If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Norwegian National Advisory Unit on Prehospital Emergency Medicine (NAKOS) and Air Ambulance Dept., Division of Prehospital Services, Oslo University Hospital and University of Oslo, Oslo, Norway
Norwegian National Advisory Unit on Prehospital Emergency Medicine (NAKOS) and Air Ambulance Dept., Division of Prehospital Services, Oslo University Hospital and University of Oslo, Oslo, Norway
Department of Clinical and Experimental Cardiology, Amsterdam UMC, Academic Medical Centre, University of Amsterdam, Amsterdam, The NetherlandsNetherlands Heart Institute, Utrecht, The Netherlands
Corresponding author at: Department of Cardiology, Heart Center, Amsterdam UMC, University of Amsterdam, Meibergdreef 15, 1105 AZ Amsterdam, The Netherlands.
Prior research suggests that the proportion of a shockable initial rhythm in out-of-hospital cardiac arrest (OHCA) declined during the last decades. This study aims to investigate if this decline is still ongoing and explore the relationship between location of OHCA and proportion of a shockable initial rhythm as initial rhythm.
Methods
We calculated the proportion of patients with a shockable initial rhythm between 2006–2015 using pooled data from the COSTA-group (Copenhagen, Oslo, Stockholm, Amsterdam). Analyses were stratified according to location of OHCA (residential vs. public).
Results
A total of 19,054 OHCA cases were included. Overall, the total proportion of cases with a shockable initial rhythm decreased from 42% to 37% (P < 0.01) from 2006 to 2015. When stratified according to location, the proportion of cases with a shockable initial rhythm decreased for OHCAs at a residential location (34% to 27%; P = 0.03), while the proportion of a shockable initial rhythm was stable among OHCAs in public locations (59%–57%; P = 0.2). During the last years of the study period (2011–2015), the overall proportion of a shockable initial rhythm remained stable (38%–37%; P = 0.45); this was observed for both residential and public OHCA.
Conclusion
We found a decline in the proportion of patients with a shockable initial rhythm in OHCAs at a residential location; this decline levelled off during the second half of the study period (2011–2015). In public locations, we observed no decline in shockable initial rhythm over time.
A shockable initial rhythm (SIR) is one of the most important predictors of survival after an out-of-hospital cardiac arrest (OHCA). Patients with a SIR are more likely to survive if they are quickly defibrillated.
Incidence and survival outcome according to heart rhythm during resuscitation attempt in out-of-hospital cardiac arrest patients with presumed cardiac etiology.
Another explanation of lower observed rates of SIR is longer delays to first electrocardiographic (ECG) recording. As time from collapse passes, SIR will eventually degenerate into asystole. Thus, patients presenting in asystole may have actually had a SIR at the time of the collapse. Such a delay (e.g. caused by longer response intervals and/or lower rates of witnessed collapse) is generally longer in OHCAs occurring at a residential location, as opposed to those occurring in a public location.
Because these OHCA patients may also face the abovementioned decline in proportion of a SIR, the question has arisen whether public access defibrillation initiatives to involve automated external defibrillators (AED) in the resuscitation effort are worthwhile for OHCA patients at a residential location.
Therefore, it is important to study whether the decline in proportion of SIR is continuing. This study aims to determine whether the proportion of SIR is still declining over a 10-year time period up to 2016 across multiple emergency medical services (EMS) in four different European countries. Secondly, this study aims to determine whether this decline in proportion of SIR is different between OHCAs occurring at a residential location and those at public locations.
Methods
COSTA group
The COSTA group is a collaboration network of resuscitation researchers in Copenhagen (Denmark), Oslo (Norway), Stockholm (Sweden), and Amsterdam (the Netherlands), which aims at joining research efforts in OHCA and early defibrillation. Data from OHCA events across all COSTA sites were merged and analyzed in a central research database within the COSTA collaboration. This study was performed in line with the Helsinki declaration.
Ethical approval was granted by the ethical boards for each study site.
Setting, study regions and emergency medical service systems
Denmark: central Copenhagen
Central Copenhagen covers 97 km2 and holds approximately 680,000 inhabitants. Incidence of OHCA was 81 cases per 100.000 in 2014. The EMS of Copenhagen is a two-tiered system comprising ambulances equipped with defibrillators providing basic life support (BLS) and physician-staffed mobile emergency care units providing advanced life support (ALS). OHCA patients with an EMS resuscitation attempt are included in the registry. Also, OHCA patients with AED defibrillation by bystanders only (either with or without return of spontaneous circulation (ROSC) at arrival of the ambulance) are included in this registry. If already deceased, OHCA patients are not included in the registry. Only physicians can decide to terminate a resuscitation effort.
Norway: Oslo
Before 2009, Oslo EMS covered a population of approximately 550,000 inhabitants in 454 km2, and was served with one-tiered ambulance service manned with paramedics and emergency medical technicians. Incidence of OHCA was 57 cases per 100.000 in 2014. After 2013, Oslo and Akershus EMS were merged to serve a population of approximately 1.2 million inhabitants in a mixed urban and rural area of 5371 km2. Before 2013, OHCA patients were included if treatment by ambulance (either CPR for at least 30 s or defibrillation) was started. After 2013 the inclusion criteria changed to: treatment by either bystanders, first responders, or ambulance. Patients already deceased were not included in the registry. Paramedics may decide to terminate the resuscitation effort without consulting a physician, and declare death. OHCA patients with AED defibrillation only and ROSC at arrival of the ambulance were not included in the registry in the study period.
Sweden: Stockholm region
Stockholm region covers 6519 km2 and has 2.1 million inhabitants. Incidence of OHCA was 38 cases per 100.000 per year in 2014. There is a two-tiered EMS system in Sweden for responses to all medical emergencies. Both tiers consist of EMS units with registered nurses in prehospital care providing ALS and with AED or manual defibrillation. All OHCA patients treated with CPR and/or defibrillation initiated by bystanders, first responders or EMS as well as OHCA patients with AED defibrillation only and ROSC at arrival of the ambulance are included in this registry. If already deceased, OHCA patients are not included in the registry. Paramedics may decide to terminate the resuscitation effort without consulting a physician, and declare death.
The Netherlands: province of North Holland
The province of North Holland covers 2404 km2 and has 2.4 million inhabitants. Incidence of OHCA was 46 cases per 100.000 in 2014. In response to an OHCA, two ambulances of a single tier are dispatched. Ambulance personnel are equipped with manual defibrillators and perform ALS. OHCA patients with AED defibrillation by bystanders or first responders only and ROSC at arrival of the ambulance are also included in this registry. If already deceased, OHCA patients are not included in the registry. Paramedics may decide to terminate the resuscitation effort without consulting a physician. If so, a physician comes to the scene to declare death. Also, OHCA patients are included in the registry only if they have been resuscitated by the EMS for >2 min.
Study question, study population and data collection
The primary purpose of this paper was to determine whether the proportion of SIR changed during the study period, and whether this differed between OHCAs occurring at a residential or public location. We included patients in whom EMS personnel attempted resuscitation for OHCA between January 1, 2006 and December 31, 2015. We excluded patients with obvious non-cardiac cause of the arrest and EMS-witnessed OHCAs. All registries used the Utstein template for data reporting.
Cardiac arrest and cardiopulmonary resuscitation outcome reports: update and simplification of the Utstein templates for resuscitation registries. A statement for healthcare professionals from a task force of the international liaison committee on resuscitation (American Heart Association, European Resuscitation Council, Australian Resuscitation Council, New Zealand Resuscitation Council, Heart and Stroke Foundation of Canada, InterAmerican Heart Foundation, Resuscitation Council of Southern Africa).
For the Oslo region, only data from 2006 to 2008 and 2012 to 2015 were included, while for the Copenhagen region, only data from 2008 to 2015 were included, for reasons of availability. All data were anonymized and pooled into a central COSTA database.
Definitions
The initial rhythm recorded by EMS manual defibrillator or AED (whichever was first) was categorized as shockable (ventricular tachycardia, ventricular fibrillation or shock by an AED) or non-shockable (asystole, pulseless electrical activity). Outcome was defined as proportion of SIR rather than incidence of SIR, in order to attenuate effects of registry differences. Location of collapse was retrieved from EMS-call or dispatch form. All non-residential locations were considered public. Long term care facilities were classified as residential location. Time to EMS-arrival was defined as the difference between the time of the EMS-call to the dispatch centre and the time of vehicle stop of the first ambulance, except for the Copenhagen region: here, time to EMS-arrival was defined as the differences between time of dispatch of the first ambulance and time of vehicle stop. Defibrillator connection time was defined as the time interval between EMS-call and connection of the defibrillator device to the patient (EMS manual defibrillator or AED, whichever came first; North Holland region only). Other resuscitation characteristics included were witnessed collapse, bystander CPR and defibrillation by AED.
Data analysis
Proportions of SIR, using the pooled COSTA database, were calculated for each study year (2006–2015), as well as for each study region separately. Categorical variables were presented as percentages and continuous variables as mean and standard deviation (SD) or as median and interquartile range (IQR) depending on the data distributions. To compare categorical data, the trend in time was tested using chi-square statistics. To compare continuous data, the trend in time was tested using the Jonckheere-Terpstra test. To determine whether trends were consistent during the study period, we separately analyzed the study years 2006–2010 (Period 1) and 2011–2015 (Period 2). In addition, analyses were stratified according to OHCA location.
Since the included OHCA cases originated from four different study regions, we performed a multilevel analysis to account for region effects. Furthermore, time trends in proportion of SIR were tested using regression analysis while adjusting for study region and/or resuscitation characterics where applicable.
Finally, proportions of SIR were analyzed in relation to defibrillator connection time. All statistical tests were two-tailed, and a P-value of <0.05 was considered to be statistically significant. Statistics were performed in SPSS 24 (IBM Corporation, Armonk, NY).
Results
A total of 19,054 EMS-treated OHCA patients were included for analysis (Table 1), of which 13,181 (69%) occurred at a residential location (eTable 1) and 5834 (31%) in a public location (eTable 1). Baseline characteristics per study region are provided in Table 2, Table 3.
Table 1Baseline and resuscitation characteristics of OHCA patients with presumed cardiac cause.
Oslo, North Holland and Stockholm regions: time EMS call to scene arrival. Copenhagen region: time of dispatch of the first ambulance to scene arrival.
Variables are denoted as cases (percentage), trend in time tested using Chi-square statistic.
140 (9%)
159 (10%)
177 (9%)
206 (11%)
278 (15%)
336 (18%)
353 (20%)
358 (18%)
383 (18%)
456 (21%)
<0.01
82 (0.4%)
Abbreviations: AED, automated external defibrillator; CPR, cardio-pulmonary resuscitation; NR not reported.
a Data represents 4 months.
b Data represents 6 months.
c Median (25th–75th percentile), trend in time tested using Jonckheere-Terpstra test.
d Variables are denoted as cases (percentage), trend in time tested using Chi-square statistic.
e Oslo, North Holland and Stockholm regions: time EMS call to scene arrival. Copenhagen region: time of dispatch of the first ambulance to scene arrival.
Median age increased from 69 to 70 years (P for trend <0.01). Patients were more often male, proportions varying between 68% to 73% (P for trend = 0.44) (Table 1). The proportion of witnessed collapse decreased from 74% to 65% (P for trend <0.01). The proportion of bystander CPR increased from 56% to 74% (P for trend <0.01), as well as the proportion of AED defibrillation which increased from 9% to 21% (P for trend <0.01) (Fig. 1A, Table 1). The proportion of OHCAs occurring at public locations (Fig. 1A) did not change over time (Table 1). Stratification according to location showed that similar trends were observed for both OHCAs occurring at residential (Fig. 1B, Table 2) and in public locations (Fig. 1C, Table 3). Both in public and residential locations, bystander CPR increased (public 69%–86%, residential: 50%–69%, both P for trend <0.01), as did AED defibrillation (public 13%–35%, residential: 7%–14%, both P for trend <0.01).
Fig. 1(A) Prehospital resuscitation characteristics per year in total. (B) Prehospital resuscitation characteristics per year at residential location. (C) Prehospital resuscitation characteristics per year in public.
Oslo, North Holland and Stockholm regions: time EMS call to scene arrival. Copenhagen region: time of dispatch of the first ambulance to scene arrival.
Variables are denoted as cases (percentage), trend in time tested using Chi-square statistic.
74 (7%)
80 (7%)
74 (6%)
86 (7%)
132 (10%)
153 (12%)
161 (13%)
161 (12%)
177 (12%)
214 (14%)
<0.01
61 (0.5%)
Abbreviations: AED, automated external defibrillator; CPR, cardio-pulmonary resuscitation; NR not reported.
a Data represents 4 months.
b Data represents 6 months.
c Median (25th–75th percentile), trend in time tested using Jonckheere-Terpstra test.
d Variables are denoted as cases (percentage), trend in time tested using Chi-square statistic.
e Oslo, North Holland and Stockholm regions: time EMS call to scene arrival. Copenhagen region: time of dispatch of the first ambulance to scene arrival.
During the total study period the proportion of SIR decreased from 42% to 37% (P < 0.01) (Fig. 2A). When stratified according to OHCA location, the proportion of SIR decreased statistically significant for OHCAs at a residential location (34%–27%; P < 0.01) (Fig. 2B), but not for OHCAs in a public location (59%–57%; P = 0.24) (Fig. 2C).
Fig. 2(A) Proportions shockable initial rhythm per year in total; and per region. (B) Proportions shockable initial rhythm per year at residential location; and per region. (C) Proportions shockable initial rhythm per year in public; and per region.
Multilevel analysis indicated that study region did not affect the slope of the association between study year and proportion of SIR (P < 0.05). Accordingly, adding ‘region’ to a logistic regression with study year as determinant of SIR did not meaningfully affect the odds ratio (OR) of study year (ORunadjusted study year = 1.019, p < 0.001; OR adjusted study year = 1.016, p = 0.003).
During Period 1, the total proportion of SIR declined (42%–37%; P < 0.01) (Fig. 2A). The proportion of SIR for OHCA at a residential location also declined (34%–28%; P = 0.03) (Fig. 2B), but remained stable for OHCA in a public location (59%–60%; P = 0.53) (Fig. 2C). However, during Period 2, there was no decrease in proportion of SIR, neither overall (38%–37%; P = 0.45) (Fig. 1A), nor when stratified according to location (residential: 30%–27%; P = 0.40; public location: 59%–58%; P = 0.50 [Fig. 2B, C]).
To further explore a possible explanation of the stabilized proportion of SIR in the residential location, we separately analyzed resuscitation characteristics in Period 1 and Period 2. During Period 1 we observed a decrease in witnessed status (P < 0.01) but proportion of bystander CPR (P = 0.33) remained stable. During Period 2, we observed a stable proportion of witnessed status (P = 0.17) but the proportion of bystander CPR increased (P < 0.01) (eTable 6). However, adding bystander CPR to a logistic regression with year as determinant of SIR did not meaningfully affect the OR of study year (ORunadjusted study year = 1.001, P = 0.944; ORadjusted study year = 1.017, P = 0.386).
Proportion of shockable initial rhythm in relation to defibrillator connection time
Fig. 3 shows the proportion of SIR in relation to defibrillator connection time. If time from EMS-call to defibrillator connection was 8−10 min, the proportion of SIR was still 48% (95% CI 46–51). For this delay, the proportion of SIR for OHCA at a residential location was 41% (95% CI 38–43) and for OHCA in a public location 71% (95% CI 67–75). Results divided in Period 1 and Period 2 are shown in eFig. 1.
Fig. 3Proportion shockable initial rhythm and time to defibrillator connection in minutes (2006–2015).
Over the total study period (2006–2015), the proportion of SIR declined in four regions from four European countries and this decline was primarily observed in OHCAs at residential locations. However, when limiting the analysis to more recent study years (2011–2015), the proportion of SIR remained stable. When time from EMS-call to defibrillator connection time was short (8−10 min), the proportion of SIR was still high with 41% for OHCA at a residential location, and 71% for OHCA in a public location.
Levelling off of decline in SIR and prior literature
Although a decline in proportion of SIR was present in Period 1, we found that the decline in SIR has levelled off during Period 2. The decline in SIR in Period 1 may result from an absolute decrease of patients at risk for SIR or from an absolute increase in non-SIR OHCAs. First, an absolute decrease of patients at risk for a SIR may be a result of primary prevention measures and/or medical changes that resulted in improved treatment of ischemic heart disease (e.g., more use of β-adrenoceptor blockers) and/or more widespread use of implantable cardioverter defibrillators.
Second, an absolute increase in non-SIR OHCAs may result from population changes such as higher aged OHCA patients. Also, it has been suggested that co-morbidities such as obstructive pulmonary disease has been associated with non-SIR OHCAs, whereby recent studies suggested an increasing disease burden in OHCA patients (in particular with advancing age).
The levelling off in the decline of the proportion of SIR observed in Period 2 may, at least partly, be influenced by an increasing rate of bystander CPR and the stabilized rate of witnessed status (while this rate decreased in Period 1). Being witnessed during collapse and an increased chance on receiving bystander CPR reduces the no-flow time, which may lead to slower degeneration of a SIR.
Guidelines 2000 for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care
Part 4: the automated external defibrillator: key link in the chain of survival. The American Heart Association in Collaboration with the International Liaison Committee on Resuscitation.
this gave rise to the question whether AED initiatives are worthwhile for OHCA patients at residential locations.
Results from the present study showed that the observed decline in SIR has levelled off, in particular in OHCAs occurring at a residential location. In addition, the present study adds important information, as we can demonstrate that the proportion of SIR in OHCA patients at a residential location with a short delay (8−10 min) between EMS-call and defibrillator connection is still high. This cut-off point of 8−10 min is important because many international EMS systems adopted a 8-minute response time for EMS units responding to life-threatening events (i.e., in order to maximize survival chances after OHCA).
Furthermore, rates of AED defibrillation doubled during the study period. As previously shown, use of AEDs at a residential location decreases time from EMS-call to defibrillator connection and is associated with increased survival rates in patients with a SIR.
We therefore recommend continuous efforts to improve resuscitation care at residential locations, with strong emphasis on introducing or extending AED initiatives to facilitate early defibrillation (such as public access defibrillation programs).
This recommendation is important, in particular, because approximately 70% of OHCAs occur at a residential location while survival rates after OHCA occurring at home are significantly lower than after OHCA occurring in public places.
Thus, for residential locations, even a modest increase in survival would have a substantial impact in absolute numbers of lives saved.
Differences in proportion of SIR across study sites
Differences in proportions of SIR have been observed across the study sites. This may be a result of regional differences in inclusion in the registry, since the definition of an EMS-treated OHCA differs across the sites. For instance, in the Netherlands, an OHCA is included in the registry if the OHCA is treated by the EMS for >2 min. The definition in Oslo includes a duration of >30 s EMS treatment and in Denmark no minimal duration is used. Therefore, the included proportion of OHCAs with a worse prognosis may have been higher in the Scandinavian regions when compared to the Dutch region, resulting in a lower proportion of SIR. Also, OHCA patients with AED defibrillation only and ROSC at arrival of the ambulance were not included in the Oslo registry, whereas these cases (with high chances of presenting with a SIR) were included in the other registries. However, the use of AEDs was very low in the Oslo registry.
Limitations
Some important limitations need to be considered; first, as with all observational studies, we were only able to study associations. A causal relation between any of the variables studied cannot be determined in this cohort. Second, missing data occurred as not every site collected data of each study year (i.e., missing data for the years 2006–2007 [Copenhagen] and 2009–2012 [Oslo]. Finally, although all study regions used the Utstein template to collect data, differences may exist because of: (1) variation in the interpretation of the Utstein definitions; (2) regional differences in the EMS system and (3) differences in inclusion to the registries, as mentioned above. To attenuate the effect of these differences between participating study regions, we chose to use proportions instead of incidences. Also, we stratified results by study region and performed a multilevel analysis.
Strengths
A major strength of the present study is the comprehensiveness of the collected data sets, including data from EMS-dispatch centre, paramedics and hospital thereby providing a complete picture of the circumstances of the OHCAs.
Conclusion
A small decline in proportion of SIR was observed in the study period 2006–2015, in particular in OHCAs occurring at a residential location. In the second half of this study period (2011–2015), this decline has levelled off.
Funding
This work has received funding from the European Union’s Horizon 2020 research and innovation program under acronym ESCAPE-NET, registered under grant agreement No. 733381 and The Netherlands CardioVascular Research Initiative (Dutch Heart Foundation, Dutch Federation of University Medical Centers, Netherlands Organization for Health Research and Development, Royal Netherlands Academy of Sciences – (CVON2018-30 Predict2) (Netherlands)). Laerdal Foundation for Acute Medicine (Denmark). The Swedish Heart–Lung Foundation and Laerdal Foundation (Stockholm), Stryker Emergency Care, Zoll Medical, Philips Medical, Cardiac Science and Defibtech,
Competing interests
None declared.
Acknowledgements
We are greatly indebted to all participating emergency medical service dispatch centers, regional ambulance services and first responders, as well as all members of the cardiac arrest registries for their cooperation and support as well as their willingness to collaborate and share their data.
Appendix A. Supplementary data
The following is Supplementary data to this article:
Incidence and survival outcome according to heart rhythm during resuscitation attempt in out-of-hospital cardiac arrest patients with presumed cardiac etiology.
Cardiac arrest and cardiopulmonary resuscitation outcome reports: update and simplification of the Utstein templates for resuscitation registries. A statement for healthcare professionals from a task force of the international liaison committee on resuscitation (American Heart Association, European Resuscitation Council, Australian Resuscitation Council, New Zealand Resuscitation Council, Heart and Stroke Foundation of Canada, InterAmerican Heart Foundation, Resuscitation Council of Southern Africa).
Guidelines 2000 for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care
Part 4: the automated external defibrillator: key link in the chain of survival. The American Heart Association in Collaboration with the International Liaison Committee on Resuscitation.
30125-8&id=gr1.jpg){kind=link}
30125-8&id=gr2.jpg){kind=link}
30125-8&id=gr3.jpg){kind=link}