Abstract
Background
Pulseless electrical activity (PEA) and asystole account for 81% of initial in-hospital cardiac arrest (IHCA) rhythms in the U.S.A. These “non-shockable” rhythms are often grouped together in resuscitation research and practice. We hypothesized that PEA and asystole are distinct initial IHCA rhythms with distinguishing features.
Methods
This was an observational cohort study using the prospectively collected nationwide Get With The Guidelines®-Resuscitation registry. Adult patients with an index IHCA and an initial rhythm of PEA or asystole between the years of 2006 and 2019 were included. Patients with PEA vs. asystole were compared with respect to pre-arrest characteristics, resuscitation practice, and outcomes.
Results
We identified 147,377 (64.9%) PEA and 79,720 (35.1%) asystolic IHCA. Asystole had more arrests in non-telemetry wards (20,530/147,377 [13.9%] PEA vs. 17,618/79,720 [22.1%] asystole). Asystole had 3% lower adjusted odds of ROSC (91,007 [61.8%] PEA vs. 44,957 [56.4%] asystole, aOR 0.97, 95%CI 0.96–0.97, P<0.01); there was no difference in survival to discharge (28,075 [19.1%] PEA vs. 14,891 [18.7%] asystole, aOR 1.00, 95%CI 1.00–1.01, P=0.63). Duration of resuscitation for those without ROSC were shorter for asystole (29.8 [±22.5] minutes in PEA vs. 26.2 [±21.5] minutes in asystole, adjusted mean difference −3.05 95%CI −3.36–−2.74, P<0.01).
Interpretation
Patients suffering IHCA with an initial PEA rhythm had patient and resuscitation level differences from those with asystole. PEA arrests were more common in monitored settings and received longer resuscitations. Even though PEA was associated with higher rates of ROSC, there was no difference in survival to discharge.
Keywords: Cardiac arrest, Cardiac rhythm, Resuscitation
Introduction
In-hospital cardiac arrests (IHCA) most commonly present with pulseless electrical (PEA) or asystole, which together account for 81% of initial rhythms in the United States 1. Not only are these “non-shockable” rhythms the most common presenting initial rhythm, the proportion of non-shockable rhythms has been steadily increasing over time 2. IHCAs with an initial non-shockable rhythm carry higher rates of mortality than those with initial “shockable” (ventricular fibrillation and pulseless ventricular tachycardia) rhythms. Previous investigations have reported survival rates in shockable rhythms to be 2 to 5 times higher than in non-shockable rhythms 3,4. Current American Heart Association/Emergency Cardiovascular Care (AHA/ECC) guidelines group the management of cardiac arrest with initial pulseless electrical activity (PEA) and initial asystole together in the same treatment algorithm due to both rhythms being categorized as “non-shockable”. The management of these rhythms relies on high-quality cardiopulmonary resuscitation (CPR), early epinephrine, and identifying and reversing the cause of the arrest 5,6.
Although the treatment algorithm for the initial rhythms of PEA and asystole are the same, there may be important differences in pre-arrest characteristics, resuscitation practices, and outcomes 7. Given the high prevalence and high mortality of IHCA with initial rhythms of PEA and asystole, identifying and understanding any differences between these rhythms is crucial. Historically, and likely as a result of identical treatment algorithms, cardiac arrest research and quality improvement initiatives do not separate these rhythms, and instead report them together as “non-shockable” 6,8–13.
In this study, we use the AHA Get With The Guidelines®–Resuscitation (GWTG®-R) database to investigate differences between PEA and asystole for patients who suffer IHCA. Our hypothesis is that these rhythms represent distinct entities, both in regard to pre-arrest/resuscitation characteristics and response to therapy.
Materials and Methods
Data Source
All data for this research was from the GWTG®-R database. The Get With The Guidelines®–Resuscitation program is provided by the American Heart Association. Hospitals participating in the registry submit clinical information regarding the medical history, hospital care, and outcomes of consecutive patients hospitalized for cardiac arrest using an online, interactive case report form and Patient Management Tool™ (IQVIA, Parsippany, New Jersey). All participating institutions were required to comply with local regulatory and privacy guidelines and, if required, to secure institutional review board approval. Because data were used primarily at the local site for quality improvement, sites were granted a waiver of informed consent under the common rule. IQVIA serves as the data collection (through their Patient Management Tool – PMT™) and coordination center for the American Heart Association/American Stroke Association Get With The Guidelines® programs. The institutional review board at the Albert Einstein College of Medicine deemed use of the de-identified GWTG®-R as non-human subjects research (IRB number 2021–13275).
Study Population
We examined adult (≥18 years old) patients with IHCA from the GWTG®-R database that occurred between the years 2006 and 2019. We included all patients who had an initial rhythm of either PEA or asystole. Non-index arrests, patients with no recorded initial rhythm (or a rhythm that was not PEA or asystole), events occurring in a pediatric area, and events without hospital-level information were excluded.
Definitions
The GWTG®-R database contains precisely defined variables derived from the Utstein in-hospital guidelines 14. IHCA in the adult database incorporates all adults (≥18 years old) including visitors, employees, and staff within a facility (including ambulatory care areas) that experience a resuscitation event. Resuscitation events must include the following elements: 1) acute respiratory compromise requiring assisted ventilation or respiratory compromise leading to cardiopulmonary arrest requiring chest compressions and/or defibrillation or cardiopulmonary arrest that requires chest compressions and/or defibrillation; 2) an emergency response by facility personnel; 3) a resuscitation record completed for the event. Those included in the PEA group were patients who had an initial rhythm documented as pulseless electrical activity; subsequent rhythms during the arrest did not impact group assignment. The same principle was applied to those patients with an initial rhythm of asystole.
Statistical Analysis
Baseline characteristics are described both for the overall population and separately for each initial rhythm. Categorical variables are presented as counts and percentages, continuous variables are presented as means with standard deviations or medians with interquartile range (IQR), as appropriate. Comparisons between categorical variables were performed using Fisher’s exact test and continuous variables were compared using either a Student’s t-test or a Wilcoxon rank-sum test, depending on the distribution of the data. A p-value of <0.05 was considered the a-priori significance threshold.
Generalized estimating equations were used to compare patient outcomes accounting for clustering by site. The final adjusted model included age, sex, race, arrest location, diabetes, renal insufficiency, hepatic insufficiency, respiratory insufficiency, congestive heart failure, malignancy, prior myocardial infarction, acute stroke, pneumonia, sepsis, acute heart failure, hypotension, acute myocardial infarction, trauma, vasopressor use, ventilator use, event year, illness category, time of day, weekend or holiday, census driven region, number of beds in the facility, teaching hospital status, and rural or urban hospital. Covariates were chosen by the authors as most likely to influence outcomes of interest. Missing data were not imputed or included in the analysis (Supplemental Table 1). Outcomes examined include return of spontaneous circulation (ROSC), survival to hospital discharge, and cerebral performance category (CPC) score. We explored duration of resuscitation (DOR) for the rhythms both in all patients and limited to those who did not achieve ROSC. When analyzing GWTG®-R quality improvement recognition measures, we limited time to epinephrine to values between 0 and 60 minutes to exclude erroneous values, and we did not examine time to defibrillation since we are evaluating only non-shockable initial rhythms, in which defibrillation is not recommended by guidelines. Finally, trends over time were explored for the outcomes by including the year of the event in the regression model and comparing outcomes in the reference year (2006) to subsequent years. The analysis of trends over time were performed for both PEA and asystole separately, and are compared visually in the figures in the manuscript. All statistical analysis was performed with STATA/SE, version 17.0 (College Station, TX, StataCorp LP, USA), and all of the figures were generated with GraphPad Prism version 9.4.1.
Results
Population characteristics
The study population included 227,097 patients who had experienced IHCA with an initial rhythm of either PEA 147,377 (64.9%) or asystole 79,720 (35.1%). See Figure 1 for details on cohort selection. Patient and hospital-level demographics can be found in Table 1. Evaluating initial rhythm by location revealed that patients with PEA had a higher proportion of arrests in the intensive care unit (ICU) (76,916/147,377 [52.2%] PEA vs. 34,451/79,720 [43.2%] asystole) and asystole had a higher proportion of arrests in non-telemetry wards (20,530/147,377 [13.9%] PEA vs. 17,618/79,720 [22.1%] asystole). There was a similar proportion of arrests occurring in the emergency department (ED) for both rhythms (16,877/147,377 [11.5%] PEA vs. 9,040/79,720 [11.3%] asystole). Patients with an initial rhythm of PEA had more diagnoses of respiratory insufficiency (66,126 [44.9%] PEA vs. 30,842 [38.7%] asystole), more peri-arrest hypotension (34,026 [23.1%] PEA vs. 14,708 [18.5%] asystole), more use of peri-arrest vasopressors (39,002 [26.5%] PEA vs. 17,624 [22.1%] asystole), more peri-arrest ventilator use (68,670 [46.6%] PEA vs. 32,205 [40.4%] asystole), and were less likely to be at night (43,809 [29.7%] PEA vs. 27,512 [34.5] asystole) as compared to asystole (Table 1).
Figure 1:
Study flow chart
Final inpatient locations that were included: adult coronary care unit, adult intensive care unit, cardiac catheterization lab, delivery suite, diagnostic intervention area, emergency department, general inpatient area, operating room, post-anesthesia recovery room, telemetry unit, step down unit
Table 1:
Cohort characteristics
| Baseline Characteristic | Initial Rhythm PEA Total 147,377 (64.9%) |
Initial Rhythm Asystole Total 79,720 (35.1%) |
Overall Total 227,097 (100%) |
|||
|---|---|---|---|---|---|---|
|
| ||||||
| Age, mean (SD) | 65.0 | (15.7) | 65.7 | (16.0) | 65.2 | (15.8) |
| Gender men, N (%) | 85,025 | (57.7) | 45,246 | (56.8) | 130,271 | (57.4) |
| Race, N (%) | ||||||
| White | 96,385 | (65.4) | 53,954 | (67.7) | 150,339 | (66.2) |
| Black | 36,540 | (24.8) | 18,043 | (22.6) | 54,583 | (24.0) |
| Other | 3,598 | (2.4) | 1,852 | (2.3) | 5,450 | (2.4) |
| Unknown | 10,854 | (7.4) | 5,871 | (7.4) | 16,725 | (7.4) |
| Admitting diagnosis, N (%) | ||||||
| Medical Cardiac | 49,719 | (33.7) | 24,676 | (31.0) | 74,395 | (32.8) |
| Medical Non-Cardiac | 68,276 | (46.3) | 38,095 | (47.8) | 106,371 | (46.8) |
| Surgical Cardiac | 7,961 | (5.4) | 4,308 | (5.4) | 12,269 | (5.4) |
| Surgical Non-Cardiac | 15,238 | (10.3) | 9,232 | (11.6) | 24,470 | (10.8) |
| Other | 6,054 | (4.1) | 3,309 | (4.2) | 9,363 | (4.1) |
| Pre-existing condition, N (%) | ||||||
| Diabetes | 45,272 | (30.7) | 23,700 | (29.7) | 68,972 | (30.4) |
| Renal insufficiency | 49,883 | (33.9) | 26,018 | (32.6) | 75,901 | (33.4) |
| Hepatic insufficiency | 11,944 | (8.1) | 6,210 | (7.8) | 18,154 | (8.0) |
| Respiratory insufficiency | 66,126 | (44.9) | 30,842 | (38.7) | 96,968 | (42.7) |
| Congestive heart failure | 30,376 | (20.6) | 14,039 | (17.6) | 44,415 | (19.6) |
| Malignancy | 16,320 | (11.1) | 9,162 | (11.5) | 25,482 | (11.2) |
| Prior myocardial infarction | 18,177 | (12.3) | 8,844 | (11.1) | 27,021 | (11.9) |
| Acute stroke | 5,354 | (3.6) | 2,782 | (3.5) | 8,136 | (3.6) |
| Pneumonia | 20,156 | (13.7) | 10,520 | (13.2) | 30,676 | (13.5) |
| Sepsis | 26,909 | (18.3) | 13,589 | (17.1) | 40,498 | (17.8) |
| Acute heart failure | 20,413 | (13.9) | 9,714 | (12.2) | 30,127 | (13.3) |
| Hypotension | 40,671 | (27.6) | 17,423 | (21.9) | 58,094 | (25.6) |
| Acute myocardial infarction | 17,377 | (11.8) | 8,036 | (10.1) | 25,413 | (11.2) |
| Trauma | 6,845 | (4.6) | 3,689 | (4.6) | 10,534 | (4.6) |
| Hospital geographic region, N (%) | ||||||
| Mid-Atlantic (North) | 23,962 | (16.3) | 13,256 | (16.6) | 37.218 | (16.4) |
| Mid-Atlantic (South) | 39,219 | (26.6) | 21,129 | (26.5) | 60,348 | (26.6) |
| Central (North) | 33,369 | (22.6) | 16,515 | (20.7) | 49,884 | (22.0) |
| Central (South) | 29,983 | (20.3) | 17,183 | (21.6) | 47,166 | (20.8) |
| Pacific | 20,844 | (14.1) | 11,637 | (14.6) | 32,481 | (14.3) |
| Hospital size, N (%) | ||||||
| Small (<200) | 12,616 | (8.6) | 8,276 | (10.4) | 20,892 | (9.2) |
| Medium (200–349) | 29,889 | (20.3) | 17,942 | (22.5) | 47,831 | (21.1) |
| Large (≥350) | 103,207 | (70.0) | 52,466 | (65.8) | 155,673 | (68.6) |
| Hospital type, N (%) | ||||||
| Teaching hospital | 133,483 | (90.6) | 70,464 | (88.4) | 203,974 | (89.8) |
| Urban hospital | 139,376 | (94.6) | 74,589 | (93.6) | 213,965 | (94.2) |
| Arrest timing, N (%) | ||||||
| Nighttime arrest | 43,809 | (29.7) | 27,512 | (34.5) | 71,321 | (31.4) |
| Weekend/holiday arrest | 47,825 | (32.5) | 26,586 | (33.4) | 74,411 | (32.8) |
| Pre-arrest support, N (%) | ||||||
| Vasopressor use | 39,002 | (26.5) | 17,624 | (22.1) | 56,626 | (24.9) |
| Ventilator use | 68,670 | (46.6) | 32,205 | (40.4) | 100,875 | (44.4) |
|
| ||||||
| Arrest Location, N (%) |
Initial Rhythm PEA Total 147,377 (64.9%) |
Initial Rhythm Asystole Total 79,720 (35.1%) |
Overall Total 227,097 (100%) |
|||
|
| ||||||
| Intensive care unit | 76,916 | (52.2) | 34,451 | (43.2) | 111,367 | (49.0) |
| Emergency department | 16,877 | (11.5) | 9,040 | (11.3) | 25,917 | (11.4) |
| Inpatient care areas | 41,739 | (28.3) | 30,040 | (37.7) | 71,779 | (31.6) |
| Inpatient telemetry | 21,209 | (14.4) | 12,422 | (15.58) | 33,631 | (14.8) |
| Inpatient non-telemetry | 20,530 | (13.9) | 17,618 | (22.1) | 38,148 | (16.8) |
| Othera | 11,803 | (8.0) | 6,167 | (7.7) | 17,970 | (7.9) |
| Unknown | 42 | (< 0.1) | 22 | (< 0.1) | 64 | (<0.1) |
|
| ||||||
| Peri-arrest vitals, mean (SD) b |
Initial Rhythm PEA Total 147,377 (64.9%) |
Initial Rhythm Asystole Total 79,720 (35.1%) |
Overall Total 227,097 (100%) |
|||
|
| ||||||
| Systolic blood pressure | 110.8 | (30.4) | 112.9 | (29.6) | 111.5 | (30.2) |
| Heart rate | 95.0 | (24.6) | 91.7 | (24.4) | 94.0 | 24.6 |
| Respiratory rate | 22.6 | (8.1) | 22.1 | (7.8) | 22.4 | 8.0 |
| Percent oxygen saturation | 93.8 | (8.9) | 94.0 | (8.7) | 93.9 | 8.9 |
| Temperature in °C | 36.7 | (1.1) | 36.7 | (1.1) | 36.7 | 1.1 |
-Other locations included cardiac catheterization lab, delivery suite, diagnostic intervention area, operating room, post-anesthesia recovery room, or documented as other in the database
-The vitals closest to the arrest but prior to it were used, values that were not consistent with physiologically possible vitals were excluded
Trends over time
The adjusted odds of having an initial rhythm of asystole, as compared to PEA, has been decreasing from 2006 to 2019 (Supplemental Figure 1). Both ROSC and survival to discharge have been increasing from 2006 to 2019 (Figure 2). Favorable neurologic function (CPC score of 1 or 2) in those that survived to discharge remained stable from 2006 to 2019 (Supplemental Figure 2). For PEA, ROSC improved from 51.8% to 65.9% and survival to discharge improved from 14.4% to 21.5%. For asystole, ROSC improved from 44.4% to 61.0% and survival to discharge improved from 12.7% to 18.2%. The mean duration of resuscitative efforts from 2006 to 2019 has been decreasing for both PEA (20.6 ± 18.7 minutes to 18.9 ± 19.4 minutes) and asystole (20.1 ± 17.6 minutes to 18.4 ± 18.6 minutes) (Figure 3). For those who did not achieve ROSC, the mean duration of resuscitation has increased from 2006 to 2019 for both PEA (26.8 ± 26.2 minutes to 30.9 ± 30.3 minutes) and asystole (23.6 ± 22.9 minutes to 27.1 ± 26.3 minutes).
Figure 2:
ROSC and survival to discharge for PEA and asystole
2a: Percentage of ROSC and survival to discharge for each initial rhythm by year
2b: Adjusted odds with 95% confidence intervals of ROSC and survival to discharge each year with respect to the reference year 2006 for each initial rhythm
Figure 3:
Duration of resuscitation for PEA and asystole
3a: Average duration of resuscitation each year both overall and for those who did not achieve ROSC, for each initial rhythm
3b: Adjusted mean difference with 95% confidence intervals of duration of resuscitation each year with respect to the year reference year 2006 for each initial rhythm
3c: Adjusted mean difference with 95% confidence intervals of duration of resuscitation each year with respect to the year reference year 2006 for each initial rhythm in only those who did not achieve ROSC
Duration of Resuscitation
The average duration of resuscitation (DOR) was 20.2 ± 20.2 minutes for those with an initial rhythm of PEA and 19.7 ± 19.1 minutes for those with an initial rhythm of asystole (adjusted mean difference −0.96 minutes 95%CI −1.17 – −0.75, P<0.01). Resuscitative efforts in those who did not achieve ROSC were longer in patients with an initial rhythm of PEA (29.8 ± 22.5 minutes in PEA vs. 26.2 ± 21.5 minutes in asystole, adjusted mean difference −3.02 minutes 95%CI −3.33 – −2.71, P<0.01).
GWTG® Quality Performance metrics
Patients with an initial rhythm of asystole were less likely to have witnessed arrests (131,793 [89.4%] PEA vs. 63,790 [80.0%] asystole, aOR 1.06, 95%CI 1.06 – 1.07, P<0.01), took longer to receive epinephrine during the arrest (2.07 ± 2.97 minutes in PEA vs. 2.44 ± 3.14 minutes in asystole, adjusted mean difference 0.14 minutes, 95%CI 0.10 – 0.17, P<0.01), and had a lower proportion of arrests where epinephrine was given in the first 5 minutes and 59 seconds (114,506 77.7% PEA vs. 58,229 73.0% asystole, aOR 0.99, 95%CI 0.99 – 0.99, P<0.01) (Supplemental Table 2).
Outcomes
A total of 91,007 (61.8%) patients with an initial rhythm of PEA achieved ROSC, and 44,957 (56.4%) of those with an initial rhythm of asystole achieved ROSC. Adjusted analysis revealed a 3% lower odds of achieving ROSC for those with an initial rhythm of asystole (91,007 [61.8%] PEA vs. 44,957 [56.4%] asystole, aOR 0.97, 95%CI 0.96 – 0.97, P<0.01) (Table 2). There was no statistically significant difference in survival to hospital discharge between the initial rhythms (28,075 [19.1%] PEA vs. 14,891 [18.7%] asystole, aOR 1.00, 95%CI 1.00 – 1.02, P=0.72). When examining only patients who achieved ROSC, those with asystole had a 1% higher adjusted odds of survival to discharge than those with PEA (28,073 [30.9%] PEA vs. 14,886 [33.1%] asystole, aOR 1.01, 95%CI 1.01 – 1.02, P<0.01). For patients discharged alive, 17,742 (63.2%) with PEA had a favorable neurologic function (CPC score of 1 or 2), and 9,351 (62.8%) with asystole had a favorable neurologic function (aOR 1.00, 95%CI 0.99 – 1.01, P=0.99). For unwitnessed arrests, 1,626 (10.4%) of those with PEA and 1,051 (6.6%) of those with asystole survived to discharge with a favorable neurologic function (CPC score of 1 or 2). Supplemental Figure 3 reports outcomes both for patients that had an unwitnessed arrest and patients that had a witnessed arrest.
Table 2:
Outcomes
| Outcome | Initial Rhythm PEA Total 147,377 (64.9%) |
Initial Rhythm Asystole Total 79,720 (35.1%) |
Adjusted Odds Ratio [95% CI] | P Value | ||
|---|---|---|---|---|---|---|
|
| ||||||
| Achieved ROSC, N (%) | 91,007 | (61.8) | 44,957 | (56.4) | 0.97 [0.96, 0.97] | < 0.01 |
| Discharged alive, N (%) | 28,075 | (19.1) | 14,891 | (18.7) | 1.00 [1.00, 1.01] | 0.72 |
| Discharged alive in patients that achieved ROSC only, N (%) | 28,075/91,007 | (30.8) | 14,891/44,957 | (33.1) | 1.01 [1.00, 1.02] | < 0.01 |
| Good CPC score (1 or 2), N (%) | 17,742 | (12.1) | 9,351 | (11.8) | 1.00 [0.99, 1.01] | 0.99 |
| Good CPC score* (1 or 2) in those that were discharged alive only, N (%) | 17,742/28,075 | (63.2) | 9,351/14,891 | (62.8) | 1.00 [0.99, 1.01] | 0.99 |
|
| ||||||
| Duration of Resuscitation Outcome |
Initial Rhythm PEA Total 147,377 (64.9%) |
Initial Rhythm Asystole Total 79,720 (35.1%) |
Adjusted Mean Difference [95% CI] | P Value | ||
|
| ||||||
| Overall duration of resuscitation in minutes, mean (SD) | 20.2 | (20.2) | 19.7 | (19.1) | −0.96 [−1.17, −0.75] | < 0.01 |
| Duration of resuscitation for those who did not achieve ROSC in minutes, mean (SD) | 29.8 | (22.5) | 26.2 | (21.5) | −3.02 [−3.33, −2.71] | < 0.01 |
Data on CPC score was missing for 5,915 (21.1%) of patients with an initial rhythm of PEA and 3,192 (21.4%) of patients with an initial rhythm of asystole that were discharged alive
Discussion
In this study, we found that although patients who suffered IHCA with an initial rhythm of PEA were more likely to achieve ROSC, they were equally likely to survive to hospital discharge when compared to patients with an initial rhythm of asystole. PEA arrests were more likely to occur in monitored areas. Duration of resuscitative efforts were longer for PEA arrests, both overall and when examining only those that did not achieve ROSC. Yet overall, the average duration of resuscitation shortened for both rhythms across each year of the study period. Rates of both ROSC and survival to discharge increased over time for both initial rhythms. The shorter resuscitations with improved outcomes likely reflect improvements in the management of cardiac arrest for PEA and asystole over time.
Few previous studies have examined PEA and asystole as distinct rhythms 3,7,15,16. Only one study by Høybye et al. provided detailed information on patient characteristics; they found IHCA with an initial rhythm of asystole to be more prevalent in the hospital wards than in the ICU, and less likely to be witnessed, as compared to IHCA with an initial rhythm of PEA, similar to our own findings. Three additional studies by Meaney et al., Nadkarni et al., and Nolan et al. also examined outcomes between PEA and asystole. These studies showed that ROSC was achieved more frequently when the initial rhythm was PEA than asystole, with which our findings were consistent. For survival to discharge, some studies found better odds after an initial rhythm of PEA 3,7,15, and one showed no difference in long-term survival between the two rhythms 16. Our investigation found no overall difference in survival to discharge between PEA and asystole. A possible explanation of these differing results for survival to discharge between investigators is the years included for the different studies. Both our investigation and the one by Høybye et al., which showed no difference in survival, included patients after the year 2006, while all of the studies that showed a difference in survival to discharge were performed with patients that arrested before the year 2006. Finally, although our study examined only IHCA, some studies have asked similar questions in the out-of-hospital cardiac arrest (OHCA) population 17,18. Similar to our investigation, these studies on OHCA found that the proportion of PEA arrests is increasing and that the survival of PEA arrests is increasing, although survival for patients with asystole has remained low 19.
PEA and asystole are frequently grouped as the “non-shockable” rhythms by cardiac arrest investigators 2,8–13,20–22, likely due to AHA guidelines placing PEA and asystole in the same management algorithm 6. However, our findings here show important differences between these two rhythms. Patients with PEA arrests were more likely to be on vasopressors, more likely to be on a ventilator, more likely to have a witnessed arrest, less likely to occur at night, more likely to occur at large hospitals and teaching hospitals, and had shorter time to delivery of epinephrine. We also found interesting results when evaluating the locations where these arrests occur. Initial PEA arrests were more likely to occur in the ICU than asystolic arrests, and asystolic arrests were more likely to occur on wards than PEA arrests. These location results are even more dramatic when splitting wards into telemetry and non-telemetry, with the proportion of asystolic arrests in non-telemetry wards being 8.2% higher than the proportion of PEA arrests in non-telemetry wards. All of these characteristics support PEA arrests being the initial rhythm in settings where there is increased monitoring (ICU or telemetry monitoring) or in settings where an arrest may be more anticipated (like for patients on vasopressors or ventilators). These findings may also reflect asystole being a common ‘terminal’ rhythm of arrests in unmonitored settings, where an arrest with an initial shockable or PEA rhythm degenerates to asystole prior to the arrest being recognized. Locations with better monitoring are able to identify arrests earlier, which likely results in more identification of PEA arrests, whereas areas with less monitoring identify arrests later and have a higher proportion of arrests where the initial PEA rhythm has devolved into asystole. We found increasing odds of PEA arrest during each subsequent year across our study period, as compared to asystole. The increasing odds of PEA may also be a result of improved monitoring which identifies arrests while there is still electrical activity, before the lack of circulation results in progression to asystole. Notably a similar trend is being recognized in the out-of-hospital arrest setting 19. The overall demographics of patients suffering arrest with initial rhythms of PEA and asystole are somewhat similar, and thus whether to group these patients together in clinical, research, and process improvement endeavors may need to be decided based on the problem being addressed.
Asystolic arrests had slightly lower rates of ROSC than PEA. Although statistically significant, the difference in ROSC was small, with only a 3% increased odds of ROSC for those with PEA. The difference in survival to discharge was even smaller, and was not statistically significant. These near-similar findings highlight that patients who arrest in the hospital with an initial rhythm of asystole do not have excessively poor rates of ROSC or survival. In fact, those with asystolic arrests had a ROSC rate of 56.4%, highlighting that this initial rhythm is still salvageable in most cases. Even for unwitnessed arrests where the initial rhythm is asystole, 6.6% survived to discharge with a favorable neurologic function; admittedly this is a low proportion, but we feel that aggressive resuscitation would still be warranted for this population. These findings of ROSC and survival for asystole are notable given that asystolic arrests in the out-of-hospital setting have a dismal survival rate 23, possibly owing to a longer duration before recognition and intervention when compared to in-hospital arrests.
For those who did not receive ROSC, resuscitations for asystolic arrests were over 3 minutes shorter (more than an entire cycle of CPR per the AHA cardiac arrest algorithm). Duration of resuscitation in patients who do not achieve ROSC is a recently described proxy for ‘intended’ duration of resuscitation in all comers 24. Shorter durations for those with asystole are very likely due to a perceived bias that patients with an initial rhythm of asystole are less likely to survive, or have a lower chance of being successfully resuscitated, an assumption that is not supported by our data, as we have shown similar rates of survival, survival with a favorable neurologic function, and only slightly lower rates of ROSC for asystolic arrests. Previous investigations have shown that patients resuscitated at hospitals with longer average resuscitation attempts have a higher rate of ROSC and survival to discharge, particularly when the arrest is due to PEA or asystole 24. These shorter resuscitations for patients with an initial rhythm of asystole are potentially modifiable practices, the improvement of which could result in increased rates of IHCA survival. Finally, when evaluating only those who obtained ROSC, survival to discharge was actually found to be better in the asystole group, although the difference was small, with 1% higher adjusted odds of survival to discharge. Again, this small difference may not be clinically meaningful, but with 290,000 in hospital cardiac arrests each year in the United States alone 1, even small improvements in survival can save thousands of lives.
Our investigation has several important strengths. This is a large study with over 200,000 patients, with data taken over a period greater than ten years, providing a detailed look at both descriptive statistics of the study population, as well as outcomes. This is also the first study we know that reports trends over time for the examined rhythms, providing new insight into trends in ROSC and survival, as well as resuscitation practices in the form of duration of resuscitation. This study has limitations. The database used does not have detailed information on the cause of the cardiac arrest or reason for admission. Our analysis is at risk of effect from resuscitation time bias25. Although this is a large database, the majority of participating hospitals have a large number of beds and are academic centers, limiting the generalizability of our results. Finally, the large number of arrests in this analysis is a strength of the study, but it also allows for the frequent finding of statistically significant but small differences, which as mentioned above may have limited clinical meaningfulness and must be interpreted in the right context.
Conclusion
In this large descriptive study, we found that patients suffering IHCA with an initial PEA rhythm had important differences regarding the location of arrest and resuscitation practices. Although arrests with an initial rhythm of asystole received shorter resuscitations, they had similar rates of survival to discharge as arrests with an initial rhythm of PEA. Overall, ROSC and survival to discharge have consistently improved for both rhythms over the years that were examined.
Supplementary Material
Supplemental Figure 1: PEA and asystole arrests over time
1a: Number of arrests by year of event
1b: Adjusted odds with 95% confidence intervals of initial rhythm being asystole as compared to PEA each year with respect to the year reference year 2006
Supplemental Figure 2: Favorable neurologic function (CPC score of 1 or 2) in those who survived to discharge for PEA and asystole
2a: Percentage favorable neurologic function (CPC score of 1 or 2) in those who survived to discharge for each initial rhythm by year
2b: Adjusted odds with 95% confidence intervals of favorable neurologic function (CPC score of 1 or 2) in those who survived to discharge each year with respect to the reference year 2006 for each initial rhythm
Supplemental Table 1: Missing data
Supplemental Table 2: Get With The Guidelines® quality metrics
Supplemental Table 3: Outcomes for patients with unwitnessed and witnessed arrests
Acknowledgements
American Heart Association’s Get With The Guidelines®-Resuscitation Adult Research Task Force members: Anne Grossestreuer PhD; Ari Moskowitz MD MPH; Joseph Ornato MD FACP FACC FACEP; Matthew Churpek MD MPH PhD; Monique Anderson Starks MD MHS; Paul Chan MD MSc; Saket Girotra MD SM; Sarah Perman MD MSCE.
Author Ari Moskowitz is funded by a grant from the National Institutes of Health/National Heart. Lung, and Blood Institute (R61 HL162980)
Abbreviation List
- AHA
American Heart Association
- CPR
cardiopulmonary resuscitation
- DOR
duration of resuscitation
- ECC
Emergency Cardiovascular Care
- ED
emergency department
- GWTG®-R
Get With The Guidelines®–Resuscitation
- ICU
intensive care unit
- IHCA
in-hospital cardiac arrest
- IQR
interquartile range
- IRB
institutional review board
- OHCA
out-of-hospital cardiac arrest
- OR
odds ratio
- PEA
pulseless electrical activity
- ROSC
return of spontaneous circulation
- SD
standard deviation
Footnotes
Conflicts of Interest
No conflicts of interest for any authors
Declaration of Interest
Declarations of interest: none
Notation
The abstract for this work was submitted to the American Thoracic Society 2022 International Conference, and was presented at the thematic poster session (C49 – Vascular and Hemodynamics).
The manuscript, related data, figures, and tables have not otherwise been published previously and the manuscript is not under consideration elsewhere.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Andersen LW, Holmberg MJ, Berg KM, Donnino MW, Granfeldt A. In-Hospital Cardiac Arrest: A Review. Jama. 2019;321(12):1200–1210. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Girotra S, Nallamothu BK, Spertus JA, Li Y, Krumholz HM, Chan PS. Trends in survival after in-hospital cardiac arrest. N Engl J Med. 2012;367(20):1912–1920. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Nadkarni VM, Larkin GL, Peberdy MA, et al. First documented rhythm and clinical outcome from in-hospital cardiac arrest among children and adults. Jama. 2006;295(1):50–57. [DOI] [PubMed] [Google Scholar]
- 4.Benjamin EJ, Virani SS, Callaway CW, et al. Heart Disease and Stroke Statistics—2018 Update: A Report From the American Heart Association. Circulation. 2018;137(12):e67–e492. [DOI] [PubMed] [Google Scholar]
- 5.Soar J, Nolan JP, Böttiger BW, et al. European Resuscitation Council Guidelines for Resuscitation 2015: Section 3. Adult advanced life support. Resuscitation. 2015;95:100–147. [DOI] [PubMed] [Google Scholar]
- 6.Panchal AR, Bartos JA, Cabañas JG, et al. Part 3: Adult Basic and Advanced Life Support: 2020 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation. 2020;142(16_suppl_2):S366–s468. [DOI] [PubMed] [Google Scholar]
- 7.Meaney PA, Nadkarni VM, Kern KB, Indik JH, Halperin HR, Berg RA. Rhythms and outcomes of adult in-hospital cardiac arrest. Crit Care Med. 2010;38(1):101–108. [DOI] [PubMed] [Google Scholar]
- 8.Sandroni C, Nolan J, Cavallaro F, Antonelli M. In-hospital cardiac arrest: incidence, prognosis and possible measures to improve survival. Intensive Care Med. 2007;33(2):237–245. [DOI] [PubMed] [Google Scholar]
- 9.Renkiewicz GK, Hubble MW, Wesley DR, et al. Probability of a shockable presenting rhythm as a function of EMS response time. Prehosp Emerg Care. 2014;18(2):224–230. [DOI] [PubMed] [Google Scholar]
- 10.Holmberg M, Holmberg S, Herlitz J. Incidence, duration and survival of ventricular fibrillation in out-of-hospital cardiac arrest patients in sweden. Resuscitation. 2000;44(1):7–17. [DOI] [PubMed] [Google Scholar]
- 11.Granfeldt A, Wissenberg M, Hansen SM, et al. Clinical predictors of shockable versus non-shockable rhythms in patients with out-of-hospital cardiac arrest. Resuscitation. 2016;108:40–47. [DOI] [PubMed] [Google Scholar]
- 12.Stankovic N, Høybye M, Holmberg MJ, Lauridsen KG, Andersen LW, Granfeldt A. Factors associated with shockable versus non-shockable rhythms in patients with in-hospital cardiac arrest. Resuscitation. 2021;158:166–174. [DOI] [PubMed] [Google Scholar]
- 13.Andersen LW, Lind PC, Vammen L, Høybye M, Holmberg MJ, Granfeldt A. Adult post-cardiac arrest interventions: An overview of randomized clinical trials. Resuscitation. 2020;147:1–11. [DOI] [PubMed] [Google Scholar]
- 14.Peberdy MA, Kaye W, Ornato JP, et al. Cardiopulmonary resuscitation of adults in the hospital: a report of 14720 cardiac arrests from the National Registry of Cardiopulmonary Resuscitation. Resuscitation. 2003;58(3):297–308. [DOI] [PubMed] [Google Scholar]
- 15.Nolan JP, Soar J, Smith GB, et al. Incidence and outcome of in-hospital cardiac arrest in the United Kingdom National Cardiac Arrest Audit. Resuscitation. 2014;85(8):987–992. [DOI] [PubMed] [Google Scholar]
- 16.Høybye M, Stankovic N, Lauridsen KG, Holmberg MJ, Andersen LW, Granfeldt A. Pulseless electrical activity vs. asystole in adult in-hospital cardiac arrest: Predictors and outcomes. Resuscitation. 2021;165:50–57. [DOI] [PubMed] [Google Scholar]
- 17.Cobb LA, Fahrenbruch CE, Olsufka M, Copass MK. Changing incidence of out-of-hospital ventricular fibrillation, 1980–2000. Jama. 2002;288(23):3008–3013. [DOI] [PubMed] [Google Scholar]
- 18.Wenzel V, Krismer AC, Arntz HR, Sitter H, Stadlbauer KH, Lindner KH. A comparison of vasopressin and epinephrine for out-of-hospital cardiopulmonary resuscitation. N Engl J Med. 2004;350(2):105–113. [DOI] [PubMed] [Google Scholar]
- 19.Bergström M, Schmidbauer S, Herlitz J, Rawshani A, Friberg H. Pulseless electrical activity is associated with improved survival in out-of-hospital cardiac arrest with initial non-shockable rhythm. Resuscitation. 2018;133:147–152. [DOI] [PubMed] [Google Scholar]
- 20.Desch S, Freund A, Akin I, et al. Angiography after Out-of-Hospital Cardiac Arrest without ST-Segment Elevation. N Engl J Med. 2021;385(27):2544–2553. [DOI] [PubMed] [Google Scholar]
- 21.Nielsen N, Wetterslev J, Cronberg T, et al. Targeted temperature management at 33°C versus 36°C after cardiac arrest. N Engl J Med. 2013;369(23):2197–2206. [DOI] [PubMed] [Google Scholar]
- 22.Dankiewicz J, Cronberg T, Lilja G, et al. Hypothermia versus Normothermia after Out-of-Hospital Cardiac Arrest. N Engl J Med. 2021;384(24):2283–2294. [DOI] [PubMed] [Google Scholar]
- 23.Andrew E, Nehme Z, Lijovic M, Bernard S, Smith K. Outcomes following out-of-hospital cardiac arrest with an initial cardiac rhythm of asystole or pulseless electrical activity in Victoria, Australia. Resuscitation. 2014;85(11):1633–1639. [DOI] [PubMed] [Google Scholar]
- 24.Goldberger ZD, Chan PS, Berg RA, et al. Duration of resuscitation efforts and survival after in-hospital cardiac arrest: an observational study. Lancet. 2012;380(9852):1473–1481. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Andersen LW, Grossestreuer AV, Donnino MW. “Resuscitation time bias”-A unique challenge for observational cardiac arrest research. Resuscitation. 2018;125:79–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental Figure 1: PEA and asystole arrests over time
1a: Number of arrests by year of event
1b: Adjusted odds with 95% confidence intervals of initial rhythm being asystole as compared to PEA each year with respect to the year reference year 2006
Supplemental Figure 2: Favorable neurologic function (CPC score of 1 or 2) in those who survived to discharge for PEA and asystole
2a: Percentage favorable neurologic function (CPC score of 1 or 2) in those who survived to discharge for each initial rhythm by year
2b: Adjusted odds with 95% confidence intervals of favorable neurologic function (CPC score of 1 or 2) in those who survived to discharge each year with respect to the reference year 2006 for each initial rhythm
Supplemental Table 1: Missing data
Supplemental Table 2: Get With The Guidelines® quality metrics
Supplemental Table 3: Outcomes for patients with unwitnessed and witnessed arrests