a b s t r a c t

Aim: Intracranial hemorrhage is an important cause of out-of-hospital cardiac arrest (OHCA), yet there are no United States (US), European, or Australian prospective studies examining its incidence in patients who sustained OHCA. This study aims to identify the incidence of ICH in US patients with OHCA who obtain return of spontaneous circulation (ROSC).

Methods: We prospectively analyzed consecutive patients with OHCA who achieved ROSC at a single US hospital over a 15-month period. Before beginning patient enrollment, we implemented a standardized emergency de- partment order-set for the initial management for all patients with ROSC after OHCA. This order-set included a non-contrast head computed-tomography (NCH-CT) scan. Patient and cardiac arrest variables were recorded, as were NCH-CT findings.

Results: During the study period, 85 patients sustained an OHCA, achieved ROSC, survived to hospital admission,

and underwent a NCH-CT. Three of these 85 patients had ICH (3.5%). Survival to discharge was seen in 23/82 (28.0%) patients without ICH and in 1/3 patients with ICH. Survival with Good neurologic outcome was seen in 14/82 (17.1%) patients without ICH and in 0/3 patients with ICH. Patients with ICH tended to be older than pa- tients without ICH.

Conclusions: In our US cohort, ICH was an uncommon finding in patients who sustained OHCA and survived to hospital admission, and no patients with ICH survived to discharge with good neurologic outcome. Additionally, the incidence of ICH was lower than reported in previous studies.

(C) 2020

1. Introduction

Over 300,000 Americans sustain an out of hospital cardiac arrest (OHCA) each year [1]. Guidelines from the American Heart Association emphasize the importance of identifying the primary etiology that re- sulted in the cardiac arrest [2]. Intracranial hemorrhage , including subarachnoid hemorrhage , is an important non-cardiac cause of cardiac arrest that can be rapidly identified in the Emergency Depart- ment (ED) through the use of non-contrast head computed tomography (NCH-CT). Cardiac arrest in these patients is believed to be related to a sudden increase in intracranial pressure resulting in a respiratory arrest secondary to brainstem herniation. Alternatively, cardiac arrest may be a result of ICH-induced Catecholamine release that result in non- cardiogenic pulmonary edema and/or malignant ventricular dysrhyth- mia [1-5].

International data from retrospective trials report a prevalence of ICH between 5% and 24% in patients with OHCA [6,7]. This wide range

* Corresponding author at: Department of Emergency Medicine, QIC #22123, Highland General Hospital, Alameda Health System, 1411 East 31st Street, Oakland, CA 94602 USA

E-mail address: [email protected] (J. Gelber).

is most likely explained by patient selection and variations in the human population due to underlying genetic differences, as well as sev- eral intervenable risk factors [8]. Only one single prospective study, completed in a primarily Japanese population, evaluated the prevalence of ICH in OHCA [9]. This 2009 trial by Inamasu et al. found that 18.3% of patients with ROSC after OHCA had a concomitant ICH, of which the ma- jority were SAH. This dataset, however, may not be generalizable to non-Japanese populations, as the incidences of cerebellar hemorrhage, SAH, and total ICH in Japan are more than double those of European countries [8,10].

In this study, we aim to prospectively identify the rate of ICH identi- fied by NCH-CT in patients with OHCA who obtained ROSC and survived to hospital admission. To enable better recognition of ICH as the cause of cardiac arrest, we recorded the clinical presentation and arrest charac- teristics of patients with ICH and compared them to the characteristics of patients without ICH.

1. Methods

We performed a prospective study on all patients with ROSC after OHCA admitted to a single ED during a 15-month period of time,

https://doi.org/10.1016/j.ajem.2020.12.015

0735-6757/(C) 2020

between December 1st, 2018 and February 29th, 2020. The study was completed in the ED of a regional Cardiac Arrest Receiving Center located in Oakland, California. Data from patients admitted to the ED after cardiac arrest were recorded according to the Utstein protocol [11]. The follow- ing data were documented on arrival: sex and age of the patient; witnessed or unwitnessed cardiac arrest; presence or absence of by- stander CPR, ROSC prior to arrival in the ED, and initial electrocardiogram (ECG) rhythm observed by any rescue worker. This study was approved by the hospital’s Institutional Review Board (IRB) prior to patient enroll- ment. Per IRB recommendations, this study met criteria for minimal risk to patients, thus there was no consent for patient enrollment.

Prior to July 2018, OHCA patients who achieved ROSC in our ED re- ceived head imaging according to physician clinical judgment. In July 2018, a “post-ROSC” order set was added to the ED workflow, standard- izing aspects of the management of post-cardiac arrest patients. This order set included a NCH-CT scan that was automatically ordered unless the medical provider cancelled the order (i.e. opt-out strategy as op- posed to an opt-in strategy). Between July 2018 and December 2018, all ED providers were recommended to use the “post-ROSC” order-set through didactic sessions at weekly educational conferences, routine emails, and in-person reminders. The delay between implementation of the “post-ROSC” order-set and the start of study enrollment was deliberate, in order to maximize provider utilization of the order-set, and to increase the percentage of OHCA patients who received a NCH- CT scan.

Each month, two of the authors (JG, MEM) manually reviewed the electronic medical record of all OHCA cases, utilizing a standardized data collection form. Patients were excluded from further analysis if they were < 18 years old, had a Traumatic cardiac arrest, or if resuscita- tion efforts were stopped because of identified “Do Not Resuscitate” sta- tus. Additionally, patients were excluded from analysis if they did not survive to the point of hospital admission from the ED. Presence or absence of ICH was determined through real-time interpretation by at- tending radiologists as entered into the patient’s clinical chart. Survival to discharge was determined through chart review. “Good neurologic outcome” was defined as a modified Rankin Score less than or equal to three in accordance with previous studies [1]. Data normality was assessed through the Shapiro-Wilk test. Data with a normal distribution is reported as mean with standard deviation. Data with a non-normal distribution is reported as median with interquartile range.

1. Results

During the 15-month enrollment period, 194 patients presented to the ED after OHCA. Of these patients, 95 (49.0%) achieved ROSC to ad- mission, and 85/95 received a NCH-CT (89.5%) (Fig. 1). Baseline and re- suscitation characteristics of patients with OHCA who survived to hospital admission are reported (Table 1). ICH was present in 3 of 85 (3.5%) OHCA patients who completed NCH-CT.

Twenty-eight of 85 patients (32.9%) had an initial “shockable” rhythm of Ventricular Fibrillation (VF) or pulseless Ventricular Tachy- cardia (pVT), including 1 of 3 patients with ICH. Of the 85 total patients, 24 survived to discharge (28.2%), including 1 of the 3 patients with ICH. Fourteen of 82 patients without ICH survived to discharge neurologi- cally intact (17.1%), while none of the 3 patients with ICH survived to discharge Neurologically intact. Patients with ICH were older than were patients without ICH (86.7 years vs. 67 years). The 82 patients without ICH were compared with the 3 patients with ICH (Table 2).

The 85 patients who achieved ROSC and completed NCH-CT were subsequently compared to the 10 patients who achieved ROSC and did not receive a NCH-CT scan, in order to evaluate for any intergroup differences (Table 3). Patients who did not receive a NCH-CT were more likely to have had an Initial shockable rhythm (6/10 or 60.0%) as compared to patients who received a NCH-CT (28/85 or 32.9%). Notably, patients who did not receive a NCH-CT were more likely to be taken

Fig. 1. Study enrollment flow. Exclusion criteria were age < 18, primary traumatic etiology of OHCA, No code or limited code due to limited goals of care. Abbreviations: OHCA: out-of- hospital cardiac arrest; ROSC: return of spontaneous circulation; CT: computed tomography; ICH: Intracranial Hemorrhage.

Table 1

Baseline and resuscitation characteristics of OHCA patients who survived to hospital admission.

Variables

Age (median with interquartile range) (n = 95) 67 years, IQR = 16 Gender (n = 95)

Male 59 patients (62.1%)

Female 36 patients (37.9%)

Witnessed OHCA? (n = 95) 65 patients (68.4%)

Bystander CPR Performed? (n = 95) 49 patients (51.6%) Initial Rhythm (n = 95)

Shockable (VT or VFib) 34 patients (35.8%)

Non-shockable (Asystole or PEA) 61 patients (64.2%) CT Head with ICH? (n = 85) 3 patients (3.5%)

Survived to Discharge? (n = 95) 27 patients (28.4%)

Neurologically intact survival? (n = 95) 17 patients (17.9%)

immediately to the catheterization-lab (6/10 or 60.0%) than were the patients who received a NCH-CT (11/85 patients or 12.9%).

1. Discussion

In this prospective study, a relatively small percentage of patients with OHCA had ICH (3.5%). Retrospective studies in the US and Europe have reported a higher incidence of ICH in patients with OHCA, between 5% to 24% [6,7]. However these retrospective studies suffer from signif- icant referral bias, as they only examined patients with OHCA who were suspected of having ICH. In a review of retrospective studies, NCH-CT was performed on 17-61% of study subjects overall, with most studies examining fewer than one third of all eligible patients [6,7,12]. These studies did not examine a large percentage of their study population, making the incidence of ICH in OHCA challenging to extrapolate.

In this trial, NCH-CT was performed in the vast majority (89.5%) of subjects. We believe that our data more closely approximates the true incidence of ICH in OHCA, as compared to retrospective studies. More- over, the subset of patients in our study who did not undergo NCH-CT had similar baseline characteristics compared to subjects who did

Table 2

Comparison of OHCA patients with and without ICH.

	Total	ICH present	ICH absent
n	85	3	82
Gender	52 M, 33F	1M, 2F	51M, 31F
Agea (years with SD/IQR)	68 (median)	86.7 (mean)	67 (median)
	IQR = 15.5	SD = 1.2	IQR = 15.5
ROSC Prior to Arrival?	42.4% (36/85)	33.3% (1/3)	42.7% (35/82)
Shockable Initial Rhythm?	32.9% (28/85)	33.3% (1/3)	32.9% (27/82)
witnessed arrest?	69.4% (59/85)	100.0% (3/3)	68.3% (56/82)
Bystander CPR?	51.8% (44/85)	66.7% (2/3)	51.2% (42/82)
Survival to Discharge?	28.2% (24/85)	33.3% (1/3)	28.0% (23/82)
Neuro intact Survival?	16.5% (14/85)	0% (0/3)	17.1% (14/82)

Abbreviations: M: male; F: female; SD: standard deviation; IQR: Interquartile range; ICH: intracranial hemorrhage.

^a For groups with normal distribution of age (“ICH present”), mean +- SD is listed. For groups with non-normal distribution of age (“Total” and “ICH absent”), median +- IQR is listed.

Table 3

Comparison of OHCA patients with and without NCH-CT scan.

Total CT done No CT done

n Gender Agea (years with SD/IQR)	95 59 M, 36F 67 (median) IQR = 16	85 52 M, 33F 68 (median) IQR = 15.5	10 7M, 3F 63.3 (mean) SD = 15.6
ROSC Prior to arrival?	41.1% (39/95)	42.4% (36/85)	30.0% (3/10)
Shockable initial rhythm?	35.8% (34/95)	32.9% (28/85)	60.0% (6/10)
Witnessed arrest?	68.4% (65/95)	69.4% (59/85)	60.0% (6/10)
Bystander CPR?	51.6% (49/95)	51.8% (44/85)	50.0% (5/10)
Went immediately to Cath Lab?	17.9% (17/95)	12.9% (11/85)	60.0% (6/10)
Survival to discharge?	28.4% (27/95)	28.2% (24/85)	30.0% (3/10)
Neuro intact survival?	17.9% (17/95)	16.5% (14/85)	30.0% (3/10)

Abbreviations: M: male; F: female; SD: standard deviation; IQR: Interquartile range; Cath: Catheterization.

^a For groups with non-normal distribution of age (“Total” and “CT done”), median +- IQR is listed. For groups with normal distribution of age (“No CT done”), mean +- SD is listed.

undergo NCH-CT (Table 3), with similar baseline characteristics of sex, age, initial rhythm, survival to discharge, and neurologically intact sur- vival. The majority of patients (6/10) who did not receive a CT scan went emergently to the catheterization lab, suggesting a primary car- diac etiology rather than ICH.

In our study, OHCA patients with ICH were older than OHCA patients without ICH. Advanced age may be an important clue in determining the etiology of OHCA in patients with unclear presentations. In fact, pre- vious literature has speculated that at least part of the explanation for Japan’s high rate of ICH is its relatively elderly population [8].

Our study reports a lower incidence of ICH as compared to the pro- spective Inamasu et al. 2009 study from Japan, which found ICH in 18.3% of patients with OHCA [9]. As previously discussed, ICH is more frequent in Japan than in other countries [10,13]. There is no consensus explanation for the higher rate of ICH in Japan, but proposed theories

have included an older population, higher smoking rates, longer work- ing hours, and lower total cholesterol [8,10].

Previous literature suggest that patients with ICH-associated OHCA are more likely to have non-shockable initial cardiac rhythms (PEA or asystole) as compared to non-ICH associated OHCA [3]. In our study, however, the frequency of shockable initial rhythm was similar in all pa- tients, regardless of ICH. Nevertheless, the low incidence of ICH in our study limits our ability to infer any association between ICH and initial cardiac rhythm.

1. Conclusions

This pilot study is the first prospective study in the USA, Australia, or Europe to examine the incidence of ICH in OHCA. The incidence of ICH in patients with OHCA who survived to hospital admission was 3.5%, lower than previously reported in retrospective studies, and much lower than noted in prospective Japanese studies.

Declaration of Competing Interest

None.

References

1. Mozaffarian D, Benjamin EJ, Go AS, Arnett DK, Blaha MJ, Cushman M, et al. Heart dis- ease and stroke statistics-2015 update : A report from the American Heart Associa- tion, vol. 131; 2015.
2. Callaway CW, Donnino MW, Fink EL, Geocadin RG, Golan E, Kern KB, et al. 2015 American Heart Association guidelines update for cardiopulmonary resuscitation and emergency cardiovascular care, part 8: post-cardiac arrest care. Circulation. 2015;132:s465-82.
3. Mitsuma W, Ito M, Kodama M, Takano H, Tomita M, Saito N, et al. Clinical and car- diac features of patients with subarachnoid haemorrhage presenting with out-of- hospital cardiac arrest. Resuscitation. 2011;82:1294-7.
4. Samuels MA. The brain-heart connection. Circulation. 2007;116:77-84.
5. Estanol BV, Marin OSM. Cardiac arrhythmias and sudden death in subarachnoid hemorrhage. Stroke. 1975;6:382-6.
6. Chelly J, Mongardon N, Dumas F, Varenne O, Spaulding C, Vignaux O, et al. Benefit of an early and systematic imaging procedure after cardiac arrest: insights from the PROCAT (Parisian region out of hospital cardiac arrest) registry. Resuscitation. 2012;83:1444-50.
7. Ruffino A, Gautam S, Patel K, Collins C, Chatterjee M, Majzun M. Prevalence of intra- cranial hemorrhage in outside hospital cardiac arrest patients. Chest. 2017;152: A326.
8. An SJ, Kim TJ, Yoon BW. Epidemiology, risk factors, and clinical features of intracere- bral hemorrhage: An update. J Stroke. 2017;19:3-10.
9. Inamasu J, Miyatake S, Tomioka H, Suzuki M, Nakatsukasa M, Maeda N, et al. Sub- arachnoid haemorrhage as a cause of out-of-hospital cardiac arrest: a prospective computed tomography study. Resuscitation. 2009;80:977-80.
10. Suzuki K, Izumi M. The incidence of hemorrhagic stroke in Japan is twice compared with western countries: the Akita Stroke registry. Neurol Sci. 2014;36:155-60.
11. Langhelle A, Nolan J, Herlitz J, Castren M, Wenzel V, Soreide E, et al. Recommended guidelines for reviewing, reporting, and conducting research on post-resuscitation care: the Utstein style. Resuscitation. 2005;66:271-83.
12. Shin J, Kim K, Lim YS, Lee HJ, Lee SJ, Jung E, et al. Incidence and clinical features of intracranial hemorrhage causing out-of-hospital cardiac arrest: a multicenter retro- spective study. Am J Emerg Med. 2016;34:2326-30.
13. De Rooij NK, Linn FHH, Van Der Plas JA, Algra A, Rinkel GJE. Incidence of subarach- noid haemorrhage: a systematic review with emphasis on region, age, gender and time trends. J Neurol Neurosurg Psychiatry. 2007;78:1365-72.