Archives of Academic Emergency Medicine. 2023; 11(1): e33 OR I G I N A L RE S E A RC H Sustained Return of Spontaneous Circulation Following Out-of-Hospital Cardiac Arrest; Developing a Predictive Model Based on Multivariate Analysis Thongpitak Huabbangyang1, Agasak Silakoon1, Pramote Papukdee1∗, Rossakorn Klaiangthong1, Chaleamlap Thongpean2, Wannakorn Pralomcharoensuk2, Weerawan Khaokaen2, Sunisa Bumrongchai2, Ratree Chaisorn3, Chomkamol Saumok3 1. Department of Disaster and Emergency Medical Operation, Faculty of Science and Health Technology, Navamindradhiraj University, Bangkok, Thailand. 2. Faculty of Medicine Vajira Hospital, Navamindradhiraj University, Bangkok, Thailand. 3. Division of Division of Emergency Medical Service and Disaster, Faculty of Medicine Vajira Hospital, Navamindradhiraj University, Bangkok, Thailand. Received: February 2023; Accepted: March 2023; Published online: 27 April 2023 Abstract: Introduction: Identifying the predictive factors of sustained return of spontaneous circulation (ROSC) following out-of- hospital cardiac arrest (OHCA) will be helpful in management of these patients. This study aimed to develop a predictive model in this regard. Methods: In a retrospective observational study, data of adult patients with OHCA, were collected from Vajira emergency medical services patient care report. Multiple logistic regression analysis with a regression co- efficient was used to develop a predictive score for a sustained ROSC at the scene. Area under the receiver operating characteristic (ROC) curve (AUC) was used to validate the accuracy of the predictive score for a sustained ROSC. Re- sults: Independent factors associated with a sustained ROSC included cardiopulmonary resuscitation (CPR) duration < 30 min (adjusted odds ratio (AOR)= 5.05, 95% confidence interval (CI): 3.34–7.65; p < 0.001); advanced airway manage- ment with an endotracheal tube (AOR= 3.06, 95% CI: 1.77–5.31; p < 0.001); advanced airway management with laryngeal mask airway (AOR= 3.42, 95% CI: 1.02–11.46; p = 0.046); defibrillation (AOR = 2.05, 95% CI: 1.31–3.2; p = 0.002); Capillary blood glucose (CBG) level < 150 mg% (AOR= 1.95, 95% CI: 1.05–3.65; p = 0.035); CBG at least 150 mg% (AOR= 2.87, 95% CI: 1.56–5.29; p = 0.001); pupil reflex (AOR = 2.96, 95% CI: 1.1–7.96; p = 0.032); and response time at most 8 min (AOR= 1.66, 95% CI: 1.07–2.57; p = 0.023). These were developed into the pupil reflex, response time, advanced airway manage- ment, defibrillation, CBG, and CPR duration (PRAD-CCPR) score. The most accurate cutoff point of score using Youden’s index was ≥ 6 with AUC of 0.759 (95% CI: 0.715–0.802; p < 0.001), sensitivity of 62.0% (95% CI: 51.2–71.9%), specificity of 75.7% (95% CI: 69.4–81.2%), positive predictive value of 51.8% (95% CI: 40.9–62.3%), and negative predictive value of 79.5% (95% CI: 73.5–84.6%). Conclusion: An optimal PRAD-CCPR score of ≥ 6 provides an acceptable accuracy of 0.759 with sensitivity of 62.0% and specificity of 75.7% in prediction of sustained ROSC following OHCA. This predictive score might help CPR commanders to prognosticate the outcome of patients with OHCA at the scene. Keywords: Emergency Medical Services; Out-of-Hospital Cardiac Arrest; Heart Arrest; Return of Spontaneous Circulation Cite this article as: Ersin Altınsoy K, Murat Oktay M. Sustained Return of Spontaneous Circulation Following Out-of-Hospital Car- diac Arrest; Developing a Predictive Model Based on Multivariate Analysis. Arch Acad Emerg Med. 2023; 11(1): e33. https://doi.org/10.22037/aaem.v11i1.2012. ∗Corresponding Author: Pramote Papukdee; Department of Disaster and Emergency Medical Operation, Faculty of Science and Health Technology, Navamindradhiraj University, Bangkok 10400, Thailand. Tel: +66 2-244-3000, Email: pramote@nmu.ac.th, ORCID: https://orcid.org/0000-0002-3644-3925. 1. Introduction Out-of-hospital cardiac arrest (OHCA) is a medical emer- gency condition requiring immediate management to save lives and an important global cause of death, especially in middle-income or developing countries. The reported num- ber of the deceased due to cardiac arrest was a million per year (1). The worldwide prevalence was 50–60 per 100,000 populations (2). In the USA, the incidence in adults was more This open-access article distributed under the terms of the Creative Commons Attribution NonCommercial 3.0 License (CC BY-NC 3.0). Downloaded from: https://journals.sbmu.ac.ir/aaem/index.php/AAEM/index T. Huabbangyang et al. 2 than 350,000 per year, the survival rate was 10.8%, and only 9% had favorable neurological outcomes (3). The prevalence rates of 17–128 and 21–29 per 100,000 population in cen- tral America and Asia, respectively, were reported in a pre- vious study (4). In Thailand, the survival rate of patients with OHCA at the scene, resuscitated by emergency medical ser- vices (EMS), was 25.6% (5). According to the report of The Pan Asian Resuscitation Out- comes Study (PAROS) in Thailand, the survival rate was 4.0% (6), indicating that patients with OHCA have different out- comes based on each context of each country, such as dif- ferences in the study design, sample size, and demographic data, especially hospital data including prehospital manage- ment by EMS and management in the hospital (7). A systematic review and meta-analysis showed that pa- tients having OHCA with shockable rhythm had better sur- vival outcomes than those with non-shockable rhythm (8). Bystander cardiopulmonary resuscitation (bystander CPR) and bystander automated external defibrillator (AED) sig- nificantly improved the survival outcomes of patients with OHCA (9). Previous studies have demonstrated that predic- tive factors for a sustained return of spontaneous circulation (sustained ROSC) included trauma cause (5, 10), response time (5, 11), bystander CPR (11), time to first chest compres- sion (7), younger age (12), arrest in a public area (12), wit- nessed arrest (12), and shockable rhythm (12). To the authors’ knowledge, three studies have developed pre- dictive scores for a sustained ROSC, namely, WATCH-CPR score (7), RACA score (13), NULL-PLEASE score (14), and P- ROSC (15). All previous studies have developed predictive scores in the context of management in the hospital or emer- gency department, which was difficult to apply in prehos- pital practice and lacked data on the time and treatment of patients at the scene, which are important in the prehospi- tal context. Some studies have used laboratory data, which could not be applied in the prehospital context. However, a modeling study for the prediction of sustained ROSC of pa- tients with OHCA at the scene would help paramedics and emergency nurse practitioners manage patients with OHCA, ethically. Thailand is a middle-income country, and no emergency physicians are working with ambulances at the scene in many areas (10). Resources are substantially limited in the prehospital con- text compared with the management at the emergency de- partment. An important issue is how to efficiently resusci- tate with the best outcome and lowest cost. Therefore, this study aimed to develop and validate a predictive model for a sustained ROSC of patients with OHCA at the scene. 2. Methods 2.1. Study design and settings This retrospective observational study was conducted in Va- jira Emergency Medical Service (V-EMS), Vajira Hospital, Faculty of Medicine, Navamindradhiraj University, Bangkok, Thailand, between January 1, 2019, and July 31, 2022. V-EMS was a leader of EMS unit zone area 1 from a total of nine area divisions of EMS in Bangkok, dispatched from Erawan Cen- ter, Bangkok, networking with public and private hospitals, a total of six hospitals, in an area of 50 km2 with 500,000 popu- lation (5, 10). During a response operation for patients with OHCA, the EMS team sent by V-EMS included at least three personnel (a paramedic or emergency nurse practitioner (ENPs) as opera- tion team leader, and two emergency medical technicians. In each response operation, paramedics or ENPs would operate under offline and online medical protocols under the orders of emergency physicians (EPs). In patients with cardiac ar- rest, the American Heart Association guidelines of 2020 were applied. All team members had passed advanced cardiovas- cular life-support training. The Standards for the Reporting of Observation Studies in Epidemiology (STROBE) statement were applied (16). 2.2. Participants Data of adult patients (aged > 18 years) with OHCA were col- lected from EMS patient care report, coded with the Thailand emergency medical triage protocol and criteria-based dis- patch (CBD) symptom group 6, which is cardiac arrest, man- aged by the V-EMS unit, Vajira Hospital, Faculty of Medicine, Navamindradhiraj University, Bangkok, Thailand, between January 1, 2019, and July 31, 2022. The patients were managed in accordance with advanced cardiovascular life support (ACLS) based on CPR guidelines (17, 18). Patients with incomplete or missing data, signs of irreversible death so that based on the judgement of the team leader no resuscitation was performed, do-not-resuscitate orders, car- diac arrest outside the scene, OHCA during transfer, i.e., pa- tients with OHCA receiving CPR starting at the scene and continuing during hospital transfer, and termination of re- suscitation at the scene were excluded. 2.3. Data collection Data of patients with OHCA were collected from the EMS pa- tient care report, which was a record of advanced EMS oper- ation, Bangkok EMS (Erawan Center), and the standard form and unit in the Bangkok advanced emergency operation unit. This form contained data of EMS operation units, patients, and all treatments by EMS teams, recorded by dispatchers and paramedics or ENPs operating at the scene. These data This open-access article distributed under the terms of the Creative Commons Attribution NonCommercial 3.0 License (CC BY-NC 3.0). Downloaded from: https://journals.sbmu.ac.ir/aaem/index.php/AAEM/index 3 Archives of Academic Emergency Medicine. 2023; 11(1): e33 Figure 1: Receiver operating characteristic (ROC) curves for the predictive model of sustained return of spontaneous circulation among pa- tients with out-of-hospital cardiac arrest at the scene: (a) derivation set and (b) validation set. AuROC: area under the ROC curve. were a part of the remuneration for EMS operation units. All data were filled in and recorded in Microsoft Excel by a prin- cipal investigator. Data comprised general characteristics of patients with OHCA including sex, age, comorbidities, loca- tion type, witnessed arrest, bystander CPR, arrest type, type of traumatic arrest, type of non-traumatic arrest, time from arrest to chest compression, CPR duration, advanced airway management, defibrillation, fluid resuscitation, medication during CPR, Capillary blood glucose (CBG), pupil reflex, re- sponse time, and sustained ROSC at the scene. 2.4. Definitions - The sustained ROSC was determined when chest compres- sions were not required for 20 minutes, and signs of circula- tion persisted for at least 20 minutes (7). - Symptom group 6 was defined as OHCA according to the emergency level screening system of Thailand, classified ac- cording to CBD 6, severity level–critical (red) 6 critical 1 or 6 red 1, defined as cardiac arrest including unconsciousness, apnea, or pulselessness (5). - Response time was defined as the duration from the emer- gency call to ambulance arrival at the scene (5). - The time from arrest to chest compression was defined as the interval between sudden cardiac collapse to chest com- pression by a bystander. - CPR duration was defined as the duration from the first medical contract (FMC) to the end of CPR based on the EMS patient care report. - The derivation set was defined as a new dataset used by the authors to develop the predictive score for a sustained ROSC of patients with OHCA at the scene. - The validation set was defined as the dataset used to vali- date the accuracy of the predictive score for a sustained ROSC of patients with OHCA at the scene. 2.5. Sample size determination The main objective of this study was to develop a predic- tive score for a sustained ROSC in patients with OHCA at the scene using multiple logistic regression analysis. Sample size estimation relied on multiple logistic regression analy- sis using the number of events per variable in a logistic re- gression analysis (19). The analysis required Enough sam- ples to include at least 10 interesting events for each inde- pendent variable. At most, 15 variables in the development of the predictive score for a sustained ROSC were expected. The number of samples with interesting events, which was sustained ROSC, of 150 events/samples was required. The rate/incidence of sustained ROSC was 25.6%. Therefore, the total sample size for analysis was at least 586 ((150*100)/25.6 = 586). After 5% of the sample size was added using the formula (nnew = 586/[1 – 0.05]), the required sam- ple size was calculated to be at least 617. Thus, the sample size for the development of the predictive score for a sus- tained ROSC was 620. This open-access article distributed under the terms of the Creative Commons Attribution NonCommercial 3.0 License (CC BY-NC 3.0). Downloaded from: https://journals.sbmu.ac.ir/aaem/index.php/AAEM/index T. Huabbangyang et al. 4 For the sample size estimation to validate the accuracy of the predictive score (validation set), the ratio of the sample size to the sample size during the development of the score (derivation set) was 1:2. Thus, sample sizes for the derivation and validation sets were 620 and the 310, respectively. Hence, the final sample size in the present study was 930, and simple random sampling was used. 2.6. Statistical analysis A descriptive analysis was performed to examine the vari- able distribution. Continuous variables are presented as mean ± standard deviation (SD) or median and interquar- tile range (IQR), and categorical variables are presented as frequencies and proportions. When comparing the two groups, differences were evaluated using independent t-test or Mann–Whitney U test for numeric variables and chi- square test or Fisher’s exact test for categorical variables. In the development of the predictive score, multivariable analysis with multiple logistic regression analysis and a back- ward stepwise selection method were used. All statistical tests were considered statistically significant at p < 0.05. Sig- nificant factors associated with a sustained ROSC were ob- tained from the univariable analysis. Regression coefficient, odds ratio (OR), and 95% confidence intervals (CI) were re- ported. The predictive score was developed using the regres- sion coefficient. The accuracy of the predictive score was validated using re- ceiver operating characteristic (ROC) curve and area under the curve (AUC), using Youden’s index to determine the best cutoff point. Sensitivity, specificity, accuracy, positive predic- tive value (PPV ), negative predictive value (NPV ), and area under the ROC curve were reported with 95% CI. IBM SPSS Statistics for Windows, version 28.0 (IBM Corp., Ar- monk, NY, USA) and Stata version 13.0 (StataCorp, College Station, TX, USA) were used. All statistical tests were consid- ered statistically significant at p < 0.05. 2.7. Ethical statement This study was conducted in accordance with the tenets of the Declaration of Helsinki 1975 and its revisions in 2000. It was approved by the Institutional Review Board of the Fac- ulty of Medicine Vajira Hospital, Navamindradhiraj Univer- sity (COA no. 217/2565). Informed consent requirement was waived because of the retrospective nature and anonymity of all patient data. 3. Results 3.1. characteristics of patients in derivation set 620 cases were studied (66.0% male). Table 1 compares the baseline characteristics of patients between cases with and without sustained ROSC. The two groups had significant dif- ferences regarding CPR duration (p < 0.001), Advanced air- way management (p = 0.001), defibrillation (p < 0.001), the median concentrations of CBG (p < 0.001), Pupil reflex (p < 0.001), and response time (p = 0.030). 3.2. Developing the predictive model Based on multivariate analysis using multiple logistic regres- sion analysis and backward stepwise selection method, in- dependent associated factors of sustained ROSC of patients with OHCA at the scene were: CPR duration < 30 min (AOR = 5.05, 95% CI: 3.34–7.65; p < 0.001); airway management at the scene with endotracheal tube (ETT) (AOR = 3.06, 95% CI: 1.77–5.31; p < 0.001) or laryngeal mask airway (LMA) (AOR = 3.42, 95% CI: 1.02–11.46; p = 0.046); defibrillation (AOR = 2.05, 95% CI: 1.31–3.2; p = 0.002); CBG level < 150 mg% (AOR= 1.95, 95% CI: 1.05–3.65; p = 0.035); CBG at least 150 mg% (AOR = 2.87, 95% CI: 1.56–5.29; p = 0.001); had pupil reflex (AOR = 2.96, 95% CI: 1.1–7.96; p = 0.032); had response time at most 8 min (AOR = 1.66, 95% CI: 1.07–2.57; p = 0.023) (Table 2). The analyzed results of the predictive formula for a sustained ROSC of patients with OHCA at the scene in the form of logit transformation were used to develop the predictive score us- ing a regression coefficient. CPR duration < 30 min was as- signed 3 points; advanced airway management with ETT, 2 points; advanced airway management with LMA, 2.5 points; defibrillation, 1.5 points; CBG <150 mg%, 1 point; CBG ≥150 mg%, 2 points; pupil reflex, 2 points; and response time ≤ 8 min, 1 point. The total score ranged from 0 to 15 points. The predictive score could predict sustained ROSC of pa- tients with OHCA at the scene significantly with AUC of 0.759 (95% CI 0.715–0.802; p < 0.001). 3.3. Validation of model 310 cases were studied (65.8% male). Table 1 compares the baseline characteristics of patients between patients with and without sustained ROSC in the validation set. Using the validation dataset with 310 patients, the predic- tive score could significantly predict sustained ROSC of pa- tients with OHCA at the scene, with AUC of 0.732 (95% CI 0.673–0.792, p < 0.001). The predictive score of at least 6 had the highest Youden’s index (0.377), which was the most suitable point in the prediction of sustained ROSC among patients with OHCA at the scene, with sensitivity of 62.0% (95% CI: 51.2–71.9), specificity of 75.7% (95% CI: 69.4–81.2), PPV of 51.8% (95% CI: 40.9–62.3), and NPV of 79.5% (95% CI: 73.5–84.6) (Table 3 and Figure 1). 4. Discussion First, in this study, the incidence of sustained ROSC at the scene was 42.5% in patients with OHCA resuscitated by EMS in Thailand, which is a middle-income country, like the pre- This open-access article distributed under the terms of the Creative Commons Attribution NonCommercial 3.0 License (CC BY-NC 3.0). Downloaded from: https://journals.sbmu.ac.ir/aaem/index.php/AAEM/index 5 Archives of Academic Emergency Medicine. 2023; 11(1): e33 Table 1: Characteristics of patients with out-of-hospital cardiac arrest with and without sustained return of spontaneous circulation (ROSC) in derivation and validation sets Characteristics Derivation set (n = 620) p-value Validation set (n = 310) p-value Total Sustained ROSC Death Total Sustained ROSC Death) (n = 620) (n = 177) (n = 443) (n = 310) (n= 218) (n = 92) Sex Male 409 (66.0) 112 (63.3) 297 (67.0) 0.371 204 (65.8) 60 (65.2) 144 (66.1) 0.887 Female 211 (34.0) 65 (36.7) 146 (33.0) 106 (34.2) 32 (34.8) 74 (33.9) Age (year) Mean ±SD 63.86 ± 18.26 62.37 ± 18.08 64.46 ± 18.32 0.197 63.14 ± 18.99 62.77 ± 19.11 63.29 ± 18.99 0.827 Comorbidities Hypertension 153 (24.7) 44 (24.9) 109 (24.6) 0.947 73 (23.5) 21 (22.8) 52 (23.9) 0.846 Diabetes mellitus 130 (21.0) 38 (21.5) 92 (20.8) 0.846 55 (17.7) 19 (20.7) 36 (16.5) 0.384 Heart disease 116 (18.7) 42 (23.7) 74 (16.7) 0.043 56 (18.1) 20 (21.7) 36 (16.5) 0.275 Dyslipidemia 29 (4.7) 7 (4.0) 22 (5.0) 0.590 11 (3.5) 1 (1.1) 10 (4.6) 0.184 Respiratory disease 87 (14.0) 10 (5.6) 77 (17.4) <0.001 53 (17.1) 7 (7.6) 46 (21.1) 0.004 Stroke 7 (1.1) 4 (2.3) 3 (0.7) 0.107 6 (1.9) 2 (2.2) 4 (1.8) 1.000 Renal diseases 29 (4.7) 9 (5.1) 20 (4.5) 0.761 17 (5.5) 7 (7.6) 10 (4.6) 0.286 Cancer 12 (1.9) 6 (3.4) 6 (1.4) 0.111 8 (2.6) 4 (4.3) 4 (1.8) 0.243 Other 19 (3.1) 7 (4.0) 12 (2.7) 0.416 12 (3.9) 6 (6.5) 6 (2.8) 0.193 Location type Non-public 509 (82.1) 146 (82.5) 363 (81.9) 0.873 242 (78.1) 69 (75.0) 173 (79.4) 0.397 Public 111 (17.9) 31 (17.5) 80 (18.1) 68 (21.9) 23 (25.0) 45 (20.6) Witnessed arrest No 310 (50.0) 96 (54.2) 214 (48.3) 0.182 131 (42.3) 38 (41.3) 93 (42.7) 0.825 Yes 310 (50.0) 81 (45.8) 229 (51.7) 179 (57.7) 54 (58.7) 125 (57.3) Bystander CPR No 315 (50.8) 99 (55.9) 216 (48.8) 0.107 144 (46.5) 47 (51.1) 97 (44.5) 0.288 Yes 305 (49.2) 78 (44.1) 227 (51.2) 166 (53.5) 45 (48.9) 121 (55.5) Type of arrest Non-trauma 587 (94.7) 168 (94.9) 419 (94.6) 0.868 293 (94.5) 87 (94.6) 206 (94.5) 0.980 Trauma 33 (5.3) 9 (5.1) 24 (5.4) 17 (5.5) 5 (5.4) 12 (5.5) Type of traumatic arrest (n = 50) Blunt 28 (84.8) 8 (88.9) 20 (83.3) 1.000 15 (88.2) 3 (60.0) 12 (100.0) 0.074 Penetrating 5 (15.2) 1 (11.1) 4 (16.7) 2 (11.8) 2 (40.0) 0 (0.0) Type of non-traumatic arrest (n = 880) Non-cardiogenic 312 (53.2) 62 (36.9) 250 (59.7) <0.001 159 (54.3) 28 (32.2) 131 (63.6) <0.001 Cardiogenic 275 (46.8) 106 (63.1) 169 (40.3) 134 (45.7) 59 (67.8) 75 (36.4) Time from arrest to chest compression < 15 min 476 (76.8) 139 (78.5) 337 (76.1) 0.513 240 (77.4) 73 (79.3) 167 (76.6) 0.598 ≥ 15 min 144 (23.2) 38 (21.5) 106 (23.9) 70 (22.6) 19 (20.7) 51 (23.4) CPR duration < 30 min 263 (42.4) 112 (63.3) 151 (34.1) <0.001 126 (40.6) 50 (54.3) 76 (34.9) 0.001 ≥ 30 min 357 (57.6) 65 (36.7) 292 (65.9) 184 (59.4) 42 (45.7) 142 (65.1) Airway management ETT 456 (73.5) 147 (83.1) 309 (69.7) 0.001 245 (79.0) 81 (88.0) 164 (75.2) 0.016 BVM 148 (23.9) 24 (13.6) 124 (28.0) 57 (18.4) 8 (8.7) 49 (22.5) LMA 16 (2.6) 6 (3.4) 10 (2.3) 8 (2.6) 3 (3.3) 5 (2.3) Defibrillation No 472 (76.1) 117 (66.1) 355 (80.1) <0.001 240 (77.4) 61 (66.3) 179 (82.1) 0.002 Yes 148 (23.9) 60 (33.9) 88 (19.9) 70 (22.6) 31 (33.7) 39 (17.9) Fluid resuscitation No 17 (2.7) 2 (1.1) 15 (3.4) 0.295 6 (1.9) 1 (1.1) 5 (2.3) 0.602 0.9% NaCl 574 (92.6) 167 (94.4) 407 (91.9) 295 (95.2) 87 (94.6) 208 (95.4) Lactate ringer 29 (4.7) 8 (4.5) 21 (4.7) 9 (2.9) 4 (4.3) 5 (2.3) Medication during CPR Adrenaline 618 (99.7) 177 (100.0) 441 (99.5) 1.000 310 (100.0) 92 (100.0) 218 (100.0) NA 7.5% SB 247 (39.8) 62 (35.0) 185 (41.8) 0.122 128 (41.3) 36 (39.1) 92 (42.2) 0.616 This open-access article distributed under the terms of the Creative Commons Attribution NonCommercial 3.0 License (CC BY-NC 3.0). Downloaded from: https://journals.sbmu.ac.ir/aaem/index.php/AAEM/index T. Huabbangyang et al. 6 Table 1: Characteristics of patients with out-of-hospital cardiac arrest with and without sustained return of spontaneous circulation (ROSC) in derivation and validation sets Characteristics Derivation set (n = 620) p-value Validation set (n = 310) p-value Total Sustained ROSC Death Total Sustained ROSC Death) (n = 620) (n = 177) (n = 443) (n = 310) (n= 218) (n = 92) Amiodarone 56 (9.0) 17 (9.6) 39 (8.8) 0.753 27 (8.7) 6 (6.5) 21 (9.6) 0.375 50% glucose 49 (7.9) 8 (4.5) 41 (9.3) 0.048 29 (9.4) 9 (9.8) 20 (9.2) 0.867 10% CG 100 (16.1) 25 (14.1) 75 (16.9) 0.391 56 (18.1) 17 (18.5) 39 (17.9) 0.902 CBG Median (IQR) 136.5 52.5–211124 22 –200 158 101–236 <0.001 129.5 24 –208 118 1 –198 153 95- 252 <0.001 Pupil reflex Non-response 598 (96.5) 162 (91.5) 436 (98.4) <0.001 298 (96.1) 84 (91.3) 214 (98.2) 0.008 Response 22 (3.5) 15 (8.5) 7 (1.6) 12 (3.9) 8 (8.7) 4 (1.8) Response time ≤ 8 min 162 (26.1) 57 (32.2) 105 (23.7) 0.030 81 (26.1) 25 (27.2) 56 (25.7) 0.786 > 8 min 458 (73.9) 120 (67.8) 338 (76.3) 229 (73.9) 67 (72.8) 162 (74.3) Data are presented as mean ± standard deviation (SD); median (interquartile range;IQR), or frequency (%). Abbreviations: BVM: bag valve mask; CPR: cardiopulmonary resuscitation; ETT: endotracheal tube; LMA: laryngeal mask airway; NA, data not applicable; CBG: capillary blood glucose; SB: sodium bicarbonate; CG: calcium gluconate. P-value corresponds to t Independent samples t-test, m Mann–Whitney U test, c Chi-square test, or fFisher’s exact test. Table 2: Multivariate logistic regression analyses to identify the independent factors associated with a sustained return of spontaneous circu- lation at the scene Factors B OR 95% CI p-value Score CPR duration < 30 min 1.620 5.05 (3.34–7.65) <0.001 3 ≥ 30 min 1.00 Reference Advanced airway management Bag valve mask 1.00 Reference Endotracheal tube 1.119 3.06 (1.77–5.31) <0.001 2 Laryngeal mask airway 1.230 3.42 (1.02–11.46) 0.046 2.5 Defibrillation No 1.00 Reference Yes 0.717 2.05 (1.31–3.2) 0.002 1.5 Capillary blood glucose (mg%) (1) 1.00 Reference < 150 0.670 1.95 (1.05–3.65) 0.035 1 ≥ 150 1.056 2.87 (1.56–5.29) 0.001 2 Pupil reflex No response 1.00 Reference Response 1.084 2.96 (1.1–7.96) 0.032 2 Response time ≤ 8 min 0.506 1.66 (1.07–2.57) 0.023 1 > 8 min 1.00 Reference Data are presented with 95% confidence interval (CI). OR: Adjusted odds ratio estimated by multiple logistic regression; CPR: cardiopulmonary resuscitation. Variables included in the multivariable model had p < 0.050 in the univariate analysis. Model summary: –2 Log likelihood = 623.963, Cox & Snell R Square = 0.173, Nagelkerke R Square = 0.248. Hosmer and Lemeshow test: Chi-square = 8.418, df = 8, p-value = 0.394. Constant: –3.704. vious study in a province in central Thailand, reporting an in- cidence of 39.0% for sustained ROSC (7). The present study demonstrated that ROSC achieved in Thailand was higher than that in a low-income country with an incidence of only 27.35% for ROSC in the emergency department (20). Previ- ous empirical evidence showed that countries with high eco- nomic and social status had a low incidence of OHCA and sig- nificantly high rate of favorable outcomes for ROSC (21)(22). Second, six predictors were associated with a sustained ROSC at the scene, including CPR duration < 30 min, advanced air- This open-access article distributed under the terms of the Creative Commons Attribution NonCommercial 3.0 License (CC BY-NC 3.0). Downloaded from: https://journals.sbmu.ac.ir/aaem/index.php/AAEM/index 7 Archives of Academic Emergency Medicine. 2023; 11(1): e33 way management with endotracheal tube (ETT) and laryn- geal mask airway (LMA), defibrillation, pupillary light reflex, and response time ≤ 8 min. The possible explanation re- garding patients with OHCA having CPR duration < 30 min, specified as the duration from the FMC to the end of CPR, was that patients with OHCA having CPR duration < 30 min had a chance of sustained ROSC, consistent with a nation- wide multicenter observational study in Japan reporting that patients with OHCA receiving CPR < 30 min would have an increased chance of ROSC at the scene and 1-month sur- vival and favorable neurological outcome defined by cerebral performance category scores 1 or 2, compared with patients with OHCA having CPR duration > 30 min. However, the au- thors explained that the outcomes were due to factors such as shockable rhythm, witnessed arrest, and bystander CPR (23) and comparable with those of a previous study finding that the most optimal cutoff for prehospital CPR duration in the non-traumatic group of patients with OHCA for obtaining ROSC and favorable neurological outcome at hospital dis- charge was 12 min, independent of cardiac rhythm. A longer CPR duration did not increase the rate of achieving ROSC (24). Therefore, CPR duration could help emergency medical personnel working prehospitally properly decide whether to terminate CPR in patients requiring prolonged CPR. The ex- planation regarding patients with OHCA obtaining prehospi- tal advanced airway management with ETT and LMA as fac- tors that helped increase sustained ROSC at the scene com- pared with bag valve masks (BVMs) was consistent with our previous study from the national EMS database of Thailand, reporting that patients with OHCA who received advanced airway management with ETT had 3.88 times higher chance of ROSC at the scene compared with those with BVMs (10). Additionally, they were similar to results of several studies re- porting that prehospital advanced airway with ETT and LMA by paramedics in patients with OHCA were associated with increased rate of ROSC at the scene, compared with patients with OHCA using BVMs (25-27), which was in contrast to a previous study finding that airway management with ETT during CPR in emergency department was associated with worsened outcome of ROSC and decreased survival until dis- charge at 28 days, compared with the non-ETT group with BVMs (28), and ETT and LMA applications during CPR for pa- tients with in-hospital cardiac arrest were associated with in- crease of no-flow interval, compared with BVMs application. A high no-flow interval is associated with worsened neuro- logical outcome, and helps emergency medical personnel in deciding for CPR initiation or termination in patients with OHCA (29). ETT insertion is the gold standard definitive airway manage- ment, but it has numerous limitations, such as local protocol; in some areas or countries, paramedics or Emergency nurse practitioners (ENPs) are not allowed to insert ETT, except when a physician is available and an ambulance is present at the scene. However, for the study area, paramedics and ENPs were allowed to insert ETT and administer medication in patients with OHCA under offline protocol (5). Defibril- lation was associated with a sustained ROSC at the scene, which was similar to a previous systematic review and meta- analysis reporting that patients with OHCA who received pre- hospital defibrillation had increased 1-month favorable neu- rological outcomes, and the rate was higher in patients pre- senting with initial shockable heart rhythms including ven- tricular fibrillation (VF) and pulseless ventricular tachycardia (VT) compared with patients without defibrillation and pre- senting with non-shockable rhythms (8). In this study, pre- hospital CBG was associated with a sustained ROSC. Even if hypoglycemia was one of the reversible causes of cardiac ar- rest in 2005 guidelines (30) and removed in ACLS 2010 (31), 2015 (32), and 2020 (33), recently the observational study in the emergency department in Thailand found that patients with OHCA who had intra-arrest blood glucose level < 100 mg/dL had decreased rate of sustained ROSC (34), which was consistent with the results of a large study retrospec- tively collecting data of patients with in-hospital cardiac ar- rest for 10 years in Taiwan. It reported that patients who had intra-arrest blood glucose level < 150 mg/dL had worse neurological outcomes and decreased rate of ROSC (35). For the study area, prehospital CBG administration is routine practice in patients with OHCA and in patients with hypo- glycemia, which is easily corrected in the prehospital con- text by 50% glucose intravenous administration and normal blood glucose level was tried to be maintained during CPR. This was believed to help improve clinical outcomes. This study revealed that patients with OHCA who had pupillary light reflex had more chances of sustained ROSC than pa- tients without pupillary light reflex. The finding was consis- tent with a previous study reporting that pupillary light re- flex could be used to predict outcomes after cardiac arrest, especially increased ROSC rate. In contrast, patients with- out pupillary light reflex would have significantly increased mortality rate (36), which contradicts the results of a recent study reporting that measurement of pupil size and pupillary light reflex could not be used to predict ROSC and the mea- surements were not associated with neurological outcomes in patients with cardiac arrest (37). Another factor used in the prediction of sustained ROSC was response time ≤ 8 min, which was defined as the interval between emergency call to ambulance arrival at the scene. This result was consis- tent with those of previous studies reporting that response time was associated with a sustained ROSC at the scene and neurological outcome, and with each minute increase in re- sponse time, the chance of ROSC would decrease in patients with OHCA (5, 15). In the study area, response time was guar- anteed to be at most 8 min in patients identified with un- This open-access article distributed under the terms of the Creative Commons Attribution NonCommercial 3.0 License (CC BY-NC 3.0). Downloaded from: https://journals.sbmu.ac.ir/aaem/index.php/AAEM/index T. Huabbangyang et al. 8 consciousness or cardiac arrest. In patients with cardiac ar- rest, staff at the dispatch center advised callers to perform dispatcher-assisted bystander CPR (DA-CPR) before the am- bulance arriving at the scene, and in the study area, the traffic police were coordinated to ensure the traffic route of ambu- lances and reduce response time. Third, this study developed a simple tool for prediction of sustained ROSC at the scene, helping prehospital emergency medical personnel, such as EPs, paramedics, and ENPs, to identify predictors of sustained ROSC in patients with OHCA, and patients’ relatives could know the prognosis in emer- gency situations. Factors that influence the management of patients with OHCA at the scene could be used in the prediction of sustained ROSC. This study developed a sim- ple tool for the prediction of sustained ROSC in patients with OHCA, called PRAD-CCPR score, including six factors, namely, pupil reflex, response time, airway management, de- fibrillation, CBG, and CPR duration. The total score ranged from 0 to 15, with simplicity and convenience in remem- brance and clinical application and appropriate to prehospi- tal context. The score of at least 6 was the most suitable cutoff point in prediction of sustained ROSC at the scene with AUC of 0.759 (95% CI 0.715–0.802, p < 0.001), sensitivity of 62.0% (95% CI 51.2–71.9), specificity of 75.7% (95% CI 69.4–81.2), PPV of 51.8% (95% CI 40.9–62.3), and NPV of 79.5% (95% CI 73.5–84.6). To the authors’ knowledge, so far, four scores have been developed for immediate prediction of ROSC outcome, comprising WATCH-CPR score (including witnessed arrest, time from arrest to chest compression, and CPR duration) (7). WATCH-CPR score ≥2 could be used to predict a chance of sustained ROSC in the emergency department. This study collected data in prehospital and ED settings. Notwithstand- ing, the WATCH-CPR score would be easy to apply, and many important data were still believed to be missed in the con- text of management by the EMS team in specific situations, for example, management of patients with OHCA, such as re- sponse time and prehospital advanced airway management. The RACA score had AUC of 0.710 (95% CI 0.697–0.724) in- cluding factors such as sex, age ≥80 years, witnessed arrest, asystole, arrest location, presumable etiology of cardiac ar- rest, bystander CPR, and time until professional arrival (13). However, the RACA score was difficult, complex, and inap- propriate to apply in emergency conditions in the context of prehospital management. Besides, the interpretation of the score was complex and unfit for application. The NULL- PLEASE score variables included non-shockable rhythm, un- witnessed arrest, long no-flow period, long low flow period, pH < 7.2, lactate > 7, end-stage renal failure on dialysis, age > 85 years, on-going CPR, and extracardiac cause; and it had AUC of 0.632 (95% CI 0.523–0.741) (14). The NULL-PLEASE score was thought to be suitable for the hospital context. However, since obtaining the results of some included pa- rameters took time, and because no blood gas test and lac- tate test are available outside the hospital and in the context of prehospital EMS in Thailand, it could not be applied in this setting. The P-ROSC score is made up of variables including age, time to EMS arrival, first rhythm, arrest witnessed, and prehospital drug administration, and has an AUC of 0.806 (95% CI 0.799 –0.814) (15). Nevertheless, the P-ROSC score only used prehospital parameters in the prediction of ROSC in patients with OHCA using only five parameters. The P- ROSC score had a total score of 100, and the score proportion was divided for each parameter with complexity, inappropri- ate to apply in emergencies, and pressured conditions as the management of patients with OHCA. Therefore, the PRAD- CCPR score was proposed, which was a simple tool for the prediction of sustained ROSC at the scene including practical parameters in the context of prehospital management, com- prising treatment at the scene by an EMS team, for assisting EPs, paramedics, and ENPs when making critical decisions for patients with OHCA regarding survival prediction. 5. Limitations This study has several limitations. First, it was a retrospec- tive observational study, collecting data in a single center, an advanced emergency medical operation unit in Bangkok, Thailand, a middle-income country. Thus, due to differences between areas, the study results may not be indicative in other areas and it is required to assess the external validity of the score. Second, this study only analyzed sustained ROSC at the scene. Additional studies are needed on the predic- tive performance of the score regarding long-term outcomes, such as survival to hospital discharge and neurological func- tion outcome in patients with OHCA. Third, all data were ret- rospectively collected from the EMS patient care report; al- though neutrality was tried to be maintained in every way possible, there might be a risk of selection bias. 6. Conclusion In this study, a score called PRAD-CCPR was developed and validated for predicting sustained ROSC in patients with OHCA at the scene in a middle-income country. An optimal score of ≥ 6 provides an acceptable area under the ROC curve of 0.759 with sensitivity and specificity of 62.0% and 75.7%, respectively. This predictive score might help CPR comman- ders to prognosticate the outcome of patients with OHCA at the scene. 7. Declarations 7.1. Acknowledgments The authors are grateful to the Navamindradhiraj Univer- sity Research Fund for Pub. We would like to thank the This open-access article distributed under the terms of the Creative Commons Attribution NonCommercial 3.0 License (CC BY-NC 3.0). Downloaded from: https://journals.sbmu.ac.ir/aaem/index.php/AAEM/index 9 Archives of Academic Emergency Medicine. 2023; 11(1): e33 Table 3: Screening performance characteristics of the PRADD-CPR model for predicting the sustained return of spontaneous circulation (ROSC) after out of hospital cardiac arrest in different cutoff points Score Sensitivity Specificity PPV NPV LR+ LR- (95% CI) (95% CI) (95% CI) (95% CI) (95% CI) (95% CI) 1.0 100(96.1–100) 1.8(0.5–4.6) 30.1(25–35.5) 100(39.8–100) 1.02(1–1.04) 0 2.0 100(96.1–100) 3.7(1.6–7.1) 30.5(25.3–36) 100(63.1–100) 1.04(1.01–1.07) 0 2.5 100(96.1–100) 4.1(1.9–7.7) 30.6(25.4–36.1) 100(66.4–100) 1.04(1.01–1.07) 0 3.0 100(96.1–100) 11.5(7.6–16.5) 32.3(26.9–38) 100(86.3–100) 1.13(1.08–1.18) 0 3.5 100(96.1–100) 12.4(8.3–17.5) 32.5(27.1–38.3) 100(87.2–100) 1.14(1.09–1.2) 0 4.0 90.2(82.2–95.4) 32.1(26–38.7) 35.9(29.7–42.5) 88.6(79.5–94.7) 1.33(1.19–1.49) 0.31(0.16–0.58) 4.5 89.1(80.9–94.7) 34.9(28.6–41.6) 36.6(30.3–43.3) 88.4(79.7–94.3) 1.37(1.21–1.54) 0.31(0.17–0.58) 5.0 73.9(63.7–82.5) 60.6(53.7–67.1) 44.2(36.2–52.4) 84.6(78–89.9) 1.87(1.53–2.3) 0.43(0.3–0.62) 5.5 69.6(59.1–78.7) 65.1(58.4–71.4) 45.7(37.3–54.3) 83.5(77.1–88.8) 2(1.59–2.5) 0.47(0.34–0.65) 6.0a 62.0(51.2–71.9) 75.7(69.4–81.2) 51.8(42.1–61.4) 82.5(76.5–87.5) 2.55(1.92–3.38) 0.5(0.38–0.66) 6.5 51.1(40.4–1.7) 79.8(73.9–84.9) 51.6(40.9–62.3) 79.5(73.5–84.6) 2.53(1.82–3.52) 0.61(0.49–0.76) 7.0 39.1(29.1–49.9) 85.3(79.9–9.7) 52.9(40.4–65.2) 76.9(71–82) 2.67(1.77–4.01) 0.71(0.6–0.85) 7.5 33.7(24.2–44.3) 87.6(82.5–91.7) 53.4(39.9–66.7) 75.8(70–80.9) 2.72(1.73–4.29) 0.76(0.65–0.88) 8.0 20.7(12.9–30.4) 95.4(91.7–97.8) 65.5(45.7–82.1) 74(68.5–79) 4.5(2.18–9.3) 0.83(0.75–0.93) 8.5 18.5(11.1–27.9) 96.8(93.5–98.7) 70.8(48.9–87.4) 73.8(68.3–78.8) 5.75(2.47–13.4) 0.84(0.76–0.93) 9.0 12(6.1–20.4) 99.1(96.7–99.9) 84.6(54.6–98.1) 72.7(67.3–77.7) 13(2.95–57.6) 0.89(0.82–0.96) 9.5 5.4(1.8–2.2) 99.5(97.5–100) 83.3(35.9–99.6) 71.4(65.9–76.4) 11.8(1.4–100) 0.95(0.9–1) 10.0 5.4(1.8–12.2) 99.5(97.5–100) 83.3(35.9–99.6) 71.4(65.9–76.4) 11.8(1.4–100) 0.95(0.9–1) Data are presented with 95% confidence interval (CI). LR: likelihood ratio; NPV: negative predictive value; PPV: positive predictive value. a The best threshold value was determined using Youden’s index (Youden index J = Sensitivity + Specificity – 1). paramedics at V-EMS, Faculty of Medicine Vajira Hospital, Navamindradhiraj University, for facilitating in data collec- tion and access in the present study, Gawin Tiyawat, MD., chief of Department of Disaster and Emergency Medical Op- eration, Faculty of Science and Health Technology, Nava- mindradhiraj University, and Chunlanee Sangketchon, MD., deputy dean of Faculty of Science and Health Technology, Navamindradhiraj University, for support and suggestions in the research development and Aniwat Berpan, MD. for sug- gestions on English for the present study. 7.2. Conflict of interest The authors have no conflicting interests to declare. 7.3. Fundings and supports This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. 7.4. Authors’ contribution Conceptualization: Thongpitak Huabbangyang, Chaleamlap Thongpean, Wannakorn Pralomcharoensuk, Weerawan Khaokaen and Sunisa Bumrongchai; Methodology: Thongpitak Huabbangyang, Chaleamlap Thongpean, Wannakorn Pralomcharoensuk, Weerawan Khaokaen and Sunisa Bumrongchai; Software: Thongpitak Huabbangyang; Validation: Thongpitak Huabbangyang; Agasak Silakoon and Pramote Papukdee; Formal analysis: Thongpitak Huabbangyang; Investigation: Thongpitak Huabbangyang, Ratree Chaisorn and Chomkamol Saumok; Resources: Thongpitak Huabbangyang, Chaleamlap Thong- pean, Wannakorn Pralomcharoensuk, Weerawan Khaokaen and Sunisa Bumrongchai; Data Curation: Thongpitak Huabbangyang; Writing – Origi- nal Draft: Thongpitak Huabbangyang; Writing - Review & Editing: Thongpitak Huabbangyang, Agasak Silakoon and Pramote Papukdee; Visualization: Thongpitak Huabbangyang and Rossakorn Klaiangthong; Supervision: Thongpitak Huabbangyang; Project administration: Thongpitak Huabbangyang; Fund- ing acquisition: Thongpitak Huabbangyang. All authors read and approved the final version of manuscript. 7.5. Data Availability The datasets generated and analyzed during the current study are available from the corresponding author on reason- able request. References 1. Gräsner J-T, Lefering R, Koster RW, Masterson S, Böttiger BW, Herlitz J, et al. EuReCa ONE- 27 Nations, ONE Eu- rope, ONE Registry: A prospective one month analysis of This open-access article distributed under the terms of the Creative Commons Attribution NonCommercial 3.0 License (CC BY-NC 3.0). Downloaded from: https://journals.sbmu.ac.ir/aaem/index.php/AAEM/index T. Huabbangyang et al. 10 out-of-hospital cardiac arrest outcomes in 27 countries in Europe. Resuscitation. 2016;105:188-95. 2. Yeeheng U. Factors associated with successful resusci- tation of out-of-hospital cardiac arrest at Rajavithi Hos- pital’s Narenthorn Emergency Medical Service Center, Thailand. Asia Pac J Public Health. 2011;23(4):601-7. 3. Foundation SCA. The following summary by the Sud- den Cardiac Arrest Foundation of selected highlights of the "American Heart Association Heart and Stroke Statis- tics - 2022 Update" focuses on out-of-hospital cardiac ar- rest in the U.S. 2022 [Available from: https://www.sca- aware.org/about-sudden-cardiac-arrest/latest-statistics. 4. Berdowski J, Berg RA, Tijssen JG, Koster RW. Global in- cidences of out-of-hospital cardiac arrest and survival rates: systematic review of 67 prospective studies. Resus- citation. 2010;81(11):1479-87. 5. Huabbangyang T, Soion T, Promdee A, Nguanjinda K, Chamchan A, Chaisorn R, et al. Factors Associated with Successful Resuscitation during Out-of-Hospital Cardiac Arrest Performed By Surgico Medical Ambulance and Rescue Team (SMART), Division of Emergency Medical Service and Disaster, Faculty of Medicine Vajira Hospital, Navamindrad. J Med Assoc Thai. 2021;104(9):1488-96. 6. Tham LP, Wah W, Phillips R, Shahidah N, Ng YY, Do Shin S, et al. Epidemiology and outcome of paediatric out- of-hospital cardiac arrests: a paediatric sub-study of the Pan-Asian resuscitation outcomes study (PAROS). Resus- citation. 2018;125:111-7. 7. Amnuaypattanapon K, Thanachartwet V, Desakorn V, Chamnanchanunt S, Pukrittayakamee S, Sahassananda D, et al. Predictive model of return of spontaneous cir- culation among patients with out-of-hospital cardiac ar- rest in Thailand: The WATCH-CPR Score. Int J Clin Pract. 2020;74(7):e13502. 8. Luo S, Zhang Y, Zhang W, Zheng R, Tao J, Xiong Y. Prognostic significance of spontaneous shockable rhythm conversion in adult out-of-hospital cardiac ar- rest patients with initial non-shockable heart rhythms: a systematic review and meta-analysis. Resuscitation. 2017;121:1-8. 9. Holmberg MJ, Vognsen M, Andersen MS, Donnino MW, Andersen LW. Bystander automated external defibrillator use and clinical outcomes after out-of-hospital cardiac arrest: A systematic review and meta-analysis. Resusci- tation. 2017;120:77-87. 10. Huabbangyang T, Sangketchon C, Ittiphisit S, Uoun K, Saumok C. Predictive Factors of Outcome in Cases of Out-of-hospital Cardiac Arrest Due to Traffic Accident In- juries in Thailand; a National Database Study. Arch Acad Emerg Med. 2022;10(1):e64. 11. Sirikul W, Piankusol C, Wittayachamnankul B, Riyapan S, Supasaovapak J, Wongtanasarasin W, et al. A retrospec- tive multi-centre cohort study: pre-hospital survival fac- tors of out-of-hospital cardiac arrest (OHCA) patients in Thailand. Resusc Plus. 2022;9:100196. 12. Nürnberger A, Sterz F, Malzer R, Warenits A, Girsa M, Stöckl M, et al. Out of hospital cardiac arrest in Vienna: incidence and outcome. Resuscitation. 2013;84(1):42-7. 13. Gräsner J-T, Meybohm P, Lefering R, Wnent J, Bahr J, Messelken M, et al. ROSC after cardiac arrest—the RACA score to predict outcome after out-of-hospital cardiac ar- rest. Eur Heart J. 2011;32(13):1649-56. 14. Ahmad R, Lumley S, Lau YC. NULL-PLEASE: A new ‘Fu- tility score’in the management of survivors of out-of- hospital cardiac arrest. Resuscitation. 2016;106:e83. 15. Liu N, Liu M, Chen X, Ning Y, Lee JW, Siddiqui FJ, et al. Development and validation of an interpretable prehos- pital return of spontaneous circulation (P-ROSC) score for patients with out-of-hospital cardiac arrest using ma- chine learning: A retrospective study. EClinicalMedicine. 2022;48:101422. 16. Von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. The Strengthening the Report- ing of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational stud- ies. Lancet. 2007;370(9596):1453-7. 17. Merchant RM, Topjian AA, Panchal AR, Cheng A, Aziz K, Berg KM, et al. Part 1: executive summary: 2020 Amer- ican Heart Association guidelines for cardiopulmonary resuscitation and emergency cardiovascular care. Circu- lation. 2020;142(16_Suppl_2):S337-S57. 18. Panchal AR, Bartos JA, Cabañas JG, Donnino MW, Drennan IR, Hirsch KG, et al. Part 3: adult ba- sic and advanced life support: 2020 American Heart Association guidelines for cardiopulmonary resuscita- tion and emergency cardiovascular care. Circulation. 2020;142(16_Suppl_2):S366-S468. 19. Peduzzi P, Concato J, Kemper E, Holford TR, Feinstein AR. A simulation study of the number of events per variable in logistic regression analysis. J Clin Epidemiol. 1996;49(12):1373-9. 20. Moosajee US, Saleem SG, Iftikhar S, Samad L. Outcomes following cardiopulmonary resuscitation in an emer- gency department of a low-and middle-income country. Int J Emerg Med. 2018;11(1):40. 21. Stankovic N, Holmberg MJ, Granfeldt A, Andersen LW. Socioeconomic status and risk of in-hospital cardiac ar- rest. Resuscitation. 2022;177:69-77. 22. Ong MEH, Do Shin S, De Souza NNA, Tanaka H, Nishi- uchi T, Song KJ, et al. Outcomes for out-of-hospital car- diac arrests across 7 countries in Asia: The Pan Asian Resuscitation Outcomes Study (PAROS). Resuscitation. 2015;96:100-8. 23. Matsuyama T, Ohta B, Kiyohara K, Kitamura T. Car- This open-access article distributed under the terms of the Creative Commons Attribution NonCommercial 3.0 License (CC BY-NC 3.0). Downloaded from: https://journals.sbmu.ac.ir/aaem/index.php/AAEM/index 11 Archives of Academic Emergency Medicine. 2023; 11(1): e33 diopulmonary resuscitation duration and favorable neu- rological outcome after out-of-hospital cardiac arrest: a nationwide multicenter observational study in Japan (the JAAM-OHCA registry). Crit Care. 2022;26(1):120. 24. Park S, Lee SW, Han KS, Lee EJ, Jang D-H, Lee SJ, et al. Op- timal cardiopulmonary resuscitation duration for favor- able neurological outcomes after out-of-hospital cardiac arrest. Scand J Trauma Resusc Emerg Med. 2022;30(1):5. 25. Shimizu K, Wakasugi M, Kawagishi T, Hatano T, Fuchigami T, Okudera H. Effect of advanced airway management by paramedics during out-of-hospital cardiac arrest on chest compression fraction and return of spontaneous circulation. Open Access Emerg Med. 2021;13:305-10. 26. Wang Y, Zhang Q, Qu GB, Fang F, Dai XK, Yu LX, et al. Ef- fects of prehospital management in out-of-hospital car- diac arrest: advanced airway and adrenaline administra- tion. BMC Health Serv Res. 2022;22(1):546. 27. Montag S, Herdtle S, John S, Lehmann T, Behringer W, Hohenstein C. Association between prehospital FPS and ROSC in adults with OHCA: A retrospective mul- ticenter study using the German Resuscitation Registry and Intubation Registry (FiPS-CPR). Anaesthesiologie. 2022;71(Suppl 2):198-203. 28. Bakhsh A, Alghoribi R, Arbaeyan R, Mahmoud R, Al- ghamdi S, Saddeeg S. Endotracheal Intubation Ver- sus No Endotracheal Intubation During Cardiopul- monary Arrest in the Emergency Department. Cureus. 2021;13(11):e19760. 29. Yeung J, Chilwan M, Field R, Davies R, Gao F, Perkins GD. The impact of airway management on quality of cardiopulmonary resuscitation: an observational study in patients during cardiac arrest. Resuscitation. 2014;85(7):898-904. 30. Association AH. 2005 American Heart Association (AHA) guidelines for cardiopulmonary resuscitation (CPR) and emergency cardiovascular care (ECC) of pediatric and neonatal patients: pediatric basic life support. Pediatrics. 2006;117(5):e989-e1004. 31. Neumar RW, Otto CW, Link MS, Kronick SL, Shus- ter M, Callaway CW, et al. Part 8: adult advanced cardiovascular life support: 2010 American Heart As- sociation guidelines for cardiopulmonary resuscita- tion and emergency cardiovascular care. Circulation. 2010;122(18_suppl_3):S729-S67. 32. Link MS, Berkow LC, Kudenchuk PJ, Halperin HR, Hess EP, Moitra VK, et al. Part 7: adult advanced cardio- vascular life support: 2015 American Heart Associa- tion guidelines update for cardiopulmonary resuscita- tion and emergency cardiovascular care. Circulation. 2015;132(18_suppl_2):S444-S64. 33. Olasveengen TM, Mancini ME, Perkins GD, Avis S, Brooks S, Castrén M, et al. Adult basic life support: 2020 international consensus on cardiopulmonary re- suscitation and emergency cardiovascular care sci- ence with treatment recommendations. Circulation. 2020;142(16_suppl_1):S41-S91. 34. Wongtanasarasin W, Ungrungseesopon N, Phinyo P. Association between Intra-Arrest Blood Glucose Level and Outcomes of Resuscitation at the Emer- gency Department: A Retrospective Study. J Clin Med. 2022;11(11):3067. 35. Wang C-H, Chang W-T, Huang C-H, Tsai M-S, Chou E, Yu P-H, et al. Associations between intra-arrest blood glu- cose level and outcomes of adult in-hospital cardiac ar- rest: a 10-year retrospective cohort study. Resuscitation. 2020;146:103-10. 36. Peluso L, Oddo M, Sandroni C, Citerio G, Taccone FS. Early neurological pupil index to predict outcome after cardiac arrest. Intensive Care Med. 2022;48(4):496-7. 37. Riker RR, Sawyer ME, Fischman VG, May T, Lord C, El- dridge A, et al. Neurological pupil index and pupillary light reflex by pupillometry predict outcome early after cardiac arrest. Neurocrit care. 2020;32:152-61. This open-access article distributed under the terms of the Creative Commons Attribution NonCommercial 3.0 License (CC BY-NC 3.0). Downloaded from: https://journals.sbmu.ac.ir/aaem/index.php/AAEM/index Introduction Methods Results Discussion Limitations Conclusion Declarations References