Hrev_master Abstract Idiopathic Out-Of-Hospital Cardiac Arrest (OHCA) requires urgent treatment. Early Computed Tomography (CT) imaging may be useful to aid diagnosis. We aimed to determine current CT imaging practice, safety, and diagnostic value in this patient popu- lation. This study was a single-centre, retrospective cohort study of patients presenting to the Emergency Department (ED) of the Royal Infirmary of Edinburgh with idiopathic non-traumatic OHCA and Return Of Spontaneous Circulation (ROSC). Between 1st January 2016 and 31st December 2019, 140 of 156 (90%) eligi- ble patients underwent 195 CT scans identifying the cause of OHCA in 6 (4%). CT head diagnosed one ischaemic and three haemorrhagic strokes, and CT pulmonary angiogram diagnosed one acute coronary syndrome and one pulmonary embolism. CT head (134), CT pulmonary angiogram (25) and CT cervical spine (16) were the commonest scans. 68 of 195 (35%) CT scans showed important pathology, mostly secondary to OHCA. CT imaging was safe with no cases of contrast nephropathy, allergic reaction, or other complications. The diagnostic value of CT imaging in this patient population was limited. However, imaging was a valuable method of identifying other important secondary pathology. Introduction Out-Of-Hospital Cardiac Arrest (OHCA) is the third leading cause of death in Europe.1 In Scotland the resuscitation of around 3000 patients with OHCA is attempted annually.2 Survival is slow- ly increasing but remains low at around 10% of cases where resus- citation was attempted.3 For patients resuscitated without an obvious cause (idiopathic OHCA), it is often difficult to identify the underlying pathology due to variety of diagnoses, lack of medical history, and intubation preventing communication.4 Urgent treatment necessitates early diagnosis. Current European Resuscitation Council (ERC) and European Society of Intensive Care Medicine guidelines for post- resuscitation care recommend investigations including an electro- cardiogram (ECG), chest radiograph, coronary angiography +/- percutaneous coronary intervention (PCI), and computed tomogra- phy (CT) brain scan or CT pulmonary angiogram (CTPA). Whole- body CT (WBCT) is currently reserved for OHCA with trauma or haemorrhage.5 Despite this, investigations fail to identify the cause in 13% of patients.6 CT imaging is becoming more common to determine the arrest cause and evaluate resuscitation complications.7-9 A recent retro- spective study found early WBCT imaging to be feasible and capa- ble of identifying acute pathology relevant to OHCA aetiology and resuscitation.10 Subsequently, a prospective study reported that early head-to-pelvis CT scanning identified time-critical diagnoses within 2 hours, directly identified 13% of OHCA causes and result- ed in significant, appropriate changes in management.11 This single-centre, retrospective cohort study was designed to determine current CT imaging practice in non-traumatic idiopathic Emergency Care Journal 2022; volume 18:10669 Correspondence: Isla Harper, Edinburgh Medical School, College of Medicine and Veterinary Medicine, 49 Little France Crescent, Edinburgh, EH16 4SB, UK. E-mail: islaharper@gmail.com Key words: Cardiac arrest; computed tomography; emergency department. Contributions: All authors were involved in the conception of the study, study design, analysis and interpretation of data and drafting the article. IH collected the data. Funding: MJR is supported by an NHS Research Scotland Career Researcher Clinician award. Conflict of interest: The authors declare no conflict of interest. Availability of data and materials: All data generated or analyzed during this study are included in this published article. Ethics approval and consent to participate: The study was deemed by the University of Edinburgh Ethics Committee to be a service evalua- tion survey and did not therefore require formal ethical approval. All patient data were anonymised prior to release from the NHS. Informed consent: Not applicable. Received for publication: 12 June 2022. Accepted for publication: 13 September 2022. This work is licensed under a Creative Commons Attribution 4.0 License (by-nc 4.0). ©Copyright: the Author(s), 2022 Licensee PAGEPress, Italy Emergency Care Journal 2022; 18:10669 doi:10.4081/ecj.2022.10669 Publisher's note: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organiza- tions, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its man- ufacturer is not guaranteed or endorsed by the publisher. [Emergency Care Journal 2022; 18:10669] [page 41] CT imaging in idiopathic out-of-hospital cardiac arrest: An assessment of current practice and diagnostic utility Isla Harper,1 Kate Easterford,2 Matthew J Reed2,3 1Edinburgh Medical School, College of Medical and Veterinary Medicine, Edinburgh; 2Emergency Medicine Research Group Edinburgh (EMERGE), Department of Emergency Medicine, Royal Infirmary of Edinburgh, Edinburgh; 3Acute Care Group, Usher Institute of Population Health Sciences and Informatics, College of Medicine and Veterinary Medicine, University of Edinburgh, Edinburgh, UK Non -co mmerc ial us e o nly OHCA in the Emergency Department (ED) of the Royal Infirmary of Edinburgh (RIE) and evaluate safety and diagnostic utility of CT imaging in identifying the aetiology of OHCA. Materials and Methods Study design, setting, and objectives This was a single-centre, retrospective cohort study. It was car- ried out at the Emergency Department (ED) of the Royal Infirmary of Edinburgh (RIE), UK. Our primary objective was to determine the diagnostic utility of CT imaging in identifying the cause of non-traumatic OHCA in adults attending the ED of the RIE, while our secondary objectives were: i) To determine current CT imaging practice in non-traumat- ic OHCA in the ED of the RIE; ii) To investigate whether CT imag- ing leads to changes in patient management; iii) To determine char- acteristics of patients who benefited from CT imaging; iv) To develop a decision-making tool to assist clinicians in determining which non-traumatic OHCA patients should be considered for CT imaging; v) To investigate CT safety including the incidence of contrast nephropathy, allergic reactions or other CT scanning com- plications, and the incidence of inappropriate further investigation or treatments. The project was deemed a service evaluation and registered with the Lothian eQuIPED Registry. It did not require formal eth- ical review. Participants Patients with a presenting diagnosis of OHCA with Return Of Spontaneous Circulation (ROSC), who presented to the ED at the RIE between 1st January 2016 and 31st December 2019 were eli- gible. The number of these patients determined sample size. Inclusion criteria were: i) Diagnosis of non-traumatic OHCA with ROSC, ii) Attended the ED of the RIE and transferred to the Intensive Care Unit, iii) >14 years of age, iv) GCS ≤8, v) no objec- tive ECG or point-of-care imaging OHCA cause and 6) clinically stable to undergo CT imaging. Exclusion criteria were: i) acute ST-elevation myocardial infarction (STEMI) or new left bundle branch block (LBBB) on ECG, ii) further investigation declined in documented Anticipatory Care Plan and iii) traumatic cause of OHCA. Data collection The RIE ED TrakCare (InterSystems) Electronic Patient Record (EPR) system containing all patients attending the RIE ED was interrogated by a data analyst to retrieve patients attending with OHCA for the past 5 years who were subsequently transferred to the Intensive Care Unit (ICU). Patients were identified by a unique study number and record- ed in the database. The study database was interrogated by a single researcher and potentially eligible patients were cross-referenced with their EPR medical notes by CHI or UHPI number (held and linked through the separate RIE ED TrakCare database) and inclu- sion and exclusion criteria applied. Reason for exclusion was doc- umented for excluded patients. Patients were not identifiable with- out access to the RIE ED TrakCare database with study number linkage. Definitions were created and applied for past medical history diagnoses. Heart disease included previous myocardial infarction, atrial fibrillation, ischaemic heart disease, valve disease, congeni- tal heart disease, cardiomyopathy, previous PCI or stenting. Chronic respiratory disease included chronic obstructive pul- monary disease, emphysema, and asthma. Chronic kidney disease was recorded. Chronic liver disease included alcoholic liver dis- ease, non-alcoholic fatty liver disease or chronic hepatitis B/C infection. Previous CerebroVascular Accident included previous stroke or transient ischaemic event. Diabetes mellitus included Type I and Type II diabetes mellitus, gestational diabetes, and maturity-onset diabetes of the young. Any previous or current malignancy was noted. Epilepsy or seizure history was recorded. Missing data was recorded as such. OHCA event data included presence of bystander CPR, ambu- lance response time, initial arrest rhythm on arrival of emergency services, total downtime, rearrest, prehospital or ED intubation, and vital signs on ED presentation. CT scans were included if they were ordered in the ED. Scans included CT head, CT cervical spine, CT thorax, CT abdomen, CT pelvis, CT chest/abdomen/pelvis, CT intracranial angiogram, CT pulmonary angiogram, CT thoracic aorta and whole-body CT. The time from admission to scan order time was recorded. CT imaging results were recorded, including the potential OHCA cause and any other important diagnoses that were identi- fied. The cause of OHCA was taken as documented by the ICU clinician or detailed on the death certificate. Outcome data included whether the patient was discharged from hospital or died, the hospital and ICU length of stay and safe- ty endpoints. CT scans were said to have changed management if they identified the OHCA cause or identified other important diag- noses such as complications of resuscitation or arrest. Study outcomes Our primary endpoint was to determine the diagnostic utility of CT imaging in identifying the OHCA aetiology. Our secondary Article Figure 1. STROBE Diagram of patient recruitment. RIE, Royal Infirmary of Edinburgh; ED, Emergency Department; OHCA, out-of-hospital cardiac arrest; ICU, Intensive Care Unit; GCS, Glasgow Coma Score; STEMI, ST-elevation myocardial infarc- tion; ECG, electrocardiogram; LBBB, left bundle branch block; NSTEMI, non-ST-elevation myocardial infarction; CT, computed tomography. *’CT head’ includes 134 CT head scans and 1 CT intracranial angiogram. ‘CT chest/abdomen/pelvis’ includes 7 CT chest/abdomen/pelvis scans, 1 CT chest scan, 2 CT abdomen scans and 2 CT pelvis scans. [page 42] [Emergency Care Journal 2022; 18:10669] Non -co mmerc ial us e o nly endpoints were to determine: i) Incidence of CT imaging leading to changes in patient management; ii) Incidence of contrast nephropathy; iii) Incidence of allergic reactions or other complica- tions of CT scanning; iv) Incidence of CT scanning resulting in inappropriate further investigation or treatment. Data analysis Data was entered into a database and analysed in Microsoft Excel. Unless otherwise stated, data are presented as median with Interquartile Range (IQR) (25th–75th percentile) for nonparamet- ric continuous variables and as simple frequencies, proportions, and percentages for categorical variables. Parametric continuous variables are presented as mean with SD. Statistical analysis The characteristics of patients who benefited from CT imaging was compared to those that did not using either Chi-Squared test with Yates correction for continuity, or Fisher Exact using two- tailed P-value, with significance set at the 5% level. Results Between 1st January 2016 and 31st December 2019, 288 patients presented to the RIE ED with ROSC after non-traumatic OHCA. 132 patients were excluded and 156 included. 140/156 (90%) received CT imaging and 16/156 (10%) did not (Figure 1). Mean age was 60 years, and 45/156 (29%) were female (Table 1). 42/156 (27%) patients experienced a documented prodrome before collapse (Table 2). The most common arrest rhythm was VF (76/156; 51%) with a median downtime of 23 (IQR 15-40) min- utes. 50/156 (32%) patients rearrested. 120/156 (78%) had bystander CPR. Median ambulance response time was 5 (IQR 5- Article Table 1. Demographics and Past Medical History of study patients. Demographics and Past All patients Patients that received Patients with a Patients with a P value Medical History (n=156) a CT scan positive CT scan* negative CT (2 decimal places) n (%) (n=140) (n=58) scan† (n=82) or mean ± SD n (%) or mean ± SD n (%) or mean ± SD n (%) or mean ± SD Age (years) 60 ± 17 60 ± 16 58 ± 18 62 ± 15 0.14 Female sex 45 (29) 39 (28) 18 (31) 21 (26) 0.61 Medical history‡ Heart disease 55 (37) 49 (37) 19 (35) 30 (38) 0.78 Hypertension 36 (24) 32 (24) 9 (17) 23 (29) 0.12 Chronic respiratory disease 19 (13) 18 (14) 7 (13) 11 (14) 1.00 Chronic kidney disease 11 (8) 8 (6) 2 (4) 6 (8) 0.47 Chronic liver disease 5 (3) 5 (4) 4 (7) 1 (1) 0.16 Cerebrovascular accident 14 (10) 12 (9) 3 (6) 9 (12) 0.36 Deep vein thrombosis 4 (3) 4 (3) 2 (4) 2 (3) 1.00 Diabetes 15 (10) 11 (8) 4 (7) 7 (9) 0.76 Malignancy 15 (10) 14 (11) 8 (15) 6 (8) 0.26 Seizures 9 (6) 9 (7) 3 (6) 6 (8) 0.74 * ‘Positive’ CT scan: identified OHCA cause or other important pathology. †‘Negative’ CT scan: no pathology. ‡148 (95%) patients had complete past medical history information. Table 2. Symptoms reported in study patients prior to collapse. Prodrome Before Collapse All patients Patients that Patients with a Patients with a P value (n=156) received a CT scan positive CT scan* negative CT scan† (2 decimal places) n (%) (n=140) (n=58) (n=82) n (%) n (%) n (%) Prodrome 42 (27) 33 (24) 14 (24) 19 (23) 1.00 Chest pain‡ 6 (14) 5 (15) 2 (14) 3 (16) 1.00 Palpitations 1 (2) 0 0 0 - Dyspnoea 20 (48) 15 (45) 7 (50) 8 (42) 0.86 Abdominal pain 2 (5) 2 (6) 1 (7) 1 (5) 1.00 Back/flank pain 1 (2) 1 (3) 0 1 (5) 1.00 Jaw pain 1 (2) 0 0 0 - Nausea/vomiting 5 (12) 5 (15) 3 (21) 2 (11) 0.65 Headache 1 (2) 1 (3) 1 (7) 0 0.41 Confusion 3 (7) 2 (6) 1 (7) 1 (5) 1.00 Dizziness 2 (5) 2 (6) 0 2 (11) 0.51 Drowsiness 3 (7) 2 (6) 1 (7) 1 (5) 1.00 Seizure 5 (12) 4 (12) 1 (7) 3 (16) 0.64 No prodrome 52 (33) 45 (32) 20 (34) 25 (30) Information not available 62 (40) 62 (44) 24 (41) 38 (46) *‘Positive’ CT scan: identified OHCA cause or other important pathology. †‘Negative’ CT scan: no pathology. ‡Symptom percentages from patients with documented prodromes. Percentages total >100% as patients could have multiple symptoms. [Emergency Care Journal 2022; 18:10669] [page 43] Non -co mmerc ial us e o nly 11) minutes. 51/156 (33%) were intubated pre-hospital and 105/156 (67%) were intubated in the ED (Table 3). Of the 140 patients who received imaging, scans identified arrest cause in 6/140 (4%). CT head scans diagnosed 3 haemorrhagic strokes and one ischaemic stroke. One acute coronary syndrome and one pul- monary embolism were identified by CT pulmonary angiogram (Table 4). The objective as to whether imaging changed management was assessed by the number of patients who received scans that identi- fied the OHCA cause, or other important pathology (‘positive’ scan) compared to no CT findings (‘negative’ scan). 58/140 (41%) patients had a positive scan and 82/140 (59%) had no pathology on imaging. The most common other important diagnosis was hypox- ic brain injury (30/140; 21%) (Table 4). Patients with asystole, patients not in VF, patients with a longer downtime, patients who rearrested and patients with unreactive pupils in the ED were all more likely to have positive findings on CT imaging (Table 3). The most common scan was CT head (135), followed by CTPA (25), CT cervical spine (16), and CT CAP (12). Overall, 6/195 (3%) of scans identified the arrest cause, 62/195 (32%) showed other pathology and 127/195 (65%) were negative (Table 5). The causes of OHCA as judged by the ICU clinician are shown in Table 6. 41/156 (26%) of patients had a cause potentially iden- tifiable by CT, of which 7/41 (17%) received imaging. 57/156 (37%) patients had a cause that would not be identifiable by imag- ing. No diagnosis was made in 58/156 (37%) of patients. 48/156 (31%) patients survived to hospital discharge, with 13/58 (22%) of those with positive CT imaging surviving com- pared to 31/82 (38%) of those with no findings on CT (p=0.08). Median hospital length of stay was 30 (2–12) days and median ICU length of stay was 6 (2–6) days. No incidents of contrast nephropathy, allergic reactions to contrast or other imaging com- plications were identified. Scans did not result in inappropriate fur- ther investigations or treatments. Discussion This retrospective cohort study found that although the diag- nostic utility of CT imaging in OHCA with ROSC was limited, CT Article [page 44] [Emergency Care Journal 2022; 18:10669] Table 3. Characteristics of the OHCA event in study patients. OHCA Event* All patients Patients that received Patients with a Patients with a P value (n=156) a CT scan positive negative n (%), (n=140) n (%), CT scan†(n=58) CT scan‡ (n=82) mean ±SD mean ±SD n (%), mean ±SD n (%), mean ±SD or median (IQR) or median (IQR) or median (IQR) or median (IQR) Time of ED presentation 12am – 6am 22 (14) 19 (14) 11 (19) 8 (10) 0.19 6am – 12pm 28 (18) 27 (19) 8 (14) 19 (23) 0.24 12pm – 6pm 56 (36) 50 (36) 24 (41) 26 (32) 0.32 6pm – 12am 50 (32) 44 (31) 15 (26) 29 (35) 0.31 Vital Signs on ED Presentation Pulse (bpm) 93 ± 24 94 ± 24 96 ± 25 92 ± 24 0.43 Systolic BP (mmHg) 119 ± 38 119 ± 38 117 ± 43 121 ± 36 0.64 Diastolic BP (mmHg) 75 ± 26 75 ± 26 77 ± 30 74 ± 24 0.72 Temperature (°C) 35.2 ± 1.3 35.2 ± 1.4 34.9 ± 1.3 35.4 ± 1.4 0.29 Left pupil size (mm) - median (IQR) 2 (2 – 4) 3 (2 – 4) 3 (2 – 4) 3 (2 – 4) 0.61 Right pupil size (mm) – median (IQR) 2 (2 – 4) 3 (2 – 4) 3 (3 – 4) 3 (2 – 4) 0.48 Left pupil reactivity Reactive 51 (53) 47 (53) 12 (35) 35 (64) 0.01§ Sluggish 14 (15) 13 (15) 3 (9) 10 (18) 0.26 Unreactive 31 (32) 29 (33) 19 (56) 10 (18) 0.006§ Right pupil reactivity Reactive 49 (52) 46 (52) 11 (32) 35 (64) 0.005§ Sluggish 14 (15) 13 (15) 3 (9) 10 (18) 0.26 Unreactive 32 (34) 30 (34) 20 (59) 10 (18) 0.003§ Arrest rhythm VF 76 (51) 67 (50) 22 (39) 46 (60) 0.05§ VT 3 (2) 2 (2) 0 2 (3) 1.00 PEA 32 (21) 31 (23) 14 (25) 17 (22) 0.79 Asystole 38 (26) 33 (25) 21 (37) 12 (16) 0.005§ Total downtime (minutes) – median (IQR) 23 (15 – 40) 30 (15 – 40) 36 (20 – 50) 22 (15 – 37) 0.02§ Patient rearrested 50 (32) 42 (30) 28 (48) 14 (17) <0.0001§ Bystander CPR 120 (78) 106 (77) 44 (76) 62 (76) 0.86 Ambulance response time (minutes) – median (IQR) 5 (5 – 11) 7 (5 – 12) 7 (4 – 12) 6 (5 – 11) 0.99 Intubated pre-hospital 51 (33) 46 (33) 19 (33) 27 (33) 0.86 Intubated in ED 105 (67) 94 (67) 39 (67) 55 (67) 0.86 *Data was complete for intubation and rearrests. 154 (99%) of all patients had data for bystander CPR, 149 (96%) for arrest rhythm, 143 (92%) for downtime, 125 (80%) for pulse, 112 (72%) for systolic blood pressure, 100 (64%) for diastolic blood pressure, 96 (62%) for pupil reactivity, 67 (43%) for pupil diameter, 48 (31%) for temperature. †’Positive’ CT scan: identified OHCA cause or other important pathology. ‡‘Negative’ CT scan: no pathology. §Significant at 5% level. Non -co mmerc ial us e o nly was a useful tool identifying other pathology mostly secondary to arrest or resuscitation. 140/156 (90%) patients underwent CT scan- ning ordered within an hour of ED arrival, most commonly CT head and CTPA. The arrest cause was identified by CT in only 6/140 (4%) patients, but 52/140 (37%) patients had other relevant pathology on imaging (4/140 (3%) had both the OHCA cause and other pathology on imaging). 82/140 (59%) had no findings. The most common other important diagnosis was hypoxic brain injury. Patients with asystole, patients not in VF, patients with a longer downtime, patients who rearrested and patients with unreactive pupils on ED admission were more likely to have positive imaging findings. CT imaging was safe with no instances of contrast nephropathy, allergic reactions or other complications recorded. Current guidelines for post-resuscitation care advise investiga- tions such as ECG, chest radiograph and coronary angiography +/- PCI. CT head or CT chest imaging is recommended for early iden- tification of neurological or respiratory causes, with whole body scanning reserved for OHCA with trauma or haemorrhage.5 Our study shows practice in the RIE ED conforms with this guidance.5 Our data support the use of head and chest CT imaging, with CT head and CTPA the only scan types to identify arrest causes. Cranial CT imaging identified all 4 patients with stroke as an arrest cause. CTPA scans diagnosed one patient with acute coronary syn- drome and one with pulmonary embolism. CT imaging therefore Article Table 4. Results of CT Imaging in study patients undergoing CT scan. Results of CT Imaging (n=140)* n (%) OHCA cause diagnosed by CT 6 (4) Haemorrhagic stroke 3 (2) Ischaemic stroke 1 (1) Acute coronary syndrome 1 (1) Pulmonary embolism 1 (1) Patients with other important diagnoses identified by CT** 56 (40) Hypoxic brain injury 30 (21) Rib fracture 20 (14) Lobar collapse/consolidation 12 (9) Cerebral oedema 10 (7) Pneumonia 8 (6) Pulmonary oedema 6 (4) Pneumothorax 4 (3) Cervical spine fracture 3 (2) Flail segment 3 (2) Raised ICP 2 (1) Left ventricular systolic dysfunction 2 (1) Intraabdominal haemorrhage 2 (1) Mesenteric ischaemia due to CA 2 (1) Traumatic brain injury 1 (1) Ischaemic stroke due to CA 1 (1) Pneumomediastinum 1 (1) Surgical emphysema 1 (1) Unruptured aortic aneurysm 1 (1) Hydronephrosis 1 (1) Patients with no findings on CT 82 (59) *4 patients had both the OHCA cause and other pathology on imaging, 2 had OHCA cause but no other pathology, 52 had no cause but other important pathology. Percentages are of patients imaged (n=140). **Total diagnoses > total patients as patients could have multiple diagnoses. Table 5. Type of CT scans performed in study patients undergoing CT scan. CT Scans* OHCA Other important Negative scans‡ Total number Time from cause diagnosis (n=127) of scans presentation identified identified† n (%) minutes (n=6) (n=62) (mean ± SD) n (%) n (%) CT head 4 (3)§ 31 (23) 100 (74) 135 35 ± 34 CT pulmonary angiogram 2 (8) 16 (64) 7 (28) 25 60 ± 124 CT cervical spine 0 2 (13) 14 (88) 16 22 ± 10 CT chest/abdomen/pelvis 0 8 (67) 4 (33) 12 69 ± 66 Whole-body CT 0 5 (100) 0 5 27 ± 28 CT thoracic aorta 0 0 2 (100) 2 72 CT coronary angiogram - - - 0 - All scans 6 (3) 62 (32) 127 (65) 195 39 ± 56 *‘CT head’ includes 134 CT head scans and 1 CT intracranial angiogram. ‘CT chest/abdomen/pelvis’ includes 7 CT chest/abdomen/pelvis scans, 1 CT chest scan, 2 CT abdomen scans and 2 CT pelvis scans. †These scans identified other important pathology but not the OHCA cause. ‡’Negative’ CT scan: no pathology. §Percentages of total number of scan type. May total >100% due to rounding. [Emergency Care Journal 2022; 18:10669] [page 45] Non -co mmerc ial us e o nly only diagnosed the cause of OHCA in 6/140 (4%) patients scanned (Table 4). This was surprising in context of previous research. The Parisian Region Out of Hospital Cardiac Arrest registry (PRO- CAT) reported 355 head and chest CT scans over ten years, diag- nosing OHCA cause in 20%.12 In 2021, Branch et al. diagnosed 39% of OHCA causes using early head-to-pelvis CT imaging of 100 patients in a prospective study.11 The difference in diagnostic utility in this study may be because only 5 patients received a whole-body CT scan. When considering the recorded clinical cause of arrest, 41/156 (26%) of causes were potentially diagnos- able by CT (Table 6). 52/140 (37%) patients had other relevant pathology on imag- ing, largely secondary to arrest or resuscitation. Hypoxic brain injury or cerebral oedema was present in 40/140 (29%), similar to rates of 35% in previous research.13 Such findings are relevant to intensive care management and prognostication.14 Chest findings likely secondary to arrest or resuscitation were common, with 20 rib fractures and 4 pneumothoraces. Plain chest radiography under- estimates such CPR-related pathologies.15 Regurgitation and aspi- ration are known risks during arrest.16,17 We found 20 reports of lobar collapse/consolidation/pneumonia, clinically likely to repre- sent aspiration.18 This is comparable to a previous retrospective study of 100 patients reporting CPR-related complications in 93%, and presumed aspiration-related pulmonary opacification in 77%.10 Overall, 58/140 (41%) of patients had ‘positive’ imaging (iden- tified the arrest cause or other important pathology) that informed patient care, for example indications for antibiotic therapy (pneu- monia), antithrombotics (pulmonary embolism) and their con- traindications (haemorrhagic stroke, intraabdominal haemorrhage) and reasons for prompt surgical referral (flail segment, intraab- dominal haemorrhage). Although it cannot be assessed in this study, imaging may have impacted care of patients without any pathological findings, for example by excluding catastrophic brain injury. It was possible to identify the characteristics of patients with ‘positive’ CT imaging that were significantly different to those with only negative scans. Patients with asystole, patients not in VF, patients with a longer downtime, patients who rearrested and patients with unreactive pupils in the ED were all more likely to have positive findings. It is unsurprising that patients with clinical indicators of a more serious arrest showed pathology on imaging. In this scenario, positive findings on CT were more likely to be associated with devastating pathologies i.e., PE and haemorrhagic stroke. Also, patients with VF cardiac arrest, which has a better prognosis, are more likely to have an underlying cardiac cause for their cardiac arrest, which is unlikely to be demonstrated on CT scan. The CASPRI scoring system is validated to predict neurolog- ical outcome after in-hospital cardiac arrest.19 Consistent with our results, asystole arrest and longer downtime indicate poor neuro- logical outcome in the score. A non-significant number of patients with positive scans died compared with those with negative scans [45/58 (86%) vs 51/82 (62%)]. Positive scans may be associated with worse survival due to the higher complication rate in this population. The benefits of CT imaging must be weighed against the potential risks. Imaging was safe, with no incidents of acute kidney injury, allergic reaction or any other complications found, and no inappropriate treatments were instated because of imaging. During this study the ERC published guidelines20 that were adopted by our emergency and critical care departments,21 recom- mending CT head and CTPA in those with no evidence of myocar- Mini Review Table 6. The causes of OHCA as judged by the ICU clinician. ICU/Clinical Cause of OHCA Cause potentially diagnosable by CT (n=41) n (%) Received CT imaging† (n=7) n (%) Acute coronary syndrome* 33 (80) 2 (6) Haemorrhagic stroke 3 (7) 3 (100) Ischaemic stroke 1 (2) 1 (100) Severe LVSD/Dilated cardiomyopathy 1 (2) 0 HOCM 1 (2) 0 Pulmonary embolism 1 (2) 1 (100) Mesenteric ischaemia 1 (2) 0 Cause not diagnosable by CT (n=57) n (%) Primary arrhythmia 32 (56) Substance use 8 (14) Respiratory arrest 5 (9) Influenza A 2 (4) Primary seizure 2 (4) Sepsis 2 (4) Electrolyte disorder 2 (4) COPD exacerbation 1 (2) Alcohol withdrawal 1 (2) Chemotherapy 1 (2) Graft vs host disease 1 (2) Unknown (n=58) *CT potentially able to detect coronary artery stenosis but not dynamic acute coronary syndrome. †Patients with a cause potentially identifiable by CT who received imaging. Percentage of patients with that diagnosis. [page 46] [Emergency Care Journal 2022; 18:10669] Non -co mmerc ial us e o nly dial ischaemia to exclude neurological or respiratory causes of arrest. They describe four criteria indicating likely poor outcome, one of which is diffuse anoxic injury on brain CT or MRI. Our study lends further weight and evidence to these recommendations. Limitations There are several limitations to this study. It is single centre and retrospective and so has the associated drawbacks. For exam- ple, patient information was limited to what was recorded at the time, and decisions around whether a patient underwent CT scan- ning, and the focus of any imaging, was at the discretion of the treating clinician at the time. This is pertinent to the prodrome before arrest, where only 94/156 (60%) patients had information on symptoms recorded. If a record of symptoms prior to arrest had been more readily available, then it may have been possible to draw comparisons between symptoms and scan results. Secondly, as a single centre study, the participants may not be representative of the wider population. One of the aims of this study was to develop a tool to aid deci- sion making in the ED. Although we did find significant differ- ences between those patients who showed pathology on imaging and those who did not, this was not a sufficient basis for a decision- making tool. In future greater sample sizes may allow the develop- ment of a decision-making tool based on vital signs and any pro- dromal symptoms. It may reveal a link between positive CT imag- ing and survival rates, informing prognostication for ongoing man- agement decisions and communication with relatives. Conclusions 9 in 10 patients presenting to the ED with idiopathic non-trau- matic OHCA and ROSC underwent CT imaging. The diagnostic value of CT in this patient population was limited, with arrest cause identified in 4% of patients. Imaging was a valuable method of identifying other relevant pathology due to arrest or resuscita- tion in 37% of patients. Patients with asystole, patients not in VF, patients with a longer downtime, patients who rearrested and patients with unreactive pupils in the ED were more likely to have positive findings on CT imaging. CT head and CTPA were most performed. Imaging was safe with no recorded cases of contrast nephropathy, allergic reactions, complications, or inappropriate treatments. References 1. Gräsner J-T, 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;105:188-95. 2. Scottish Out-of-Hospital Cardiac Arrest data linkage project: 2018/19 results. APS Group; 2020. 3. Gräsner J-T, Wnent J, Herlitz J, et al. Survival after out-of-hos- pital cardiac arrest in Europe - Results of the EuReCa TWO study. Resuscitation 2020;148:218-26. 4. Pokorna M, Necas E, Skripsky R, et al. How accurately can the aetiology of cardiac arrest be established in an out-of-hospital setting? Analysis by “Concordance in Diagnosis Crosscheck Tables”. Resuscitation 2010;82:391-7. 5. Nolan JP, Soar J, Cariou A, et al. European Resuscitation Council and European Society of Intensive Care Medicine 2015 guidelines for post-resuscitation care. Int Care Med 2015;41:2039-56. 6. Hawkes C, Booth S, Ji C, et al. Epidemiology and outcomes from out-of-hospital cardiac arrests in England. Resuscitation 2017;110:133-40. 7. Dunham GM, Perez-Girbes A, Bolster F, et al. Use of whole body CT to detect patterns of CPR-related injuries after sudden cardiac arrest. Eur Radiol 2018;28:4122-7. 8. Choi SJ, Kim HS, Kim EY, et al. Thoraco-abdominal CT examinations for evaluating cause of cardiac arrest and com- plications of chest compression in resuscitated patients. Emerg Radiol 2014;21:485-90. 9. Kashiwagi Y, Sasakawa T, Tampo A, et al. Computed tomog- raphy findings of complications resulting from cardiopul- monary resuscitation. Resuscitation 2015;88:86-91. 10. Viniol S, Thomas RP, König AM, et al. Early whole-body CT for treatment guidance in patients with return of spontaneous circulation after cardiac arrest. Emerg Radiol 2020;27:23-9. 11. Branch KRH, Strote J, Gunn M, et al. Early head�to�pelvis computed tomography in out�of�hospital circulatory arrest without obvious etiology. Acad Emerg Med 2021;28:394-403. 12. Chelly J, Mongardon N, Dumas F, 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. 13. Naples RMD, Ellison EMD, Brady WJMD. Cranial computed tomography in the resuscitated patient with cardiac arrest. Am J Emerg Med 2009;27:63-7. 14. Metter RB, Rittenberger JC, Guyette FX, Callaway CW. Association between a quantitative CT scan measure of brain edema and outcome after cardiac arrest. Resuscitation 2011;82:1180-5. 15. Lederer W, Mair D, Rabl W, Baubin M. Frequency of rib and sternum fractures associated with out-of-hospital cardiopul- monary resuscitation is underestimated by conventional chest X-ray. Resuscitation 2004;60:157-62. 16. Oschatz E, Wunderbaldinger P, Sterz F, et al. Cardiopulmonary resuscitation performed by bystanders does not increase adverse effects as assessed by chest radiography. Anesthesia Analgesia 2001;93:128-33. 17. Lawes EG, Baskett PJF. Pulmonary aspiration during unsuc- cessful cardiopulmonary resuscitation. Intensive Care Med 1987;13:379-82. 18. Franquet T, Giménez A, Rosón N, et al. Aspiration diseases: findings, pitfalls, and differential diagnosis. RadioGraphics 2000;20:673-85. 19. Girotra S, Nallamothu BK, Chan PS. Using Risk Prediction Tools in Survivors of In-hospital Cardiac Arrest. Current Cardiol Reports 2014;16:1-8. 20. Nolan JP, Sandroni C, Böttiger BW, et al. European Resuscitation Council and European Society of Intensive Care Medicine Guidelines 2021: Post-resuscitation care. Resuscitation 2021;161:220-69. 21. Waterson F, Fleming G, Young N, et al S. Out of Hospital Cardiac Arrest Guideline 2021 V5.0. 2022. Mini Review [Emergency Care Journal 2022; 18:10669] [page 47] Non -co mmerc ial us e o nly