCA occurred after the arrival of emergency rescue service personnel were excluded.
The data sets included for analysis were sorted by the cause of OHCA documented in the primary care module. Related causes that were practically impossible to distinguish from one another in the out-of-hospital setting were amalgamated (table 1). In accordance with the Utstein Resuscitation Registry templates for OHCA, cases of unknown or other cause were assigned to the category of cardiac events (12). Separating the OHCA by cause, we began by exploring the associations between lay resuscitation and the following endpoints:
Table 1. Documented causes of cardiac arrest in the German Resuscitation Registry and classification for analysis in this study.
| Documented cause (Utstein style) | Classification for analysis |
| – Cardiac – Other/unknown |
– Cardiac, other, and unknown |
| – Trauma – Massive bleeding |
– Trauma and bleeding to death |
| – Hypoxia | – Hypoxia |
| – Drowning | – Drowning |
| – Intoxication | – Intoxication |
| – Intracranial bleeding (including subarachnoid hemorrhage) –Apoplexy (ischemic) |
– Central nervous system disorders |
| – Sepsis | – Sepsis |
| – Sudden infant death syndrome – Metabolic |
Excluded due to low case numbers |
Return of spontaneous circulation (ROSC)
Survival for 30 days or discharge from the hospital alive
Good neurological function at discharge (cerebral performance category [CPC] 1/2)
We then used multivariate binary logistic regression analysis to ascertain the effect of lay resuscitation on the endpoint 30 days’ survival/discharged alive—adjusting for previously identified confounders (for details see eMethods) (13).
The statistical analyses were carried out with the aid of IBM SPSS Statistics, Version 24. Categorical variables were evaluated using the chi-squared test, while the results of multivariate binary logistic regression analysis were expressed as odds ratios (OR) with 95% confidence interval. A p-value < 0.05 was considered to show a statistically significant difference.
Extensive description of the data acquisition with information about the documented cause and details of data evaluation and statistical analysis can be found in the eMethods.
Results
In the period 2007–2019, a total of 43 905 OHCA were fully documented in the GRR, observing the criteria for reference centers. Following exclusion of the cases in which the OHCA only occurred after arrival of the emergency rescue services, 40 604 data sets were included for analysis.
Lay resuscitation took place in 35.1% of cases, and in 1.4% an AED was used. Patients resuscitated by bystanders were younger, more likely to be male, and the cardiac arrest was more likely to be an observed event. These patients were significantly more likely to have a cardiac rhythm amenable to defibrillation when emergency rescue services personnel arrived, had a significantly higher ROSC rate, and their 30-day survival rate was almost twice as high. Patients resuscitated by laypersons were also almost twice as likely to be discharged with good neurological function, defined as CPC 1/2 (table 2).
Table 2. Analysis of data from the German Resuscitation Registry, 2007–2019 *1.
| (Assumed) cause | Lay resuscitatation | Number of cases | Male sex | Mean age (years) | Cardiac arrest observed | Use of AED by layperson | Time before start of CPR (min) | Rhythm defibrillatable at ERS arrival (VF/VT) | ROSC | Thirty days’ survival/discharged alive | CPC 1/2 on discharge |
| All causes | Yes | 14 245 | 67.9% *2 | 67.0 ± 18.0 *2 | 62.1% *2 | 2.5% | 2.8 *2 | 32.5% *2 | 48.1% *2 | 17.0% *2 | 11.5% *2 |
| No | 26 359 | 64.2% | 70.1 ± 16.5 | 38.7% | 0.9% | 9.2 | 19.2% | 41.0% | 9.5% | 6.1% | |
| Cardiac, other, and unknown | Yes | 11 530 | 69.7% *2 | 68.6 ± 15.7 *2 | 63.3% *2 | 2.8% | 2.8 *2 | 38.5% *2 | 47.5% *2 | 18.0% *2 | 12.5% *2 |
| No | 20 590 | 65.0% | 71.8 ± 14.7 | 39.3% | 1.0% | 9.3 | 23.1% | 39.9% | 9.8% | 6.6% | |
| Trauma and bleeding to death | Yes | 472 | 73.7% | 56.4 ± 21.4 *4 | 57.0% *2 | 0.6% | 3.1 *2 | 6.1% | 26.5% | 4.2% | 2.8% |
| No | 1389 | 69.3% | 59.1 ± 20.9 | 38.3% | 0.1% | 9.3 | 4.5% | 26.8% | 3.0% | 1.5% | |
| Hypoxia | Yes | 1603 | 55.9% | 64.6 ± 22.9 | 60.5% *2 | 0.8% | 2.7 *2 | 5.4% | 55.1% | 12.1% *4 | 5.5% |
| No | 3159 | 57.2% | 68.6 ± 18.6 | 39.6% | 0.4% | 8.9 | 4.2% | 52.3% | 10.0% | 5.0% | |
| Drowning | Yes | 100 | 69.0% | 34.5 ± 29.9 *2 | 26.0% | 0.1% | 3.6 *2 | 6.0% | 56.0% *4 | 24.0% *4 | 16.0% *4 |
| No | 105 | 69.5% | 56.0 ± 23.2 | 22.9% | 1.9% | 18.2 | 8.6% | 38.1% | 11.4% | 6.7% | |
| Intoxication | Yes | 216 | 71.3% | 41.9 ± 16.5 *4 | 38.4% *2 | 0.9% | 3.2 *2 | 8.3% | 54.6% *4 | 29.2% *2 | 21.8% *2 |
| No | 378 | 70.9% | 44.9 ± 16.0 | 18.3% | 0.3% | 9.7 | 6.9% | 45.0% | 14.6% | 10.1% | |
| Central nervous system disorder | Yes | 147 | 45.6% | 67.7 ± 17.5 | 67.4% *2 | 1.4% | 2.7 *2 | 17.7% *3 | 67.3% *3 | 18.4% *4 | 13.6% *3 |
| No | 294 | 52.7% | 70.6 ± 15.5 | 40.5% | 0.7% | 8.6 | 9.2% | 53.7% | 10.2% | 5.8% | |
| Sepsis | Yes | 47 | 57.5% | 67.3 ± 21.9 | 63.8% *2 | 0.0% | 2.5 | 10.6% | 46.8% | 8.5% | 2.1% |
| No | 101 | 64.4% | 71.0 ± 17.0 | 24.8% | 1.0% | 6.6 | 5.0% | 49.5% | 3.0% | 1.0% |
*1 The cases investigated were out-of-hospital cardiac arrests that occurred before the arrival of ERS personnel and for which data on subsequent treatment were available. The data included comprise demographic information, basic characteristics, and patient outcome by underlying cause. Statistical analysis by means of chi-squared test was carried out separately for each cause (lay resuscitation versus no lay resuscitation).
*2 p < 0.001; *3 p < 0.01; *4 p < 0.05
AED, Automatic external defibrillator; CPC, cerebral performance category; CPR, cardiopulmonary resuscitation; ERS, emergency rescue services; ROSC, return of spontaneous circulation; VF, ventricular fibrillation; VT, ventricular tachycardia
With regard to cause, the rate of lay resuscitation was highest for drowning and lowest for trauma/ bleeding to death. Significantly higher ROSC rates following lay resuscitation were seen for cardiac origin, drowning, intoxication, and central nervous system disorders. The endpoint 30 days’ survival/discharged alive was significantly higher after lay resuscitation for all causes except trauma/bleeding to death and sepsis.
After adjusting for all potential confounders, multivariate logistic regression analysis revealed a significantly higher survival rate following lay resuscitation for the whole study population, for cardiac origin, and for intoxication. For all other causes—with the exception of hypoxia—there was a tendency towards better survival after lay resuscitation. Only five patients survived OHCA caused by sepsis, too small a group to enable regression analysis (figure).
Figure.
Analysis of data from the German Resuscitation Registry, 2007–2019
The cases investigated were out-of-hospital cardiac arrests that occurred before the arrival of ERS personnel and for which data on subsequent treatment were ?available. Presented here are the endpoints “30 days’ survival”/ “discharged alive” by underlying cause.
Multivariate logistical regression analysis was performed for the effect of lay resuscitation (adjusted for age, sex, health status before CA, CA observed or not, place of CA occurrence, time before arrival of ERS, first recorded ECG rhythm) and analyzed separately for each cause: “no lay resuscitation” = 1 (Nagelkerke’s R-squared 0.39); *1 p < 0.001; *2 p < 0.05
CA, Cardiac arrest; CI, confidence interval; ECG, electrocardiogram; ERS, emergency rescue services
Discussion
The study presented here was the first to explore the associations between lay resuscitation and both survival following OHCA and neurological recovery, related to the cause as diagnosed by the attending emergency physician. For certain causes, the endpoint 30 days’ survival/discharged alive was significantly better after lay resuscitation. Moreover, the proportion of patients discharged from the hospital with good neurological function is almost twice as high after lay resuscitation.
The data analyzed in this study came from the GRR. At the time of the study period, the registry received data from emergency rescue services covering areas of Germany with a total population of around 27 million. Because participation was voluntary, it can only be assumed that the sample is representative; however, earlier analyses of GRR data demonstrated comparability with other registry studies that partly included patients from the whole country (1, 4, 14).
The average lay resuscitation rate during the study period was 35.1%. This figure has risen steadily in recent years, reaching 42.1% by 2019 (3). Despite this increase, the rate of lay resuscitation in Germany is below the European average. A Europe-wide investigation of resuscitations that was performed as part of the EuReCa TWO study in 2017 found an average rate of 58% with a peak of 82% (15). The type of resuscitation carried out by bystanders seems to be associated with survival following OHCA. The survival rate was 14% for chest compressions with mouth-to-mouth resuscitation, significantly higher than the 8% found for chest compressions alone (15).
A positive effect between performance of lay resuscitation following OHCA and both survival of the event and neurological function at discharge was found by Bürger et al. in a study on the effect of the time between OHCA and the arrival of emergency rescue service personnel (5). However, the study adjusted only for cardiac causes in the multivariate analysis.
The largest category of OHCA in our study was “cardiac, other, and unknown,” accounting for 79.1% of the total. The analysis showed a significant benefit of lay resuscitation in this group, even after adjusting for potential influencing factors. Similar findings had already been made in a number of earlier studies, with regard to both 30-day survival (5, 6) and 1-year survival (7), so the result was not surprising. However, the OR found in our study was 1.16, showing a much less pronounced benefit of lay resuscitation than found in the study by Hasselqvist-Ax et al. (OR 2,94) (6). Assuming that our sample was representative, and with an estimated 52 000 out-of-hospital resuscitations, this would nevertheless mean that consistent implementation of lay resuscitation could result in the survival of 400 more patients in Germany each year after OHCA. Barnard et al. had similar findings: although in their study lay resuscitation had no impact on the rate of hospital admission, nevertheless significantly more patients resuscitated by bystanders were discharged alive (16). As a possible reason, the authors assumed that the main cause of death after admission could have been secondary brain damage owing to the longer time without CPR, while ROSC could still be attained because of the somewhat better tolerance of ischemia.
Lay resuscitation also had a significant survival benefit in OHCA due to intoxication. Here the adjusted effect was even higher, with an OR of 1.81. It may be that the cardiac arrest in intoxication is often due to respiratory arrest, which at an early stage is amenable to treatment with CPR. In contrast to OHCA caused by hypoxia due to a pulmonary disorder, in cases of intoxication there is usually no relevant disturbance of gas exchange. This would explain why—after adjusting for potential influencing factors— lay resuscitation is of no benefit in hypoxia-related OHCA, as seen in this study and described in an earlier publication (17).
Our study shows that the survival rate following cardiac arrest caused by drowning is twice as high with lay resuscitation; after adjustment, however, this effect is not significant. The patients resuscitated by bystanders were much younger. Without lay resuscitation the time before CPR was 18 min. The lack of benefit following adjustment could be explained by the presence of a pathomechanism comparable to hypoxia, with impairment of gas exchange decreasing the usefulness of lay resuscitation. This would be in accordance with the findings of a study in which the use of chest compressions together with addition of mouth-to-mouth resuscitation achieved no improvement in neurological outcome versus chest compressions alone (18). However, taking into account the tendency we found towards a benefit of lay resuscitation, with an OR of 1.72 and a wide confidence interval, it seems more likely that the number of cases (205 patients) is too low to permit statistically significant results. A number of studies published in the past 5 years, partly with much higher case numbers, demonstrated a significant benefit of lay resuscitation in drowned patients (19– 21), albeit partly with extremely low survival rates (0.4% versus 0.8%) (21).
In OHCA caused by central nervous system disorders (ischemic and hemorrhagic apoplexy, subarachnoid hemorrhage, basilar artery thrombosis), we found no correlation of lay resuscitation with improvement of the endpoint 30 days’ survival/discharged alive. In this case the limiting factor for survival is not the successful restoration of circulation but the severity of the brain injury, which is ultimately what leads to death. Studies on cardiac arrest owing to subarachnoid hemorrhage showed highly elevated cerebral pressure in initial survivors, and in one study this led to brain death in 46% of cases (22, 23).
Because the number of survivors of OHCA due to sepsis was small in our study, we did not carry out regression analysis on the effect of lay resuscitation. However, the outcome was much worse than that for the sample as a whole (survival rate 3.0% versus 8.5%). The reason for this may well be that these patients often have septic cardiomyopathy and/or septic multiple organ failure, severely reducing the chance of successful treatment (24).
Trauma-associated OHCA had both the lowest ROSC rate and the lowest 30-day survival rate in our study, independent of whether lay resuscitation was carried out. These low survival rates have been described several times before (4, 25– 27), and are understandable in view of the fact that around 85% of these patients have injuries incompatible with survival (28). Nevertheless, there remains a proportion of around 15% of patients in whom survival would be possible or even probable (28). To achieve this, however, the underlying causes of the cardiac arrest must be eliminated swiftly and decisively. Algorithms for this purpose have been published by various groups (29– 31) and since 2015 have formed part of the ERC resuscitation guidelines (11). Because a layperson has no means of treating these causes, it seems understandable that lay resuscitation in this group of patients is not associated with significantly better survival.
Limitations
As is usually the case in retrospective analyses, the results must be interpreted cautiously with regard to causality. In this review, the prospective data acquisition and the analysis with the aid of multivariate logistic regression, adjusting for relevant confounders, help to reduce potential sources of error. Nevertheless, it cannot be excluded that other, unconsidered factors might influence the results. For ethical and logistical reasons, a randomized or blinded prospective design does not come into question for a study on this topic.
A further limitation is the validity of the documented causes of OHCA on which the analysis of the data was based. In the case of hospital admission and subsequent diagnostic evaluation, the cause entered in the registry by the attending emergency physician is most likely correct, because the emergency physician and hospital staff have the opportunity to discuss the underlying cause later, at which time the registry entry can be confirmed or corrected (by the emergency physician: the hospital cannot alter the cause documented in the registry’s primary care data set).
If the patient dies before arrival of the emergency physician, the latter has to determine the underlying cause of cardiac arrest on the basis of medical history, symptoms, and the known diagnostic details. It is not possible to use an algorithm for diagnosis following cardiac arrest, as proposed for example by Chan et al., owing to the lack of opportunity for diagnostic investigation at the location of the event (32). To our knowledge, there have been no systematic studies of the accuracy of diagnosis in these circumstances. Kürkciyan et al., however, showed that in initially successfully resuscitated patients the diagnosis made in the shock room, taking account of the patient’s medical history and symptoms, agreed with the clinical diagnosis in 83% of cases (33). These findings suggest that the quality of the data used in this study was not essentially affected by the fact that the cause was determined by the emergency physician at the place of deployment. Furthermore, to increase the validity of the documentation in this respect, causes that are difficult to differentiate in the out-of-hospital setting, such as trauma/bleeding to death or intracranial hemorrhage/apoplexy/subarachnoid hemorrhage, were amalgamated for analysis (table 1).
Summary
Overall, a significant positive association was found between lay resuscitation and 30-day survival for OHCA caused by cardiac events or intoxication, corresponding to 81% of the cases analyzed. For other causes—with the exception of hypoxia—there was also a tendency towards a positive effect of resuscitation by bystanders; however, the differences were not statistically significant on multivariate analysis. Here, treatment of the underlying cause in the context of extended resuscitation measures seems more urgent. No negative effects of lay resuscitation are discernable, however, so—to keep the algorithm simple—resuscitation by laypersons should continue to be recommended and performed, regardless of the cause of cardiac arrest.
Supplementary Material
eMethods
Data acquisition
The German Resuscitation Registry (GRR; Deutsches Reanimationsregister) is a nationwide registry for documentation and scientific analysis of out-of-hospital and in-hospital cardiac arrests (CA). Participation by out-of-hospital emergency rescue services and hospitals is voluntary. In 2019, data were contributed by 88 emergency rescue centers covering a population of 27 million (3). The GRR data set comprises three modules:
The basis for the data acquired is the Utstein Style Protocol (12). For better comparability, and to enable integration of the GRR into the European Registry of Cardiac Arrest (EuReCa), some segments and parameters were documented in more detail (34). A full description of the GRR and all parameters recorded can be found in the articles by Gräsner et al. (14, 35).
The data on out-of-hospital cardiopulmonary resuscitation (CPR) in the GRR’s primary care module are entered by the emergency rescue services, based on the emergency care physician’s deployment report. For patients who die at the place where their CA occurred, the cause of death is determined by the attending emergency care physician. In the event of hospital admission, the cause of the out-of-hospital CA (OHCA) established by hospital staff can be transmitted to the emergency rescue service, whereupon the data set can be altered if necessary. The causes available for submission to the GRR are listed in Table 1. No studies have been carried out on the validity of the documented causes. However, the organizers of the GRR strive to attain the highest possible data quality. Internal evaluations are performed for this purpose. In 2019, for the first time, the GRR therefore selected so-called reference centers, which had to fulfill certain criteria (incidence, success rate, completeness of data sets and information on subsequent care). The aim was to prevent selection bias arising from the entry of only a few (special) cases. The conditions that had to be met by these reference centers are described in full by Fischer et al. (3).
At hospitals which participate voluntarily in the GRR, for each patient admitted after OHCA a member of the hospital staff will enter the data into the module “further clinical treatmente.” How the registry is managed is decided by each hospital individually, only the data set is prescribed. Because these data are entered and validated independently of the emergency rescue service concerned, the person entering them is blinded for the emergency rescue service center with regard to the outcome.
Every participating emergency rescue service center and every participating hospital is responsible for the submission of its data. The out-of-hospital data are usually entered into the registry by the attending emergency care physician on the basis of their report. Each data set is then checked for plausibility before it is approved and transmitted to the GRR, either by the medical director of the emergency rescue service center concerned or by a person nominated by the medical director. Data submission by hospitals is comparable, each data set being checked by a superior before it is approved and sent. The data documentation and the procedure for reporting to the GRR have also been described in a recently published article by our research team (4).
Data analysis
The data sets from all out-of-hospital resuscitations documented in the GRR for the period 2007–2019 were provided in anonymized form. The cases included for analysis were all those from the reference centers fulfilling the criteria described above for which a complete module “primary care” was present—plus, for those involving hospital admission, a complete module “further clinical treatment” with information on patient outcome. Cases in which the OHCA did not occur until after the arrival of emergency rescue service personnel were excluded. To minimize selection bias, one of the criteria for reference centers was that the the proportion of data sets on subsequent care should exceed 30% of the hospital admissions (as otherwise possibly only isolated or special or positive cases could be reported). This proportion was evaluated separately for each emergency rescue service center and each year of participation.
The remaining data sets were classified by documented cause of OHCA. Similar entities that could be differentiated only with difficulty or not at all in the out-of-hospital setting, e.g., trauma and bleeding to death or different central nervous system disorders, were amalgamated. Causes found in fewer than 100 cases over the whole period were not analyzed further. According to the stipulations of the Utstein Resuscitation Registry Templates, cases in which the cause was “unknown” or “other” were assigned to the category “cardiac events” (12). The causal categories used in analysis are shown in Table 1.
For each category separately, the association of lay resuscitation with the following endpoints was investigated:
In a second step, multivariate logistic regression analysis was used to assess the effect of lay resuscitation on the endpoint 30 days’ survival/discharged alive, adjusting for potential confounders (age, sex, health status before CA, whether CA was observed, location where CA occurred, time between notification and arrival of emergency rescue services, first recorded ECG rhythm). The health status before CA was documented as pre-emergency status, a categorical variable on a scale of 1 to 5. Comparable to the determination of the Cardiac Arrest Survival Score (CRASS), this variable was divided into three categories (no, minor, or severe/unknown prior morbidity) (36).
Statistical analysis
Microsoft Excel 2016 (Microsoft Corp., Redmond, WA, USA) was used for data processing and for presentation of the results, and the statistical analyses were performed using IBM SPSS Statistics, Version 24. The categorical variables were analyzed by means of the chi-squared test. The effect of lay resuscitation was ascertained by means of multivariate binary logistic regression analysis, adjusting for potential confounders, with calculation of odds ratios and 95% confidence intervals. Metric values were expressed as arithmetic mean and standard deviation. For both chi-squared tests for individual characteristics and logistic regression analysis for the primary outcome (30 days’ survival), a p-value < 0.05 was taken to show a statistically significant difference.
Primary care
Further clinical treatment
Long-term care
Return of spontaneous circulation (ROSC)
Survival for 30 days or discharge from the hospital alive
Good neurological function at discharge (cerebral performance category [CPC] 1/2)
Acknowledgments
Translated from the original German by David Roseveare
Footnotes
Conflict of interest statement Prof. Fischer is a member of the organizing committee of the German Resuscitation Registry.
The remaining authors declare that no conflict of interest exists.
References
- 1.Gräsner JT, Lefering R, Koster RW, et al. EuReCa ONE—27 Nations, ONE Europe, ONE Registry A prospective one month analysis of out-of-hospital cardiac arrest outcomes in 27 countries in Europe. Resuscitation. 2016;105:188–195. doi: 10.1016/j.resuscitation.2016.06.004. [DOI] [PubMed] [Google Scholar]
- 2.Berdowski J, Berg RA, Tijssen JG, Koster RW. Global incidences of out-of-hospital cardiac arrest and survival rates: systematic review of 67 prospective studies. Resuscitation. 2010;81:1479–1487. doi: 10.1016/j.resuscitation.2010.08.006. [DOI] [PubMed] [Google Scholar]
- 3.Fischer M, Wnent J, Gräsner JT, et al. Jahresbericht des Deutschen Reanimationsregisters - Außerklinische Reanimation 2019. Anästh Intensivmed. 2020;61:V89–V93. [Google Scholar]
- 4.Gässler H, Fischer M, Wnent J, Seewald S, Helm M. Outcome after pre-hospital cardiac arrest in accordance with underlying cause. Resuscitation. 2019;138:36–41. doi: 10.1016/j.resuscitation.2019.02.039. [DOI] [PubMed] [Google Scholar]
- 5.Bürger A, Wnent J, Bohn A, et al. The effect of ambulance response time on survival following out-of-hospital cardiac arrest An analysis from the German resuscitation registry. Dtsch Ärztebl Int. 2018;115:541–548. doi: 10.3238/arztebl.2018.0541. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Hasselqvist-Ax I, Riva G, Herlitz J, et al. Early cardiopulmonary resuscitation in out-of-hospital cardiac arrest. N Engl J Med. 2015;372:2307–2315. doi: 10.1056/NEJMoa1405796. [DOI] [PubMed] [Google Scholar]
- 7.Kragholm K, Wissenberg M, Mortensen RN, et al. Bystander efforts and 1-year-outcome in out-of-hospital cardiac arrest. N Engl J Med. 2017;376:1737–1747. doi: 10.1056/NEJMoa1601891. [DOI] [PubMed] [Google Scholar]
- 8.Waalewijn RA, Nijpels MA, Tijssen JG, Koster RW. Prevention of deterioration of ventricular fibrillation by basic life support during out-of-hospital cardiac arrest. Resuscitation. 2002;54:31–36. doi: 10.1016/s0300-9572(02)00047-3. [DOI] [PubMed] [Google Scholar]
- 9.Christenson J, Andrusiek D, Everson-Stewart S, et al. Chest compression fraction determines survival in patients with out-of-hospital ventricular fibrillation. Circulation. 2009;120:1241–1247. doi: 10.1161/CIRCULATIONAHA.109.852202. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Pollack RA, Brown SP, Rea T, et al. Impact of bystander automated external defibrillator use on survival and functional outcomes in shockable observed public cardiac arrests. Circulation. 2018;137:2104–2113. doi: 10.1161/CIRCULATIONAHA.117.030700. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Truhlár A, Deakin CD, Soar J, et al. European Resuscitation Council Guidelines for Resuscitation 2015 Section 4. Cardiac arrest in special circumstances. Resuscitation. 2015;95:148–201. doi: 10.1016/j.resuscitation.2015.07.017. [DOI] [PubMed] [Google Scholar]
- 12.Perkins GD, Jacobs IG, Nadkarni VM, et al. Cardiac arrest and cardiopulmonary resuscitation outcome reports: update of the utstein Resuscitation Registry Templates for out-of-hospital cardiac arrest. Resuscitation. 2015;96:328–340. doi: 10.1016/j.resuscitation.2014.11.002. [DOI] [PubMed] [Google Scholar]
- 13.Gräsner JT, Meybohm P, Lefering R, et al. ROSC after cardiac arrest - the RACA score to predict outcome after out-of-hospital cardiac arrest. Eur Heart J. 2011;32:1649–1656. doi: 10.1093/eurheartj/ehr107. [DOI] [PubMed] [Google Scholar]
- 14.Gräsner JT, Seewald S, Bohn A, et al. Deutsches Reanimationsregister: Wissenschaft und Reanimationsforschung. Anaesthesist. 2014;63:470–476. doi: 10.1007/s00101-014-2324-9. [DOI] [PubMed] [Google Scholar]
- 15.Gräsner JT, Wnent J, Herlitz J, et al. Survival after out-of-hospital cardiac arrest in Europe—results of the EuReCa TWO study. Resuscitation. 2020;148:218–226. doi: 10.1016/j.resuscitation.2019.12.042. [DOI] [PubMed] [Google Scholar]
- 16.Barnard EBG, Sandbach DD, Nicholls TL, Wilson AW, Ari Ercole A. Prehospital determinants of successful resuscitation after traumatic and non-traumatic out-of-hospital cardiac arrest. Emerg Med J. 2019;36:333–339. doi: 10.1136/emermed-2018-208165. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Fukuda T, Ohashi-Fukuda N, Kondo Y, Sera T, Yahagi N. Effect of rescue breathing by lay rescuers for out-of-hospital cardiac arrest caused by respiratory disease: a nationwide, population-based, propensity score-matched study. Intern Emerg Med. 2017;12:493–501. doi: 10.1007/s11739-016-1472-0. [DOI] [PubMed] [Google Scholar]
- 18.Fukuda T, Ohashi-Fukuda N, Hayashida K, Kondo Y, Kukita I. Bystander-initiated conventional vs compression-only cardiopulmonary resuscitation and outcomes after out-of-hospital cardiac arrest due to drowning. Resuscitation. 2019;145:166–174. doi: 10.1016/j.resuscitation.2019.08.026. [DOI] [PubMed] [Google Scholar]
- 19.Claesson A, Lindqvist J, Herlitz J. Cardiac arrest due to drowning - changes over time and factors of importance for survival. Resuscitation. 2014;85:644–648. doi: 10.1016/j.resuscitation.2014.02.006. [DOI] [PubMed] [Google Scholar]
- 20.Tobin JM, Ramos WD, Pu Y, Wernicki PG, Quan L, Rossano JW. Bystander CPR is associated with improved neurologically favourable survival in cardiac arrest following drowning. Resuscitation. 2017;115:39–43. doi: 10.1016/j.resuscitation.2017.04.004. [DOI] [PubMed] [Google Scholar]
- 21.Fukuda T, Ohashi-Fukuda N, Hayashida K, Kukita I. Association of bystander cardiopulmonary resuscitation and neurological outcome after out-of-hospital cardiac arrest due to drowning in Japan, 2013-2016. Resuscitation. 2019;141:111–120. doi: 10.1016/j.resuscitation.2019.06.005. [DOI] [PubMed] [Google Scholar]
- 22.Skrifvars MB, Parr MJ. Incidence, predisposing factors, management and survival following cardiac arrest due to subarachnoid haemorrhage: a review of the literature. Scand J Trauma Resusc Emerg Med. 2012;20 doi: 10.1186/1757-7241-20-75. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Shapiro S. Management of subarachnoid hemorrhage patients who presented with respiratory arrest resuscitated with bystander CPR. Stroke. 1996;27:1780–1782. doi: 10.1161/01.str.27.10.1780. [DOI] [PubMed] [Google Scholar]
- 24.Morgan RW, Fitzgerald JC, Weiss SL, Nadkarni VM, Sutton RM, Berg RA. Sepsis-associated in-hospital cardiac arrest: epidemiology, pathophysiology, and potential therapies. J Crit Care. 2017;40:128–135. doi: 10.1016/j.jcrc.2017.03.023. [DOI] [PubMed] [Google Scholar]
- 25.Escutnaire J, Genin M, Babykina E, et al. Traumatic cardiac arrest is associated with lower survival rate vs medical cardiac arres—results from the French national registry. Resuscitation. 2018;131:48–54. doi: 10.1016/j.resuscitation.2018.07.032. [DOI] [PubMed] [Google Scholar]
- 26.Djarv T, Axelsson C, Herlitz J, Stromsoe A, Israelsson J, Claesson A. Traumatic cardiac arrest in Sweden 1990-2016—a population-based national cohort study. Scand J Trauma Resusc Emerg Med. 2018;26 doi: 10.1186/s13049-018-0500-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Evans CC, Petersen A, Meier EN, et al. Prehospital traumatic cardiac arrest: Management and outcomes from the resuscitation outcomes consortium epistry-trauma and PROPHET registries. J Trauma Acute Care Surg. 2016;81:285–293. doi: 10.1097/TA.0000000000001070. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Kleber C, Giesecke MT, Tsokos M, Haas NP, Buschmann CT. Trauma-related preventable deaths in Berlin 2010: need to change prehospital management strategies and trauma management education. World J Surg. 2013;37:1154–1161. doi: 10.1007/s00268-013-1964-2. [DOI] [PubMed] [Google Scholar]
- 29.Lockey DJ, Lyon RM, Davies GE. Development of a simple algorithm to guide the effective management of traumatic cardiac arrest. Resuscitation. 2013;84:738–742. doi: 10.1016/j.resuscitation.2012.12.003. [DOI] [PubMed] [Google Scholar]
- 30.Kleber C, Giesecke MT, Lindner T, Haas NP, Buschmann CT. Requirement for a structured algorithm in cardiac arrest following major trauma: epidemiology, management errors, and preventability of traumatic deaths in Berlin. Resuscitation. 2014;85:405–410. doi: 10.1016/j.resuscitation.2013.11.009. [DOI] [PubMed] [Google Scholar]
- 31.Gässler H, Helm M, Lampl L, Kulla M. Kardiopulmonale Reanimation nach Trauma. Der Notarzt. 2015;31:210–216. [Google Scholar]
- 32.Chen N, Callaway CW, Guyette FX, et al. Arrest etiology among patients resuscitated from cardiac arrest. Resuscitation. 2018;130:33–40. doi: 10.1016/j.resuscitation.2018.06.024. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Kürkciyan I, Meron G, Behringer W, et al. Accuracy and impact of presumed cause in patients with cardiac arrest. Circulation. 1998;98:766–771. doi: 10.1161/01.cir.98.8.766. [DOI] [PubMed] [Google Scholar]
- 34.Gräsner JT, Herlitz J, Koster RW, Rosell-Ortiz F, Stamatakis L, Bossaert L. Quality management in resuscitation—towards a European Cardiac Arrest Registry (EuReCa) Resuscitation. 2011;82:989–994. doi: 10.1016/j.resuscitation.2011.02.047. [DOI] [PubMed] [Google Scholar]
- 35.Gräsner JT, Meybohm P, Fischer M, et al. A national resuscitation registry of out-of-hospital cardiac arrest in Germany—a pilot study. Resuscitation. 2009;80:199–203. doi: 10.1016/j.resuscitation.2008.10.008. [DOI] [PubMed] [Google Scholar]
- 36.Seewald S, Wnent J, Lefering R, et al. CaRdiac Arrest Survival Score (CRASS)—a tool to predict good neurological outcome after out-of-hospital cardiac arrest. Resuscitation. 2020;146:66–73. doi: 10.1016/j.resuscitation.2019.10.036. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eMethods
Data acquisition
The German Resuscitation Registry (GRR; Deutsches Reanimationsregister) is a nationwide registry for documentation and scientific analysis of out-of-hospital and in-hospital cardiac arrests (CA). Participation by out-of-hospital emergency rescue services and hospitals is voluntary. In 2019, data were contributed by 88 emergency rescue centers covering a population of 27 million (3). The GRR data set comprises three modules:
The basis for the data acquired is the Utstein Style Protocol (12). For better comparability, and to enable integration of the GRR into the European Registry of Cardiac Arrest (EuReCa), some segments and parameters were documented in more detail (34). A full description of the GRR and all parameters recorded can be found in the articles by Gräsner et al. (14, 35).
The data on out-of-hospital cardiopulmonary resuscitation (CPR) in the GRR’s primary care module are entered by the emergency rescue services, based on the emergency care physician’s deployment report. For patients who die at the place where their CA occurred, the cause of death is determined by the attending emergency care physician. In the event of hospital admission, the cause of the out-of-hospital CA (OHCA) established by hospital staff can be transmitted to the emergency rescue service, whereupon the data set can be altered if necessary. The causes available for submission to the GRR are listed in Table 1. No studies have been carried out on the validity of the documented causes. However, the organizers of the GRR strive to attain the highest possible data quality. Internal evaluations are performed for this purpose. In 2019, for the first time, the GRR therefore selected so-called reference centers, which had to fulfill certain criteria (incidence, success rate, completeness of data sets and information on subsequent care). The aim was to prevent selection bias arising from the entry of only a few (special) cases. The conditions that had to be met by these reference centers are described in full by Fischer et al. (3).
At hospitals which participate voluntarily in the GRR, for each patient admitted after OHCA a member of the hospital staff will enter the data into the module “further clinical treatmente.” How the registry is managed is decided by each hospital individually, only the data set is prescribed. Because these data are entered and validated independently of the emergency rescue service concerned, the person entering them is blinded for the emergency rescue service center with regard to the outcome.
Every participating emergency rescue service center and every participating hospital is responsible for the submission of its data. The out-of-hospital data are usually entered into the registry by the attending emergency care physician on the basis of their report. Each data set is then checked for plausibility before it is approved and transmitted to the GRR, either by the medical director of the emergency rescue service center concerned or by a person nominated by the medical director. Data submission by hospitals is comparable, each data set being checked by a superior before it is approved and sent. The data documentation and the procedure for reporting to the GRR have also been described in a recently published article by our research team (4).
Data analysis
The data sets from all out-of-hospital resuscitations documented in the GRR for the period 2007–2019 were provided in anonymized form. The cases included for analysis were all those from the reference centers fulfilling the criteria described above for which a complete module “primary care” was present—plus, for those involving hospital admission, a complete module “further clinical treatment” with information on patient outcome. Cases in which the OHCA did not occur until after the arrival of emergency rescue service personnel were excluded. To minimize selection bias, one of the criteria for reference centers was that the the proportion of data sets on subsequent care should exceed 30% of the hospital admissions (as otherwise possibly only isolated or special or positive cases could be reported). This proportion was evaluated separately for each emergency rescue service center and each year of participation.
The remaining data sets were classified by documented cause of OHCA. Similar entities that could be differentiated only with difficulty or not at all in the out-of-hospital setting, e.g., trauma and bleeding to death or different central nervous system disorders, were amalgamated. Causes found in fewer than 100 cases over the whole period were not analyzed further. According to the stipulations of the Utstein Resuscitation Registry Templates, cases in which the cause was “unknown” or “other” were assigned to the category “cardiac events” (12). The causal categories used in analysis are shown in Table 1.
For each category separately, the association of lay resuscitation with the following endpoints was investigated:
In a second step, multivariate logistic regression analysis was used to assess the effect of lay resuscitation on the endpoint 30 days’ survival/discharged alive, adjusting for potential confounders (age, sex, health status before CA, whether CA was observed, location where CA occurred, time between notification and arrival of emergency rescue services, first recorded ECG rhythm). The health status before CA was documented as pre-emergency status, a categorical variable on a scale of 1 to 5. Comparable to the determination of the Cardiac Arrest Survival Score (CRASS), this variable was divided into three categories (no, minor, or severe/unknown prior morbidity) (36).
Statistical analysis
Microsoft Excel 2016 (Microsoft Corp., Redmond, WA, USA) was used for data processing and for presentation of the results, and the statistical analyses were performed using IBM SPSS Statistics, Version 24. The categorical variables were analyzed by means of the chi-squared test. The effect of lay resuscitation was ascertained by means of multivariate binary logistic regression analysis, adjusting for potential confounders, with calculation of odds ratios and 95% confidence intervals. Metric values were expressed as arithmetic mean and standard deviation. For both chi-squared tests for individual characteristics and logistic regression analysis for the primary outcome (30 days’ survival), a p-value < 0.05 was taken to show a statistically significant difference.
Primary care
Further clinical treatment
Long-term care
Return of spontaneous circulation (ROSC)
Survival for 30 days or discharge from the hospital alive
Good neurological function at discharge (cerebral performance category [CPC] 1/2)