Title
Author
DOI
Article Type
Special Issue
Volume
Issue
Interaction effect between prehospital hydration and initial cardiac rhythm in traumatic out-of-hospital cardiac arrest: a nationwide observational study
1Department of Public Health Care Service, Seoul National University Bundang Hospital, 13620 Seongnam, Republic of Korea
2Laboratory of Emergency Medical Services, Seoul National University Hospital Biomedical Research Institute, 03080 Seoul, Republic of Korea
3Department Emergency Medicine, College of Medicine, Seoul National University, 03080 Seoul, Republic of Korea
4Department Emergency Medicine, Seoul National University Hospital, 03080 Seoul, Republic of Korea
5Department Emergency Medicine, Seoul Metropolitan Government Boramae Medical Center, 07061 Seoul, Republic of Korea
6Department Emergency Medicine, Seoul National University Bundang Hospital, 13620 Seongnam, Republic of Korea
*Corresponding Author(s):ssberg@snu.ac.kr (Ki Jeong Hong)
| History | Submitted: 15 July 2023 | Accepted: 22 August 2023 | Published: 08 February 2024 |
| Copyright: | ©2024 The Author(s). Published by MRE Press. |
Traumatic cardiac arrest (TCA) is different in etiology compared to medical cardiac arrest. In case of TCA, it is important to initiate early fluid resuscitation. Initial cardiac rhythm serves as an indicator of outcomes in case of cardiac arrest. We aimed to find the association between prehospital hydration and outcomes of TCA according to initial cardiac rhythm. This is a retrospective, observational, cross-sectional study. An examination was undertaken involving patients afflicted with TCA within the timeframe of 2014 to 2019. Exposure was defined to encompass prehospital hydration; interactive exposure was categorized by initial cardiac rhythm (non-shockable vs. shockable); the primary outcome was defined as good neurological status at discharge, whereas the secondary outcome was defined as survival to discharge. Multivariable logistic regression analysis was used to calculate adjusted odds ratios (AORs) with 95% confidence intervals (CIs). A comprehensive analysis was conducted on a cumulative of 20,247 patients. Rates of good neurological status and survival to discharge were 0.2% and 8.3% (non-shockable rhythm group) and 3.0% and 16.7% (shockable rhythm group), respectively. However, rates of good neurological status and survival to discharge were 0.2% and 7.9% (non-prehospital hydration group) and 0.3% and 10.0% (hydration group), respectively. Compared to the non-hydration group, the AORs for good neurological status at discharge was 1.44 (95% CI: 0.77–2.69) for the hydration group. Moreover, compared to the non-shockable rhythm group, the AORs for good neurological status at discharge was 19.74 (95% CI: 10.46–27.26) in the shockable rhythm group. The interaction analysis conducted between prehospital hydration and initial rhythm unveiled the efficacy of prehospital hydration in promoting favorable survival to discharge outcomes in the non-shockable rhythm group. Therefore, prehospital hydration is recommended for those with TCA characterized by a non-shockable rhythm before transport from the incident location.
Cite this article
Dae Kon Kim, Sang Do Shin, Young Sun Ro, Kyoung Jun Song, Joo Jeong, Ki Jeong Hong. Interaction effect between prehospital hydration and initial cardiac rhythm in traumatic out-of-hospital cardiac arrest: a nationwide observational study. Signa Vitae. 2024; 20(2): 27-37. doi: 10.22514/sv.2023.128
1. Introduction
Out-of-hospital cardiac arrest (OHCA) is a leading cause of mortality globally [1]. The majority of OHCA cases are presumed to have a cardiac etiology, whereas traumatic cardiac arrest (TCA) accounts only for a small proportion [2]. However, previously reported mortality rates in cases of TCA were higher than those observed in OHCA cases with a medical etiology [3]. The survival rate of TCA has been reported to range from 1% to 12% according to the research settings [4, 5].
The mechanism of TCA primarily involves decompensated hypovolemic and hypoxic shock after the initial injury [6]. Distinct etiological differences necessitate varied approaches for presumed medical causes and trauma. For instance, strategies include early chest compressions with electro-cardioversion for medical causes, in contrast to addressing acute reversible causes. Reversible causes of TCA encompass several factors. Hypoxia arising from airway obstruction attributed to severe traumatic brain injury (treated by airway protection), hypovolemia due to major vessel injuries or pelvic fractures (treated by hydration, blood products transfusion and hemostasis), and obstructive shock stemming from cardiac tamponade or pneumothorax (treated by needle thoracotomy) [7]. In case of TCA, the utmost priority lies in establishing a substantial intravenous (IV) access point and initiate a rapid transfusion of blood products. However, many prehospital emergency medical services (EMS) encounter challenges in obtaining blood products due to difficulties in storage and transportation involving temperature control, as well as product expiry issues [8]. Therefore, instead of blood transfusion, immediate fluid resuscitation can be administered during the prehospital phase to restore circulating volume and maintain organ perfusion in TCA [9, 10, 11]. However, there remains an insufficiency of comprehensive evidence regarding the effect of prehospital hydration on TCA.
The initial cardiac rhythm serves as a well-known prognostic indicator not only in presumed cardiac OHCA but also in TCA [6, 12, 13]. This significance arises from the recognition that an initial shockable rhythm has been linked to short no-flow time (time between patient collapse and start of cardiopulmonary resuscitation (CPR)) in previous studies [14, 15]. Nevertheless, this concept has been proposed mainly in presumed cardiac arrest cases and has not been investigated in TCA. As aforementioned, differences in cardiac arrest etiologies require different interpretations and strategies even when the initial cardiac rhythm remains consistent.
We hypothesized that prehospital hydration would have a different effect according to initial cardiac rhythms on outcomes of TCA. We aimed to evaluate the association between prehospital hydration and clinical outcomes based on initial cardiac rhythm in cases of TCA.
2. Materials and methods
This study was conducted as a retrospective, observational, cross-sectional study, using the Korean national OHCA registry. South Korea has a population of nearly 50 million and a land of 99,000 km2. The EMS is based on a single-tiered, fire-based and government-sponsored system operated by the National Fire Agency. For emergency assistance, the telephone number of the EMS is “119” across the entire country. EMS providers adhere to the EMS CPR protocol, which is based on the 2020 American Heart Association guidelines. Moreover, this protocol is followed at the scene and during transportation of the patient [16, 17]. A Korean EMS provider holds a role similar to that of an intermediate emergency medical technician (EMT-I) in the United States. These providers can perform CPR with basic life support at the scene and during transportation, coupled with the utilization of automatic external defibrillation. Additionally, IV fluid administration and advanced airway management, such as supraglottic airway and endotracheal intubation performed under direct medical control, are also allowed to be executed by the EMS providers. They can also call doctors from the scene using a smartphone for direct medical oversight. There is no definite protocol or guideline about EMS IV hydration in South Korea. We defer to the judgement of the on-scene EMS providers to determine whether to administer hydration before transportation or while in transit. Patients with OHCA must be transported to an emergency department (ED). Advanced cardiac life support medications are available at an ED; however, their availability is limited in most prehospital areas. Furthermore, in cases involving major trauma, EMS providers are not authorized to perform needle thoracotomy for tension pneumothorax. The nationwide OHCA registry, which includes the basic EMS run sheet, prehospital EMS cardiac arrest registry, and hospital record review, was employed in this study [18, 19, 20, 21]. The EMS run sheet and prehospital EMS cardiac arrest registry, including Utstein factors, such as demographics and prehospital EMS management variables, are recorded by the on-duty EMS provider.
All EMS-treated TCA cases from 2014 to 2019 were initially enrolled for analysis. Patients with OHCA of medical etiology; those with non-traumatic cardiac arrest owing to reasons such as burns, fire and suffocation; those that did not receive CPR at the scene; and those of unknown age, exposures and outcomes were excluded.
The main exposure was defined as “prehospital hydration status at the scene”. Prehospital hydration was considered positive when EMS providers successfully administered IV fluid at the scene. The interactive exposure was characterized as “the initial cardiac rhythm first checked at the scene”. Shockable rhythm included pulseless ventricular tachycardia and ventricular fibrillation, whereas non-shockable rhythm encompassed pulseless electrical activity and asystole. Various factors were considered including Utstein factors, demographic variables, community factors, as well as prehospital and hospital variables. These encompassed age, sex, weekends, EMS call time (day or night), response time interval (RTI), scene time interval (STI), transport time interval (TTI), EMS time interval, place of arrest (public or private), metropolitan area (with a population of more than one million), witnessed status, bystander CPR, mechanism of trauma (traffic accident, fall, blunt, penetrating or gunshot), intent of trauma (accidental or intentional), prehospital airway management (endotracheal intubation, combitube, laryngeal mask airway, king airway, I-gel, others or none), prehospital adrenaline (epinephrine) administration, prehospital restraint (cervical, whole spine, extremity splint or head), prehospital wound management (hemostasis or dressing), multi-tiered response status, referred hospital level (regional level 1 ED, local level 2 ED, local level 3 ED or local level 4), death on arrival at ED, prehospital return of spontaneous circulation (ROSC), and ED ROSC were collected. RTI was characterized as the span between the moment when the emergency call was received by a call-taker in the 119 dispatch center and the point when the ambulance arrived at the OHCA scene. Similarly, STI was defined as the duration commencing from the instance the ambulance arrived at the scene to the time when the ambulance departed from the same location. TTI was defined as the span starting from the point when the ambulance departed the scene to the time when the ambulance arrived at the ED. However, EMS time interval was defined as the duration commencing from the moment the emergency call was received by a call-taker in 119 dispatch center to the time when the ambulance arrived at the ED.
The primary outcome was a good neurological status at discharge, and the secondary outcome was survival to discharge. Good neurological status was constructed to be the attainment of a cerebral performance category (CPC) score of 1 or 2 at hospital discharge.
Patient demographics and several factors related to TCA, such as place of arrest, witness status, bystander, time of arrest, mechanism of trauma, and prehospital EMS management, were compared between the variables of “prehospital hydration” and “initial cardiac rhythm”. The categorical variables were described using counts and proportions and compared using the chi-square test. The continuous variables were presented as the mean and standard deviation (SD) or median and interquartile range (IQR) using the Mann-Whitney U test. Additionally, a multivariable logistic regression analysis was performed to test the association between prehospital hydration and initial cardiac rhythm and the resultant outcomes. Potential confounders, such as age, sex, year, weekend, time of arrest, place of arrest, witness, bystander, mechanism of trauma, intent of trauma, metropolitan and RTI were adjusted. The AORs and 95% CIs were calculated for outcomes. An interaction analysis was performed to compare the effect of the initial cardiac rhythm and the prehospital hydration on the outcomes. All of the analyses were performed using SAS version 9.4 (SAS©, Cary, NC, USA). p-values of 0.05 were considered to be statistically significant.
3. Results
From 181,495 eligible OHCA patients, 20,247 patients were finally assessed after excluding patients with medical etiology (n = 145,787), non-TCA owing to reasons such as burns, fire and suffocation (n = 13,392), those that did not receive CPR at the scene (n = 2036), and those with unknown age (n = 26), exposures (n = 0) and outcomes (n = 7) (Fig. 1).
Fig. 1.Study flow chart. EMS: emergency medical service; OHCA: out-of-hospital cardiac arrest; CPR: cardiopulmonary resuscitation.
A cumulative of 6413 patients, constituting 31.7% of the overall 20,247 patient cohort, received prehospital hydration; 6.4%, 10.7%, 16.3%, 15.8%, 25.5% and 25.3% of the patients were administered prehospital hydration from 2014 to 2019, respectively. The median STI was 7 (IQR 5–10) min for the no prehospital hydration group and 9 (IQR 6–11) min for the prehospital hydration group. TCA occurred in a metropolitan region in 25.5% of patients of the no prehospital hydration group and 44.9% of the prehospital hydration group. The proportion of prehospital airway management was 27.1% in the no prehospital hydration group and 73.3% in the prehospital hydration group. I-gel was the most frequently used in both groups: 14.9% in the no prehospital hydration group and 39.8% in the prehospital hydration. The proportion of patients with initial shockable rhythm was 2.7% in the no prehospital hydration group and 3.0% in the prehospital hydration group. The proportion of patients with outcomes of good neurological status at discharge and survival to discharge were 0.2% and 7.9% in the no prehospital hydration group and 0.3% and 10.0% in prehospital hydration, respectively (Table 1).
| Prehospital hydration status | p-value | |||||||
| Hydration (−) | Hydration (+) | |||||||
| All | % | n | % | n | % | |||
| All | N | 20,247 | 100.0 | 13,834 | 100.0 | 6413 | 100.0 | |
| Year | ||||||||
| 2014 | 3204 | 15.8 | 2792 | 20.2 | 412 | 6.4 | <0.001 | |
| 2015 | 3443 | 17.0 | 2760 | 20.0 | 683 | 10.7 | ||
| 2016 | 3299 | 16.3 | 2252 | 16.3 | 1047 | 16.3 | ||
| 2017 | 3267 | 16.1 | 2252 | 16.3 | 1015 | 15.8 | ||
| 2018 | 3513 | 17.4 | 1880 | 13.6 | 1633 | 25.5 | ||
| 2019 | 3521 | 17.4 | 1898 | 13.7 | 1623 | 25.3 | ||
| Age | ||||||||
| 0∼18 | 766 | 3.8 | 545 | 3.9 | 221 | 3.4 | 0.11 | |
| 19∼65 | 13,491 | 66.6 | 9166 | 66.3 | 4325 | 67.4 | ||
| >66 | 5990 | 29.6 | 4123 | 29.8 | 1867 | 29.1 | ||
| Median (IQR) | 55 (40∼68) | 55 (40∼68) | 54 (40∼67) | 0.30 | ||||
| Sex | Male | 14,573 | 72.0 | 9915 | 71.7 | 4658 | 72.6 | 0.16 |
| Weekend | Yes | 5569 | 27.5 | 3828 | 27.7 | 1741 | 27.1 | 0.44 |
| Daytime | 8A∼8P | 11,853 | 58.5 | 8072 | 58.3 | 3781 | 59.0 | 0.41 |
| RTI (call∼contact) | ||||||||
| ≤4 | 2206 | 10.9 | 1560 | 11.3 | 646 | 10.1 | <0.001 | |
| 4∼8 | 8243 | 40.7 | 5399 | 39.0 | 2844 | 44.3 | ||
| >8 | 8546 | 42.2 | 5783 | 41.8 | 2763 | 43.1 | ||
| Median (IQR) | 8 (6∼12) | 8 (6∼12) | 8 (6∼12) | 0.38 | ||||
| STI (contact∼depart) | ||||||||
| ≤8 | 11,360 | 56.1 | 8229 | 59.5 | 3131 | 48.8 | <0.001 | |
| 8∼16 | 6215 | 30.7 | 3641 | 26.3 | 2574 | 40.1 | ||
| >16 | 1500 | 7.4 | 931 | 6.7 | 569 | 8.9 | ||
| Median (IQR) | 8 (5∼10) | 7 (5∼10) | 9 (6∼11) | <0.001 | ||||
| TTI (depart∼ED arrival) | ||||||||
| ≤6 | 8546 | 42.2 | 5880 | 42.5 | 2666 | 41.6 | 0.35 | |
| 6∼12 | 6450 | 31.9 | 4368 | 31.6 | 2082 | 32.5 | ||
| >12 | 5177 | 25.6 | 3549 | 25.7 | 1628 | 25.4 | ||
| Median (IQR) | 8 (5∼13) | 8 (5∼13) | 8 (5∼13) | 0.06 | ||||
| EMS TI (call∼ED arrival) | ||||||||
| ≤15 | 2524 | 12.5 | 1975 | 14.3 | 549 | 8.6 | <0.001 | |
| 15∼25 | 8035 | 39.7 | 5468 | 39.5 | 2567 | 40.0 | ||
| >25 | 9688 | 47.8 | 6391 | 46.2 | 3297 | 51.4 | ||
| Median (IQR) | 25 (19∼35) | 24 (18∼34) | 26 (20∼35) | <0.001 | ||||
| Distance (km) | Median (IQR) | 3 (1.5∼5.8) | 3 (1.7∼6) | 2.5 (1.5∼5) | <0.001 | |||
| Place of arrest | ||||||||
| Public | 12,297 | 60.7 | 8592 | 62.1 | 3705 | 57.8 | <0.001 | |
| Private | 6191 | 30.6 | 4033 | 29.2 | 2158 | 33.7 | ||
| Others | 1759 | 8.7 | 1209 | 8.7 | 550 | 8.6 | ||
| Metropolitan | 6412 | 31.7 | 3534 | 25.5 | 2878 | 44.9 | <0.001 | |
| Witness | 8505 | 42.0 | 5617 | 40.6 | 2888 | 45.0 | <0.001 | |
| Bystander | 6498 | 32.1 | 4257 | 30.8 | 2241 | 34.9 | <0.001 | |
| EMS AED | 20,055 | 99.1 | 13,661 | 98.7 | 6394 | 99.7 | <0.001 | |
| Mechanism of trauma | ||||||||
| TA | 11,668 | 57.6 | 8265 | 59.7 | 3403 | 53.1 | <0.001 | |
| Fall | 6892 | 34.0 | 4448 | 32.2 | 2444 | 38.1 | ||
| Blunt | 1203 | 5.9 | 817 | 5.9 | 386 | 6.0 | ||
| Penetrating | 473 | 2.3 | 296 | 2.1 | 177 | 2.8 | ||
| Gun shot | 11 | 0.1 | 8 | 0.1 | 3 | 0.0 | ||
| Intent | ||||||||
| Accident | 14,346 | 70.9 | 10,040 | 72.6 | 4306 | 67.1 | <0.001 | |
| Intentional | 3177 | 15.7 | 2079 | 15.0 | 1098 | 17.1 | ||
| Assault | 305 | 1.5 | 200 | 1.4 | 105 | 1.6 | ||
| Prehospital Airway | ||||||||
| None | 10,417 | 51.4 | 8707 | 62.9 | 1710 | 26.7 | <0.001 | |
| Endotracheal Tube | 552 | 2.7 | 197 | 1.4 | 355 | 5.5 | ||
| Combitube | 16 | 0.1 | 6 | 0.0 | 10 | 0.2 | ||
| LMA | 873 | 4.3 | 481 | 3.5 | 392 | 6.1 | ||
| King Airway | 259 | 1.3 | 207 | 1.5 | 52 | 0.8 | ||
| I-gel | 4613 | 22.8 | 2062 | 14.9 | 2551 | 39.8 | ||
| Others | 3234 | 16.0 | 2017 | 14.6 | 1217 | 19.0 | ||
| Prehospital Adrenaline use | 77 | 0.4 | 2 | 0.0 | 75 | 1.2 | <0.001 | |
| Restraint | ||||||||
| Cervical | 15,234 | 75.2 | 9781 | 70.7 | 5453 | 85.0 | <0.001 | |
| Whole spine | 9775 | 48.3 | 5903 | 42.7 | 3872 | 60.4 | <0.001 | |
| Extremity splint | 3651 | 18.0 | 1959 | 14.2 | 1692 | 26.4 | <0.001 | |
| Head | 6504 | 32.1 | 3579 | 25.9 | 2925 | 45.6 | <0.001 | |
| Wound management | ||||||||
| Hemostasis | 5469 | 27.0 | 3153 | 22.8 | 2316 | 36.1 | <0.001 | |
| Dressing | 4009 | 19.8 | 2224 | 16.1 | 1785 | 27.8 | <0.001 | |
| Initial shockable rhythm | 569 | 2.8 | 375 | 2.7 | 194 | 3.0 | 0.21 | |
| Hydration amount | Mean (SD) | 500 (1000) | 0 (0) | 500 (1000) | N/A | |||
| Multi-tiered response | 9350 | 46.2 | 5393 | 39.0 | 3957 | 61.7 | <0.001 | |
| Transported hospital level | ||||||||
| 1 | 4834 | 23.9 | 2918 | 21.1 | 1916 | 29.9 | <0.001 | |
| 2 | 8842 | 43.7 | 5945 | 43.0 | 2897 | 45.2 | ||
| 3 | 5617 | 27.7 | 4193 | 30.3 | 1424 | 22.2 | ||
| 4 | 614 | 3.0 | 526 | 3.8 | 88 | 1.4 | ||
| DOA at ED | 9823 | 48.5 | 7096 | 51.3 | 2727 | 42.5 | <0.001 | |
| Prehospital ROSC | 340 | 1.7 | 179 | 1.3 | 161 | 2.5 | <0.001 | |
| ED ROSC | 2957 | 14.6 | 1895 | 13.7 | 1062 | 16.6 | <0.001 | |
| Good CPC at discharge | 51 | 0.3 | 31 | 0.2 | 20 | 0.3 | 0.25 | |
| Survival to discharge | 1738 | 8.6 | 1094 | 7.9 | 644 | 10.0 | <0.001 | |
| IQR: interquartile range; RTI: response time interval; STI: scene time interval; TTI: transport time interval; EMS TI: emergency medical service time interval; AED: automated external defibrillator; TA: traffic accident; LMA: laryngeal mask airway; SD: standard deviation; DOA: death on arrival; ED: emergency department; ROSC: return of spontaneous circulation; CPC: cerebral performance category. |
The number (percentage) of patients with initial shockable rhythm was 569 (2.8%). The median STI was 8 (IQR 5–10) min in the non-shockable rhythm group and 8 (IQR 5–10.5) min in the shockable rhythm group. The rate of witnessed arrest and bystander CPR were 41.9% and 31.9%, respectively, in the non-shockable rhythm group and 46.7% and 39.7%, respectively, in the shockable rhythm group. The proportions of patients with statuses of death on arrival at ED, prehospital ROSC and ED ROSC were 48.8%, 1.5% and 14.5% in the non-shockable rhythm group and 38.7%, 8.6% and 16.5% in the shockable rhythm group, respectively. The proportion of patients with outcomes of good neurological status at discharge and survival to discharge were 0.2% and 8.3% in the non-shockable rhythm group and 3.0% and 16.7% in the shockable rhythm group, respectively (Table 2).
| Initial cardiac rhythm | p-value | |||||||
| Shockable (−) | Shockable (+) | |||||||
| All | % | n | % | n | % | |||
| All | N | 20,247 | 100.0 | 19,678 | 100.0 | 569 | 100.0 | |
| Year | ||||||||
| 2014 | 3204 | 15.8 | 3100 | 15.8 | 104 | 18.3 | 0.08 | |
| 2015 | 3443 | 17.0 | 3337 | 17.0 | 106 | 18.6 | ||
| 2016 | 3299 | 16.3 | 3197 | 16.2 | 102 | 17.9 | ||
| 2017 | 3267 | 16.1 | 3196 | 16.2 | 71 | 12.5 | ||
| 2018 | 3513 | 17.4 | 3418 | 17.4 | 95 | 16.7 | ||
| 2019 | 3521 | 17.4 | 3430 | 17.4 | 91 | 16.0 | ||
| Age | ||||||||
| 0∼18 | 766 | 3.8 | 743 | 3.8 | 23 | 4.0 | 0.94 | |
| 19∼65 | 13,491 | 66.6 | 13,114 | 66.6 | 377 | 66.3 | ||
| ≥65 | 5990 | 29.6 | 5821 | 29.6 | 169 | 29.7 | ||
| Median (IQR) | 55 (40∼68) | 55 (40∼68) | 54 (40∼66) | 0.58 | ||||
| Sex | Male | 14,573 | 72.0 | 14,150 | 71.9 | 423 | 74.3 | 0.20 |
| Weekend | Yes | 5569 | 27.5 | 5419 | 27.5 | 150 | 26.4 | 0.54 |
| Daytime | 8A∼8P | 11,853 | 58.5 | 11,496 | 58.4 | 357 | 62.7 | 0.04 |
| RTI (call∼contact) | ||||||||
| ≤4 | 2206 | 10.9 | 2129 | 10.8 | 77 | 13.5 | 0.05 | |
| 4∼8 | 8243 | 40.7 | 8020 | 40.8 | 223 | 39.2 | ||
| >8 | 8546 | 42.2 | 8328 | 42.3 | 218 | 38.3 | ||
| Median (IQR) | 8 (6∼12) | 8 (6∼12) | 8 (5∼11) | 0.01 | ||||
| STI (contact∼depart) | ||||||||
| ≤8 | 11,360 | 56.1 | 11,064 | 56.2 | 296 | 52.0 | 0.45 | |
| 8∼16 | 6215 | 30.7 | 6033 | 30.7 | 182 | 32.0 | ||
| >16 | 1500 | 7.4 | 1459 | 7.4 | 41 | 7.2 | ||
| Median (IQR) | 8 (5∼10) | 8 (5∼10) | 8 (5∼10.5) | 0.49 | ||||
| TTI (depart∼ED arrival) | ||||||||
| ≤6 | 8546 | 42.2 | 8316 | 42.3 | 230 | 40.4 | 0.4 | |
| 6∼12 | 6450 | 31.9 | 6272 | 31.9 | 178 | 31.3 | ||
| >12 | 5177 | 25.6 | 5018 | 25.5 | 159 | 27.9 | ||
| Median (IQR) | 8 (5∼13) | 8 (5∼13) | 8 (5∼13.5) | 0.22 | ||||
| EMS TI (call∼ED arrival) | ||||||||
| ≤15 | 2524 | 12.5 | 2446 | 12.4 | 78 | 13.7 | 0.06 | |
| 15∼25 | 8035 | 39.7 | 7836 | 39.8 | 199 | 35.0 | ||
| >25 | 9688 | 47.8 | 9396 | 47.7 | 292 | 51.3 | ||
| Median (IQR) | 25 (19∼35) | 25 (19∼35) | 26 (19∼34) | 0.98 | ||||
| Distance (km) | Median (IQR) | 3 (1.5∼5.8) | 3 (1.6∼5.9) | 2.9 (1.5∼5) | 0.10 | |||
| Place of arrest | ||||||||
| Public | 12,297 | 60.7 | 11,940 | 60.7 | 357 | 62.7 | 0.49 | |
| Private | 6191 | 30.6 | 6030 | 30.6 | 161 | 28.3 | ||
| Others | 1759 | 8.7 | 1708 | 8.7 | 51 | 9.0 | ||
| Metropolitan | 6412 | 31.7 | 6227 | 31.6 | 185 | 32.5 | 0.66 | |
| Witness | 8505 | 42.0 | 8239 | 41.9 | 266 | 46.7 | 0.05 | |
| Bystander | 6498 | 32.1 | 6272 | 31.9 | 226 | 39.7 | <0.001 | |
| EMS AED | 20,055 | 99.1 | 19,486 | 99.0 | 569 | 100.0 | 0.02 | |
| Mechanism of trauma | ||||||||
| TA | 11,668 | 57.6 | 11,338 | 57.6 | 330 | 58.0 | 0.13 | |
| Fall | 6892 | 34.0 | 6694 | 34 | 198 | 34.8 | ||
| Blunt | 1203 | 5.9 | 1168 | 5.9 | 35 | 6.2 | ||
| Penetrating | 473 | 2.3 | 468 | 2.4 | 5 | 0.9 | ||
| Gun shot | 11 | 0.1 | 10 | 0.1 | 1 | 0.2 | ||
| Intent | ||||||||
| Accident | 14,346 | 70.9 | 13,928 | 70.8 | 418 | 73.5 | 0.08 | |
| Intentional | 3177 | 15.7 | 3104 | 15.8 | 73 | 12.8 | ||
| Assault | 305 | 1.5 | 301 | 1.5 | 4 | 0.7 | ||
| Prehospital Airway | ||||||||
| None | 10,417 | 51.4 | 10,134 | 51.5 | 283 | 49.7 | 0.03 | |
| Endotracheal Tube | 552 | 2.7 | 534 | 2.7 | 18 | 3.2 | ||
| Combitube | 16 | 0.1 | 15 | 0.1 | 1 | 0.2 | ||
| LMA | 873 | 4.3 | 838 | 4.3 | 35 | 6.2 | ||
| King Airway | 259 | 1.3 | 245 | 1.2 | 14 | 2.5 | ||
| I-gel | 4613 | 22.8 | 4495 | 22.8 | 118 | 20.7 | ||
| Others | 3234 | 16.0 | 3138 | 15.9 | 96 | 16.9 | ||
| Prehospital Adrenaline Use | 77 | 0.4 | 75 | 0.4 | 2 | 0.4 | 0.91 | |
| Restraint | ||||||||
| Cervical | 15,234 | 75.2 | 14,823 | 75.3 | 411 | 72.2 | 0.09 | |
| Whole spine | 9775 | 48.3 | 9519 | 48.4 | 256 | 45.0 | 0.11 | |
| Extremity splint | 3651 | 18.0 | 3572 | 18.2 | 79 | 13.9 | 0.01 | |
| Head | 6504 | 32.1 | 6333 | 32.2 | 171 | 30.1 | 0.28 | |
| Wound management | ||||||||
| Hemostasis | 5469 | 27.0 | 5346 | 27.2 | 123 | 21.6 | 0.003 | |
| Dressing | 4009 | 19.8 | 3923 | 19.9 | 86 | 15.1 | 0.004 | |
| Prehospital Hydration | ||||||||
| Yes | 6413 | 31.7 | 6219 | 31.6 | 194 | 34.1 | 0.21 | |
| Mean (SD) | 500 (1000) | 0 (0) | 500 (1000) | 0.02 | ||||
| Multi-tiered response | 9350 | 46.2 | 9112 | 46.3 | 238 | 41.8 | 0.03 | |
| Transported hospital level | ||||||||
| 1 | 4834 | 23.9 | 4704 | 23.9 | 130 | 22.8 | 0.91 | |
| 2 | 8842 | 43.7 | 8592 | 43.7 | 250 | 43.9 | ||
| 3 | 5617 | 27.7 | 5453 | 27.7 | 164 | 28.8 | ||
| 4 | 614 | 3.0 | 596 | 3.0 | 18 | 3.2 | ||
| DOA at ED | 9823 | 48.5 | 9603 | 48.8 | 220 | 38.7 | <0.001 | |
| Prehospital ROSC | 340 | 1.7 | 291 | 1.5 | 49 | 8.6 | <0.001 | |
| ED ROSC | 2957 | 14.6 | 2863 | 14.5 | 94 | 16.5 | 0.19 | |
| Good CPC at discharge | 51 | 0.3 | 34 | 0.2 | 17 | 3.0 | <0.001 | |
| Survival to discharge | 1738 | 8.6 | 1643 | 8.3 | 95 | 16.7 | <0.001 | |
| IQR: interquartile range; RTI: response time interval; STI: scene time interval; TTI: transport time interval; EMS TI: emergency medical service time interval; AED: automated external defibrillator; TA: traffic accident; LMA: laryngeal mask airway; SD: standard deviation; DOA: death on arrival; ED: emergency department; ROSC: return of spontaneous circulation; CPC: cerebral performance category. |
Compared with the no prehospital hydration group, the unadjusted ORs and AORs (95% CIs) for good neurological status at discharge of the prehospital hydration group were 1.39 (0.79–2.45) and 1.44 (0.77–2.69), respectively, and those for survival to discharge were 1.30 (1.17–1.44) and 1.32 (1.18–1.48), respectively. Compared with the non-shockable rhythm group, the unadjusted ORs and AORs (95% CIs) of the shockable rhythm group for good neurological status at discharge were 17.79 (9.88–32.05) and 19.74 (10.46–37.26), respectively, and for survival to discharge, they were 2.20 (1.76–2.76) and 2.06 (1.61–2.63), respectively (Table 3).
| Outcomes | ||||||
| All | n | % | Unadjusted OR (95% CI) | Adjusted OR (95% CI)* | ||
| Good CPC at discharge | ||||||
| Hydration (−) | 13,834 | 31 | 0.2 | 1.00 | 1.00 | |
| Hydration (+) | 6413 | 20 | 0.3 | 1.39 (0.79∼2.45) | 1.44 (0.77∼2.69) | |
| Total | 20,247 | 51 | 0.3 | |||
| Shockable (−) | 19,678 | 34 | 0.2 | 1.00 | 1.00 | |
| Shockable (+) | 569 | 17 | 3.0 | 17.79 (9.88∼32.05) | 19.74 (10.46∼37.26) | |
| Total | 20,247 | 51 | 0.3 | |||
| Survival to discharge | ||||||
| Hydration (−) | 13,834 | 1094 | 7.9 | 1.00 | 1.00 | |
| Hydration (+) | 6413 | 644 | 10.0 | 1.30 (1.17∼1.44) | 1.32 (1.18∼1.48) | |
| Total | 20,247 | 1738 | 8.6 | |||
| Shockable (−) | 19,678 | 1643 | 8.3 | 1.00 | 1.00 | |
| Shockable (+) | 569 | 95 | 16.7 | 2.20 (1.76∼2.76) | 2.06 (1.61∼2.63) | |
| Total | 20,247 | 1738 | 8.6 | |||
| OR: odds ratio; CI: confidence interval; CPC: cerebral performance category. *Adjusted for age, sex, year, weekend, daytime, place of arrest, witness, bystander, mechanism of injury, intent, metropolitan and response time interval. |
In the interaction analysis for good neurological status at discharge according to initial rhythm, AORs (95% CIs) of the prehospital hydration group were 1.49 (0.70–3.17) among those with non-shockable rhythm and 1.08 (0.36–3.30) among those with shockable rhythm. AORs (95% CIs) for survival to discharge of prehospital hydration group were significantly different according to initial cardiac rhythm: 1.27 (1.13–1.42) in the non-shockable rhythm group and 1.52 (0.93–2.47) in the shockable rhythm group (Table 4).
| Outcomes | Shockable (−) | Shockable (+) | |||||
| Good CPC at discharge | AOR | 95% CI | AOR | 95% CI | |||
| Hydration (−) | 1.00 | 1.00 | |||||
| Hydration (+) | 1.49 | 0.70 | 3.17 | 1.08 | 0.36 | 3.30 | |
| Survival to discharge | *AOR | 95% CI | AOR | 95% CI | |||
| Hydration (−) | 1.00 | 1.00 | |||||
| Hydration (+) | 1.27 | 1.13 | 1.42 | 1.52 | 0.93 | 2.47 | |
| AOR: adjusted odds ratio; CI: confidence interval; CPC: cerebral performance category. *Adjusted for age, sex, year, weekend, daytime, place of arrest, witness, bystander, mechanism of injury, intent, metropolitan and response time interval. |
4. Discussion
We found that prehospital hydration was associated with higher survival to discharge rate in patients with TCA with initial non-shockable rhythm. However, prehospital hydration had no effect, regardless of initial cardiac rhythm, on good neurological status at discharge. To the best of our knowledge, this is the first study to evaluate the interaction analysis between prehospital hydration and initial cardiac rhythm in TCA.
TCA can be caused by several reversible factors, such as hypoxia, tension pneumothorax, cardiac tamponade or hypovolemia [22]. In this study, hypovolemia was the cause of interest. It is known that TCA due to hypovolemia does not occur immediately after traumatic events. As the patient’s blood volume decreases, blood flow is diverted to vital organs, such as the brain and heart. Lactic acid accumulates within cells over time due to organ ischemia, leading to TCA [23]. In a previous study, Xavier et al. [14] have stated that shockable rhythms are associated with no-flow duration. Considering the pathophysiology of TCA, initial non-shockable rhythms indicate prolonged no-flow duration and less intravascular volume. We think that prehospital hydration was effective in patients with TCA with non-shockable rhythm because it restored effective circulatory volume, which led to an increase of survival to discharge rate. In future research, the relationship between prehospital fluid amounts and electrocardiogram rhythms will need to be analyzed.
TCA is a time-dependent condition that requires a well-coordinated chain of events for survival from prehospital management to specialized trauma center care [24, 25]. TCA management must focus on the correction of reversible causes soon after prioritized life support has been administered. Herein, we selected patients with blunt and penetrating trauma, which was consistent with other studies on this topic [26, 27, 28, 29]. Prehospital hydration has been considered as one of the basic treatment factors to restore circulating blood volume in previous studies [30, 31]. However, Evans et al. [32] have stated that prehospital hydration—including IV and intraosseous administration—was ineffective for survival to discharge. Contrarily, our results show that prehospital hydration was associated with higher survival to discharge after adjusting for confounders in the logistic regression analysis. This may be attributed to the fact that the database used in Evans et al.’s [32] study was the North American EMS system, where paramedics could select which patients to provide interventions based on existing protocols and clinical judgement. In our study, the EMS providers should perform high-quality CPR to patients with all causes of arrest, unless the signs of death (i.e., decapitation, trunk amputation or rigor mortis) were evident. This selection bias could have led to the contrary results. Another reason may be that there is a difference in the EMS provider procedure levels between different countries. In our study, the EMS providers could perform advanced airway management techniques and administer fluid intravenously; however, they could not perform needle thoracotomy and blood product transfusion, which is possible in some North American regions. These differences in prehospital procedure level may have functioned as confounders in logistic regression and affected the outcomes. Further studies are therefore required to determine the effectiveness of prehospital hydration in patients with TCA.
The mechanism of injury in this study was mainly blunt trauma, including traffic accidents, falls and blunt injuries, consisting of up to 97.5% of all TCAs. In a previous study, the proportions of cases with blunt trauma and gunshot injuries were nearly 67% and 25%, respectively [32]. As civilian gun ownership is legally prohibited in South Korea, there have been rare instances of TCAs caused by firearm-related penetrating injuries. Keizer et al. [33] found that penetrating injuries required a higher rate of surgical intervention in major trauma cases compared to blunt injuries, even though there were no differences in mortality and complication rates. This implies that differences in the mechanism of injuries contribute to different pathophysiologies and consequently require different treatment plans. The homogenous mechanism of injury in our study population requires caution in terms of interpretation of the results. Further research requires stratified analysis that includes the mechanism of injury.
In our study, IV hydration was not administered in 68.3% of TCA cases (Table 1). The proportion of those who received hydration increased from 6.4% in 2014 to 25.3% in 2019 because a multi-tiered response (MTR) protocol was started with a pilot study in 2013 and has been implemented nationwide since 2015 [34]. The implementation of the MTR protocol has been accompanied by education on procedures, which was mandatory for EMS providers, including IV hydration and airway management. The prehospital procedure affects STI, and in this study, we found a median of a 2 min delay of departure of the ambulance from the scene in the hydration group. The optimal STI in TCA is still controversial, but the AOR of the hydration group on survival to discharge may allude to the fact that prehospital hydration is required, even though it delays transportation to the hospital from the scene. In case of poor peripheral condition because of extremity fractures or collapsed venous vessels, a maximum of 2 min of additional IV hydration attempts on the scene can be acceptable according to this study.
The effect of shockable rhythm as a good prognostic factor in OHCA of all causes was observed again in our study. Unlike prehospital hydration, initial shockable rhythm was associated with higher outcomes in both survival to discharge and good neurological status. However, the effect of initial shockable rhythm in TCA needs further discussion because it is difficult to distinguish true TCA from a case of major trauma followed by medical cardiac arrest. For example, if the driver collapsed while driving because of acute coronary syndrome, the resulting diagnosis will be major trauma caused by road traffic injury. In-depth surveys, such as injury severity score and evaluation of first injury severity by EMS providers on the scene, must be reviewed together to confirm true TCA. In the interaction analysis, prehospital hydration was not associated with better outcomes in shockable rhythm. This implies that early transportation is recommended for patients with initial shockable rhythm, rather than spending time on the scene in the prehospital phase. Further research must be followed by randomized control trials to determine the effect of prehospital hydration in patients with TCA with shockable rhythm.
This study had some limitations. First, patient data from 2020 to 2022 were not included because during this period, prehospital EMS protocol was affected by the coronavirus disease pandemic, and inclusion of this data may have introduced a bias. Second, the prehospital hydration amount was not fully recorded in many cases; although it was recorded in the EMS run sheet, it was not easy to record CPR situations exactly on the scene while performing high-quality CPR. Furthermore, the EMS run sheet was routinely recorded after the transport is over. A possible recall bias could have been incorporated by the EMS providers in recording exact prehospital hydration amounts, which in turn could have affected the outcomes. Third, this nationwide OHCA registry does not specify trauma information and therefore lacks information on injury severity scores. Further research must include the injury severity score as a potential confounder in association analyses. Fourth, data on post-arrest care and in-hospital treatment, such as hemodynamic support, emergency operation, angio-intervention and targeted temperature management, is missing; these interventions might have affected the outcomes. Fifth, the traumatic etiology (e.g., hypovolemia, hypoxia, tension pneumothorax and cardiac tamponade), which caused TCA, was not clear in this database. Further research requires association analysis according to specified populations. Sixth, this study setting involved a nationally operated EMS by the National Fire Agency of Korea. CPR protocols and available medications at the prehospital stage were different compared to those of other countries, as these are determined according to the local EMS act. The difference in the EMS environment along with that of other resources can limit the generalizability of our findings.
5. Conclusions
The interaction analysis between prehospital hydration and initial cardiac rhythm revealed that prehospital hydration was effective in improving the survival to discharge outcome in patients with TCA with non-shockable rhythm. Prehospital hydration was not associated with better survival to discharge in patients with shockable rhythm and good neurological status at discharge, regardless of the initial cardiac rhythm. Therefore, prehospital hydration is recommended for patients with TCA with non-shockable rhythm before transport from the scene. For those with shockable rhythm, early transportation is recommended.
Abbreviations
OHCA, Out-of-hospital cardiac arrest; TCA, Traumatic cardiac arrest; IV, Intravenous; EMS, Emergency medical services; CPR, Cardiopulmonary resuscitation; ED, Emergency department; RTI, Response time interval; STI, Scene time interval; TTI, Transport time interval; ROSC, Return of spontaneous circulation; SD, Standard deviation; IQR, Interquartile range; AOR, Adjusted odds ratio; CI, Confidence interval; MTR, Multi-tiered response.
Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.
Author contributions
DKK, YSR and SDS—designed the research study. DKK and KJH—performed the formal analysis and wrote the manuscript. KJS and JJ—performed the data curation, validation and methodology. SDS and YSR—were responsible for the project administration. All of the authors contributed to editorial changes in the manuscript and read and approved the final manuscript.
Ethics approval and consent to participate
This study was approved by the Seoul National University Hospital Institutional Review Board with a waiver of informed patient consent (IRB No. 1103-153-357).
Acknowledgment
The authors thank Jeong Ho Park (Department of emergency medicine, Seoul National University Hospital, South Korea) for providing statistical support for research analysis. The authors also thank Tae Han Kim (Department of emergency medicine, Seoul Metropolitan Government Boramae Medical Center, South Korea) for the help of data acquisition.
Funding
This research received no external funding.
Conflict of interest
The authors declare no conflict of interest.
