Signa Vitae. 2022; 18(3): 91-100. doi: 10.22514/sv.2021.245
Original Research

Characteristics of patients with carbon monoxide poisoning due to smoke inhalation and pre-hospital factors related to intensive care unit admission of these patients: a nationwide observational study

Sung Min Lee1, Dong Hun Lee1, Jun Ho Han1, Hyoung Youn Lee1, Tag Heo1,2,*,

*Corresponding Author(s):terran034@gmail.com (Tag Heo)

History Submitted: 18 July 2021 | Accepted: 08 September 2021 | Published: 08 May 2022
Copyright:  ©2022  The Author(s). Published by MRE Press.
This is an open access article under the CC BY 4.0 license (https://creativecommons.org/licenses/by/4.0/).

Collapse table of contents

Abstract

This study aimed to investigate the pre-hospital clinical status of patients with carbon monoxide (CO) poisoning by smoke inhalation and the pre-hospital factors associated with these patients’ admission to the intensive care unit (ICU). In this observational study from January 2016 to December 2018, the National Fire Agency’s first aid activity log on patients with smoke inhalation was matched with National Emergency Department Information System’s patient data with CO poisoning and further analyzed retrospectively. Multiple logistic regression analysis was conducted to identify the relevant pre-hospital associative factors for the decision to admit a patient with CO poisoning to the ICU. Of the 4422 patients with CO poisoning included in the study, 358 (8.09%) were admitted to the ICU. In such patients transported by pre-hospital emergency medical services, age (odds ratio [OR], 1.020; 95% confidence interval [CI], 1.010–1.029), verbal (OR, 3.564; 95% CI, 2.390–5.315), pain (OR, 4.011; 95% CI, 2.661–6.045), unconsciousness (OR, 5.728; 95% CI, 2.708–12.113), SBP (OR, 0.979; 95% CI, 0.969–0.989), HR (OR, 1.011; 95% CI, 1.004–1.018), SpO2 (OR, 0.965; 95% CI, 0.946–0.985), O2 supply (OR, 1.725; 95% CI, 1.143–2.603), use of nasal prongs (OR, 0.504; 95% CI, 0.281–0.905), and intentional inhalation (OR, 2.282; 95% CI, 1.659–3.139) were independently associated with ICU admission. Our study demonstrated that age, mental change, SBP, HR, SPO2, O2 supply, use of nasal prongs, and intentional inhalation in patients with CO poisoning were associated with their ICU admission.

Keywords:Smoke inhalation;Carbon monoxide poisoning;Pre-hospital emergency medical services;Intensive care unit
PDF(821.98 KB)|EndNote (RIS)|BibTeX|RefMan|RefWorks

Cite this article

Sung Min Lee, Dong Hun Lee, Jun Ho Han, Hyoung Youn Lee, Tag Heo. Characteristics of patients with carbon monoxide poisoning due to smoke inhalation and pre-hospital factors related to intensive care unit admission of these patients: a nationwide observational study. Signa Vitae. 2022; 18(3): 91-100. doi: 10.22514/sv.2021.245

1. Introduction

Patients inhaling smoke due to fire, explosion, or other accidents are exposed to various toxic substances. Substances having the most serious influence during smoke inhalation include carbon monoxide (CO) and cyanide (CN) [1, 2]. Most people worldwide die or have sequelae due to CO poisoning following smoke inhalation [3, 4]. Hypoxia is the leading cause of pre-hospital death for accidents such as fires or explosions, by adversely affecting the oxygen transport and utilization systems at the intracellular level owing to toxic substances such as CO or CN, rather than flames or burns [5, 6]. The only similarity between CO and CN is their ability to avidly bind iron ions in the different hemoproteins. The mechanisms of these compounds vary in other aspects. CO impairs the ability of erythrocytes to transfer O2 while CN does not. Acute CO poisoning primarily alters neurological cognitive function and even consciousness without significant alteration of vital functions, while CN primarily alters consciousness and leads to an early onset of severe alterations of vital functions. O2 is the only treatment available for CO poisoning, while antidotes in combination with O2 are more efficient than O2 alone in the treatment of CN poisoning [7]. In most cases, hypoxia leads to ischemic injury of oxygen-sensitive organs, and in particular, damages the brain and heart [8, 9, 10, 11]. In the United States, there are 50,000 carbon monoxide poisoning cases per year and 1000–2000 deaths per year, and a study from Utah in the United States reports that 17.3 per 100,000 cases per year occur [12, 13]. In Europe, 2.24 cases per 100,000 deaths per year were reported [14]. A study from Taiwan, reported 1.13 deaths per 10,000 people per year due to CO poisoning [15]. In the Republic of Korea, according to a report published in 2019 by the Ministry of Health and Welfare, the number of CO poisoning cases increased from 5300 in 2016 to 7152 in 2018, with 1.25 incidences per 10,000 cases. The mortality rate from CO poisoning was declared as 1.3% [16].

Early removal of carboxyhemoglobin (COHb) from the body improves prognosis in CO poisoning [8]. The half-life of CO by spontaneous respiration, normobaric oxygen therapy, and hyperbaric oxygen (HBO2) therapy are 5–6 h, 74 ± 25 min, and 2–30 min, respectively. The first intervention implemented by rescuers and in-hospital physicians can largely influence the time needed to remove CO from the body [17, 18, 19]. Moreover, the protean and non-specific clinical manifestations of CO poisoning can lead to misdiagnosis even experienced medical staff [20]. General guidelines for exposure and poisoning of hazardous substances are available within the standard guidelines for on-site first-aid for 119 emergency personnel of the National Fire Agency (NFA); however, specialized guidelines for smoke inhalation or CO intoxication in patients in the pre-hospital setting are lacking [21]. Previous studies have investigated the clinical characteristics and prognosis of patients with CO poisoning at the in-hospital stage. However, comprehensive research regarding the treatment of patients with CO poisoning by smoke inhalation at the pre-hospital stage and the clinical outcomes of such patients after hospital admission are unavailable [22, 23, 24].

In this study, the initial clinical status of the patients, the status of first aid by NFA personnel, and the clinical status and prognosis of patients at the hospital stage were analyzed. This study is aimed at identifying the major pre-hospital prognostic factors in CO poisoning which are associated to the need for admission to the Intensive Care Unit (ICU). Such analysis could provide guidance in developing new specialized first-aid guidelines for optimizing classification, treatment, and transfer of CO-poisoned patients, thereby improving their prognosis.

2. Materials and methods

2.1 Study design and setting

This study was conducted retrospectively based on the matched data among adults (≥18 years) who were transported to the emergency department (ED) after being checked for smoke inhalation in the first-aid activity log of the NFA from January 1, 2016 to December 31, 2018, and from the NEDIS data on patients diagnosed with CO poisoning [code (T58)] in the ED. The present study has been approved by the Chonnam National University Hospital Institutional Review Board (CNUH-EXP-2019-147). We examined the data covering anonymous patient visits; hence, no informed consent was required.

2.2 Data collection and analysis

The NFA first-aid activity log data has been collected nationwide since 1998. The NEDIS, which currently registers nationwide emergency medical information, has been in place in the EDs since June 2003. However, there is no official linkage between these data. In this study, we merged 10791 cases of NFA first-aid activity log data (No. 5901565) and 18802 cases of NEDIS data (N20191720715). The variables used in the merge were the date and time of ED visit, sex, and age of the patients, and the date and time of transfer to the hospital, sex, and age in the NFA first-aid activity log. The following pre-hospital data were analyzed: age, sex, location of smoke inhalation, reasons for smoke inhalation, medical history, mental state of the scene, systolic blood pressure (SBP), diastolic blood pressure (DBP), heart rate (HR), respiration rate, arterial O2 saturation by pulse oximetry (SpO2), body temperature, blood sugar, NFA first-aid, distance from the scene to the ED, time from call to scene arrival, time from patient contact to scene departure, time from scene departure to ED arrival, the total time from call to ED arrival, the result of a patient transfer by NFA following smoke inhalation, and the need for multiplace hyperbaric chamber [10, 25, 26, 27]. The following hospital data were analyzed: age, sex, initial mental state in the ED, initial vital sign in the ED, Korea Triage Acuity Score (KTAS) (KTAS was implemented in our ED in May 2016; it is based on the Canadian Triage and Acuity Scale with modifications to adapt it to the Korean emergency medical environment. The KTAS consists of a five-level system [level 1, resuscitation; level 2, emergent; level 3, urgent; level 4, less urgent; and level 5, non-urgent] that classifies patients using a combination of variables, including vital signs and chief complaints), HBO2 therapy in the ED, HBO2 therapy after hospital admission, and the result of ED admission. First, we analyzed the epidemiological and clinical status of patients following smoke inhalation in the pre-hospital stage, and the status of first-aid treatment, along with the status in the ED and patient prognosis in the hospital stage. Second, we analyzed the factors related to ICU admission by comparing the ICU admission group and combined group of patients admitted to the general ward and those discharged as a result of ED prognosis.

2.3 Statistical analysis

Continuous variables were presented as means with standard deviations as all continuous variables had a normal distribution. Categorical variables were presented as frequencies with percentages. Continuous variables between independent groups were compared using the two-sample t-test with unequal variances, while categorical variables were compared using the chi-square or Fisher’s exact tests as appropriate. To determine the relevant pre-hospital factors for the decision to admit a patient with smoke inhalation to the ICU, we constructed a series of multiple logistic regression models. In the first of these, we included variables that were univariately predictive of the ICU admission at the 0.05 significance level. Multicollinearity was assessed, and none of the variables had a variance inflation factor of >5. The results of the logistic regression analyses were reported as odds ratios (ORs) with its corresponding 95% confidence intervals (CIs). Data were analyzed using Stata/SE version 16.1 software for Windows (StataCorp, College Station, Texas, USA). A two-sided p-value of 0.05 was considered significant.

3. Results

A total of 5011 adult patients aged >18 years were diagnosed with CO poisoning after being transferred to the ED by NFA Emergency Medical Service (EMS) in the Republic of Korea following smoke inhalation for three years from January 2016 to December 31, 2018. A total of 3355 patients were discharged following ED treatment, 709 patients were admitted to the general ward, 358 patients were admitted to the ICU, 38 were dead on arrival, and 551 patients were transferred to other hospitals. The last two cases (patients who were transferred to other hospitals and patients who died) were excluded, and finally, 4422 patients were included in this study (Fig. 1).

Patients with smoke inhalation between January 2016 and December 2018. NFA, National fire agency; NEDIS, National emergency department information system; ICU, intensive care unit.

Fig. 1.Patients with smoke inhalation between January 2016 and December 2018. NFA, National fire agency; NEDIS, National emergency department information system; ICU, intensive care unit.

3.1 Pre-hospital stage characteristics of patients with CO poisoning due to smoke inhalation

The average age of patients following smoke inhalation was 39.49 (±18.42) years, and that of the ICU patients was 43.83 (±16.67) years (Table 1). Overall, 2474 (55.95%) were males, and 1948 (44.05%) were females, while in the ICU-admitted group, 62.57% were males. The main places where smoke inhalation took place were private houses, roads, commercial facilities, and gathering facilities (military, dormitory, and prison). In addition, the proportion of patients who inhaled smoke from homes, roads, and sea/river/mountains in the ICU admission group was relatively higher than in the other groups. Smoke inhalation was intentional in 1232 (31.79%) patients and unintentional in 2463 (68.21%) patients. However, in the ICU group, 198 (63.26%) patients experienced intentional smoke inhalation, almost twice as high as that of the unintentional group, 115 (36.74%). The proportion of patients with cerebrovascular disease was higher in the ICU admission group. The mental status at the scene was alert, verbal, in pain, and unconscious for 3736 (84.52%), 296 (6.7%), 290 (6.56%), and 98 (2.22%), respectively. The vital signs showed that both SBP and DBP were lower, and HR was higher in the ICU group. SpO2 was also found to be lower in the ICU group. In the first-aid by NFA paramedics, the need for airway manual manipulation and the use of the oropharyngeal device were more frequent in the ICU group. The frequency of O2-supply implementation and the average provided flow rate of O2 were also higher in this group. The use of nasal prongs was less common in the ICU group, while the use of a non-rebreather mask and bag valve mask was more frequent. The rate of supraglottic airway and intubation was also higher in this group. The need for ECG check and for intravenous fluid supply were also more frequent in the ICU group, as were the rate of cases receiving medical direction in the field, and CPR.

Table 1.A comparative analysis of factors related to pre-hospital demographic and clinical status between the combined general ward admission and discharged group, and ICU group in patients with carbon monoxide poisoning due to smoke inhalation.
Carbon monoxide poisoning due to smoke inhalationp-value
Total (n = 4422)Ward + Discharge (n = 4064)Intensive care unit (n = 358)
Age, mean (SD), y39.49 (18.42)39.10 (18.52)43.83 (16.67)<0.001
Sex, n (%)0.008
Male2474 (55.95)2250 (55.36)224 (62.57)
Female1948 (44.05)1814 (44.64)134 (37.43)
Location of smoke inhalation, n (%)<0.001
House2694 (60.92)2467 (60.7)227 (63.41)
Commercial facility487 (11.01)457 (11.25)30 (8.38)
Road230 (5.2)194 (4.77)36 (10.06)
Factory/industry/construction facility193 (4.36)185 (4.55)8 (2.23)
Dormitory/military/prison facility187 (4.23)175 (4.31)12 (3.35)
Medical facility85 (1.92)80 (1.97)5 (1.4)
Sea/river/mountain54 (1.22)46 (1.13)8 (2.23)
Others492 (11.13)460 (11.32)32 (8.94)
Reasons for smoke inhalation, n (%)38753562313<0.001
Unintentional2643 (68.21)2528 (70.97)115 (36.74)
Intentional1232 (31.79)1034 (29.03)198 (63.26)
Medical History, n (%)
Hypertension298 (6.74)268 (6.59)30 (8.38)0.196
Diabetes mellitus142 (3.21)132 (3.25)10 (2.79)0.640
Cerebrovascular accident37 (0.84)30 (0.74)7(1.96)0.015
Cardiovascular disease61 (1.38)58 (1.43)3 (0.84)0.360
Malignancy41 (0.93)36 (0.89)5 (1.4)0.334
Chronic lung disease22 (0.5)22 (0.54)0 (0.0)0.163
Chronic liver disease4 (0.09)4 (0.1)0 (0.0)0.553
Chronic renal disease2 (0.05)1 (0.2)1 (0.28)0.03
Allergy1 (0.02)1 (0.02)0 (0.0)0.767
Mental state, n (%)<0.001
Alert3736 (84.52)3597 (88.55)139 (38.83)
Verbal296 (6.7)224 (5.51)72 (20.11)
Pain290 (6.56)196 (4.83)94 (26.26)
Unconsciousness98 (2.22)45 (1.11)53 (14.8)
Vital signs
Systolic blood pressure, mean (SD), mmHg, n127.10 (19.45), 3750127.83 (18.94), 3434119.48 (23.00), 316<0.001
Diastolic blood pressure, mean (SD), mmHg, n81.20 (13.78), 367281.70 (13.84), 336175.70 (15.73), 311<0.001
Heart rate, mean (SD), min−1, n96.95 (19.69), 402696.45 (19.19), 3698102.60 (23.98), 328<0.001
Respiration rate, mean (SD), frequency/min, n18.47 (4.38), 395218.44 (4.34), 363118.7 (4.82), 3210.300
O2 saturation, mean (SD), (%), n96.53 (5.00), 405196.93 (4.50), 371991.99 (7.53), 332<0.001
Temperature, mean (SD), ∘C, n36.60 (0.50), 379936.61 (0.48), 349236.60 (0.71), 3070.775
Blood sugar, mean (SD), mg/dL, n144.45 (76.32), 296145.24 (79.26), 82141.19 (63.36), 580.680
NFA personnel for first aid
Manual airway maneuver, n (%)1498 (33.88)1321 (32.5)177 (49.44)<0.001
Oral airway, n (%)144 (3.26)84 (2.07)60 (16.76)<0.001
O< supply, n (%)2955 (66.82)2657 (65.38)298 (83.2)<0.001
Total (n = 4422)Ward + Discharge (n = 4064)Intensive care unit (n = 358)
O2 flow, mean (SD), L/min, n10.33 (5.07), 295510.04 (5.12), 265712.84 (3.77), 298<0.001
Nasal prongs, n (%)690 (15.6)670 (16.49)20 (5.59)<0.001
Facial mask, n (%)693 (15.67)637 (15.67)56 (15.64)0.987
Non-rebreathing mask, n (%)1721 (38.92)1485 (36.54)236 (65.92)<0.001
Bag valve mask, n (%)68 (1.54)34 (0.84)34 (9.5)<0.001
Supraglottic airway device, n (%)18 (0.41)8 (0.2)10 (2.79)<0.001
Tracheal intubation, n (%)6 (0.14)3 (0.07)3 (0.84)<0.001
ECG check (AED), n (%)1055 (23.86)857 (21.09)198 (55.31)<0.001
Fluid supply, n (%)82 (1.85)51 (1.25)31 (8.66)<0.001
Fluid dose, mean (SD), mL, n464.21 (398.68), 82513.53 (387.53), 51383.06 (409.77), 310.152
Drug supply, n (%)6 (0.14)5 (0.12)1 (0.28)0.441
AED monitoring, n (%)285 (6.45)204 (5.02)81 (22.63)<0.001
AED shock, n (%)5 (0.11)4 (0.1)1 (0.28)0.329
Medical direction579 (13.09)455 (11.2)124 (34.64)<0.001
CPR, n (%)26 (0.59)10 (0.25)16 (4.47)<0.001
Stop CPR, n (%)3 (0.07)1 (0.02)2 (0.56)<0.001
Distance and time related to smoke inhalation patient transfer
Distance from scene to ED, mean (SD), km9.56 (11.94)9.42 (11.91)11.24 (12.25)0.006
Time from call to scene arrival, mean (SD), min15.34 (18.48)15.77 (19.01)10.51 (9.32)<0.001
Time from patient contact to scene departure, mean (SD), min8.97 (10.82)9.06 (11.01)7.95 (8.23)0.017
Time from scene departure to ED arrival, mean (SD), min14. 17 (12.95)14.01 (12.44)15.96 (17.65)0.041
Total time from call to ED arrival, mean (SD), min45.84 (27.77)46.52 (28.13)38.12 (21.92)<0.001
Smoke inhalation patient hospital transfer result
First transfer HBO2 available hospital, (%)909 (20.56)805 (19.81)104 (29.05)<0.001
Retransfer, (%)47 (1.06)42 (1.03)5 (1.4)0.521
Retransfer HBO2 available hospital, (%)9 (19.15)8 (19.05)1 (20)0.959
Critical patients with smoke inhalation requiring multiplace hyper-baric chamber, (%)609 (13.77)449 (11.05)160 (44.69)<0.001
Transfer of critical patients with smoke inhalation for multiplace hyperbaric chamber, (%)158 (25.94)108 (24.05)50 (31.25)0.075
ICU, Intensive Care Unit; NFA, National Fire Agency; ECG, electrocardiogram; AED, automated external defibrillator; IV, Intravenous; CPR, Cardiopulmonary resuscitation; ED, Emergency department; HBO2, Hyperbaric oxygenation.

Regarding the NFA rescue missions, the transfer distance from the scene to the hospital was, overall, of 9.56 (±11.94) km, while it was longer for the ICU group (11.24 (±12.25) km). Overall, the total time from dispatch call to the arrival on the scene was 15.34 (±18.48) min, whereas the ICU group’s arrival time on the scene was shorter (10.51 (±9.32) min). The total time to the departure from the scene was, overall, of 8.97 (±10.82) min, while it was shorter, in the ICU group (7.95 (±8.23) min). The overall time from departure to the ED was 14.18 (±12.95) min, while it was longer in the ICU group (15.96 (±17.65) min). The total transfer time from reporting of the patient following smoke inhalation to the arrival at the ED was 45.84 (±27.77) min, while the ICU admitted group took a shorter transfer time of 38.12 (±21.92) min; 909 patients with smoke inhalation were immediately transferred to HBO2-therapy capable hospitals, corresponding to the 20.56% of the patients; this quote was slightly higher in the ICU group (29.05%). In addition, among the total patients with smoke inhalation transferred, 609 patients (13.77%) required multiplace hyperbaric chamber at the scene, while the proportion of patients requiring multiplace hyperbaric chamber in the ICU group was far higher (44.69%) (Table 1).

3.2 Hospital stage characteristics of patients with carbon monoxide poisoning due to smoke inhalation

At the ED visit, the mental status was alert in 84.57% of the patients, verbal in 8.85%, pain in 5.5%, and unconscious in 1.08%, and the proportion of patients showing mental change was lower than in the pre-hospital stage (Table 2). As for the vital signs assessed in the ED, both SBP and DBP, were found to be lower in the ICU group; the HR and respiration rate were higher, and SpO2 was lower in this group. The results of KTAS showed: first, second, third, fourth, and fifth-grade classification in 99 (2.38%), 709 (17.08%), 2795 (67.33%), 490 (11.8%), and 55 (1.32%) patients, respectively. HBO2 therapy was performed in 227 (5.13%) patients in the ED and in 210 (4.75%) after admission. Finally, the outcomes of discharge after hospitalization through the ED were: standard discharge in 720 (68.25%) cases, discharge against medical advice in 211 (20.0%), transfer in 98 (9.29%), death in 18 (1.71%), and “other” in 8 (0.76%) (Table 2).

Table 2.A comparative analysis of hospital factors between the combined general ward admission and discharged group, and ICU group in patients with carbon monoxide poisoning due to smoke inhalation.
Carbon monoxide poisoning due to smoke inhalationp-value
Total (n = 4422)Ward + Discharge (n = 4064)ICU (n = 358)
ED mental state, n (%)40903744346<0.001
Alert3459 (84.57)3344 (89.32)115 (33.24)
Verbal362 (8.85)281 (7.51)81 (23.41)
Pain225 (5.5)106 (2.83)119 (34.39)
Unconsciousness44 (1.08)13 (0.35)31 (8.96)
Vital signs
Systolic blood pressure, mean (SD), mmHg, n131.15 (22.23), 3887132.10 (21.77), 3559120.89 (24.44), 328<0.001
Diastolic blood pressure, mean (SD), mmHg, n80.43 (14.95), 388480.83 (14.58), 355976.05 (17.95), 325<0.001
Heart rate, mean (SD), (min−1), n94.56 (18.71), 402993.88 (18.24), 3697102.15 (21.97), 332<0.001
Respiration rate, mean (SD), frequency/min, n20.48 (4.18), 402020.37 (3.92), 370021.75 (6.31), 320<0.001
O2 saturation, mean (SD), %, n97.63 (3.01), 376993.08 (21.16), 361593.42 (15.30), 3370.706
Temperature, mean (SD), ∘C, n36.66 (0.51), 406536.67 (0.49), 372736.53 (0.73), 338<0.001
KTAS41513803348<0.001
1, n (%)99 (2.38)27 (0.71)72 (20.69)
2, n (%)709 (17.08)534 (14.04)175 (50.29)
3, n (%)2795 (67.33)2705 (71.13)90 (25.86)
4, n (%)490 (11.8)485 (12.75)5 (1.44)
5, n (%)55 (1.32)49 (1.29)6 (1.72)
8, n (%)3 (0.07)3 (0.08)0 (0.0)
HBO2 result, n (%)
HBO2 therapy in ED227 (5.13)182 (4.48)45 (12.57)<0.001
HBO2 therapy in hospital admission210 (4.75)153 (3.76)57 (15.92)<0.001
Result of hospital admission, n (%)1055702353<0.001
Normal discharge720 (68.25)526 (74.93)194 (54.96)
Discharge against medical advice211 (20.0)135 (19.23)76 (21.53)
Transfer98 (9.29)34 (4.84)64 (18.13)
Death18 (1.71)1 (0.14)17 (4.82)
Others8 (0.76)6 (0.85)2 (0.57)
ICU, Intensive Care Unit; ED, Emergency department; KTAS, Korea Triage Acuity Score; HBO2, Hyperbaric oxygenation.

3.3 Pre-hospital factors associated with intensive care unit admission of patients with carbon monoxide poisoning due to smoke inhalation transported by NFA

Table 3 shows the results of the multiple logistic regression analysis of the ICU admission. In patients who underwent smoke inhalation, and were transported by the 119 NFA ambulance, age (OR, 1.020; 95% CI, 1.010–1.029), verbal (OR, 3.564; 95% CI, 2.390–5.315), pain (OR, 4.011; 95% CI, 2.661–6.045), unconsciousness (OR, 5.728; 95% CI, 2.708–12.113), SBP (OR, 0.979; 95% CI, 0.969–0.989), HR (OR, 1.011; 95% CI, 1.004–1.018), SpO2 (OR, 0.965; 95% CI, 0.946–0.985), O2 supply (OR, 1.725; 95% CI, 1.143–2.603), use of nasal prongs (OR, 0.504; 95% CI, 0.281–0.905), and intentional inhalation (OR, 2.282; 95% CI, 1.659–3.139) were independently associated with ICU admission.

Table 3.Multiple logistic regression analyses demonstrating pre-hospital factors associated with intensive care unit (ICU) admission in patients with carbon monoxide poisoning due to smoke inhalation.
VariablesICU admission
Odds Ratio95% CIp-value
Epidemiology factor
Sex1.1130.824–1.5030.485
Age1.0201.011–1.029<0.001
Distance from scene to ED, km0.9850.971–1.0000.055
Time from call to scene arrival, min0.9970.986–1.0090.660
Time from patient contact to scene departure, min1.0030.986–1.0200.733
Time from scene departure to ED arrival, min1.0100.997–1.0230.124
Smoke inhalation region
HouseReference
Sea/river/mountain1.8270.731–4.5630.197
Medical facility1.8060.594–5.4870.297
Road1.2120.741–1.9840.444
Factory/Industry/Construction facility1.7270.743–4.0140.204
Commercial facility1.2010.727–1.9840.475
Dormitory/military/Prison facility1.0870.508–2.3230.830
Others1.0380.642–1.6780.879
Past history
Cerebrovascular disease1.4570.420–5.0570.554
Scene mental state
AlertReference
Verbal3.5642.390–5.315<0.001
Pain4.0112.662–6.046<0.001
Unconsciousness5.7282.708–12.114<0.001
Scene vital sign
Systolic blood pressure0.9790.969–0.989<0.001
Heart rate1.0111.004–1.0180.002
Oxygen saturation0.9650.946–0.9850.001
NFA personnel first aid
Manual airway maneuver0.9910.738–1.3300.951
Oral airway1.2010.712–2.02670.492
Tracheal intubation1.3760.092–20.6600.818
O2 supply1.7241.143–2.6030.009
Nasal prongs0.5040.281–0.9050.022
Non-rebreathing mask1.0380.716–1.5050.842
Bag valve mask1.3720.557–3.3820.492
ECG(AED) check1.1700.848–1.6150.339
AED monitoring1.4220.945–2.1380.091
Medical direction1.2690.897–1.7930.178
Reasons for smoke inhalation
UnintentionalReference
Intentional2.2821.659–3.139<0.001
The multiple logistic regression model included all variables with a p-value of <0.05 in the univariate analyses except variables presumed to have interactions with other variables. ICU, intensive care unit; CI, confidence interval; ECG, electrocardiogram; AED, automated external defibrillator; NFA, National fire agency; ED, Emergency department.

4. Discussion

Through this study, we were able to identify the pre-hospital epidemiology, vital signs, emergency treatment, hospital transfer status, and clinical outcomes of patients with CO poisoning after smoke inhalation. We further showed that the age, mental change, SBP, HR, SpO2, O2 supply, use of nasal prongs, and intentional inhalation were associated with ICU admission for patients with smoke inhalation. The relative affinities of CO and O2 for hemoglobin, myoglobin, and cytochrome p450 proteins are very different. The CO/O2 affinity ratio is 200–250 for hemoglobin, 20–25 for myoglobin, and <1 for the cytochrome system [28]. In case of CO poisoning due to smoke inhalation, CO gets strongly bound to hemoglobin, making pulse oximetry unreliable in estimating O2Hb saturation in CO-exposed patients. Hence, it is necessary to carefully observe the clinical symptoms of the patient. In addition, it is difficult to accurately measure the COHb level in the field, and the measured value does not reliably reflect the patient’s severity [29]. Thus, overall observations, including mental status, SpO2, and ECG of the patient, are required.

In this study, the first-aid results by NFA personnel at the pre-hospital stage revealed oxygen administration performed in 66.82% of all patients, and 83.2% in the ICU group (p < 0.001). In addition, the average oxygen dose was 10.33 (±5.07) L/min in all patients, and 12.84 (±3.77) L/min in the ICU group (p < 0.001). Previous studies on suspected CO poisoning recommended 100% oxygen supply by a tight-fitting face mask at normal atmospheric pressure at the scene or during transport [29, 30]. In addition, administration of 100% O2 as early as possible is recommended for all patients with a relevant suspected diagnosis (in alert patients, for example, using non-invasive continuous airway pressure, respiration using a mask with a demand valve, or administration of 15 L/min O2 through a reservoir mask) [31]. However, in our study, 33.18% of patients with smoke inhalation did not even receive oxygen supply, and 15.6% of them were treated with nasal prongs as the only oxygen supply method. These results suggest that medical recommendations were less strongly enforced in the pre-hospital setting. In addition, the ECG check rates were quite low (23.86% overall, and 55.31% in the ICU group). In case of suspected CO poisoning by smoke inhalation the heart may be affected, and if an ECG abnormality is observed, it may represent an indication for HBO2 therapy; therefore, it should be checked at the scene [32].

Patients suspected of CO poisoning should be transferred to a hospital capable of HBO2 therapy in case of a possible neurological change, which mandates for HBO2 therapy [33, 34]. HBO2 should be critically considered and initiated within 6 h in patients with neurologic deficits, cardiac ischemia, pregnancy, and/or a very high COHb concentration. If HBO2 therapy is administered, it should be initiated within 6 h, but under no circumstances, it should be initiated after more than 24 h [31]. Although it is not directly related to the outcomes of this study, some important considerations must be made for pregnant women suspected of CO poisoning, especially with respect to indications for HBO2 therapy. The elimination COHb half-life is 2 h for the mother and 7 h for the fetus. In the fetal system, saturation and elimination, possibly occurs at a slower rate than those in the maternal system. Therefore, some researchers regard pregnancy as a strict indication for HBO2, especially in the presence of neurological symptoms, signs of fetal distress, syncope, or high COHb levels [35, 36].

In this study, the evaluation of consciousness status was investigated through the alert, verbal, pain, unresponsive score alone, and not the Glasgow Coma Scale score. We considered consciousness below verbal as a relevant threshold for mental-status change, and logistic regression analysis showed a positive correlation between the need for intensive care and mental-status change, with the strength of this correlation being proportional to the severity of that change (unconsciousness > pain > verbal). Some studies have shown that consciousness changes are associated with aspiration pneumonia and poor prognosis in patients with CO poisoning. In addition, in patients with CO poisoning undergoing major changes in consciousness and aspiration pneumonia, the rate of ventilator use was found to be higher, and the duration of ventilator use and hospital stay longer [37]. If a patient with initial smoke inhalation shows an altered mental status in the field, it will be helpful for the patient’s prognosis to be transferred to a hospital where HBO2 therapy and intensive care are available.

In previous studies, patients with intentional smoke inhalation had higher mortality rates and higher ICU admission rates [38, 39, 40]. Similarly, in our study, intentional smoke inhalation was highly related to ICU admission. In the last years, the suicide rate in the Republic of Korea has decreased slightly until 2017, the highest level being 31.7 incidents per 100,000 population in 2011. However, in 2018, the value increased again to a level of 26.6 per 100,000 population. In addition, within the Organization for Economic Co-operation and Development member countries, referring to the most recent record of 2016, the Republic of Korea has the world’s second-highest suicide rate of 24.6 per 100,000 population (beaten only by Lithuania, with a 26.7 per 100,000 rate). As of 2018, the most common means of suicide in the Republic of Korea were: hanging (52.1%), fall (16.6%), and gas poisoning (15.7%). There is a consistent number of CO-poisoning cases in the Republic of Korea caused by intentional smoke inhalation [41]. According to our findings, patients with CO poisoning caused by intentional smoke inhalation are more likely to have a poor prognosis; hence, it appears to be crucial to assess intentionality at the scene.

Finally, HBO2 therapy to quickly remove CO from the body in patients with CO poisoning caused by smoke inhalation should be adequately provided as soon as possible. However, this study showed that the overall rate of direct transfer from the scene to a hospital capable of HBO2 was 20.56%, and as low as 29.05% in the ICU group. In addition, 13.77% of the patients required multiplace hyperbaric chamber, and 25.94% of them were transferred to a hospital equipped with a multiplace hyperbaric chamber. Previous studies have shown that monoplace hyperbaric chambers with certified medical equipment in high-pressure environments can provide adequate treatment even for critically ill CO poisoning patients [26, 42]. Therefore, it should be transferred to a hospital that can at least a monoplace hyperbaric chamber treatment, but the total transfer rate to a hospital with a hyperbaric oxygen chamber was only 20.56%. So, there was a problem in the selection of hospitals by paramedics. In addition, the transfer time of the total patients with smoke inhalation was 45.84 (±27.77) min, 33.16 min being the national average for other general acute conditions. The transfer time was longer than that for cardiovascular disease (25 min) or cerebrovascular disease (27 min) [43]. To date, there is no established cut-off value for the start time of HBO2 therapy in patients with CO poisoning, although HBO2 therapy shows great effectiveness in dissociating CO from hemoglobin in the early stages of CO poisoning. In addition, HBO2 therapy is known to help reduce delayed neurological sequelae and to remove cyanide, thermal burns, or upper airway inhalation burn injuries. Thus, this treatment is particularly important for patients with CO poisoning caused by smoke inhalation [8, 10, 44, 45, 46]. In the future, based on such considerations and on the findings of our study, we think that, in the development of future specialized guidelines, clear indications regarding HBO2 therapy must be deemed as a necessary and critical factor for optimizing the management of patients with smoke inhalation and for guiding their timely transfer to the most suitable hospital.

This study has some limitations. First, this study used retrospective data of patients transferred to the ED by the NFA ambulance at the pre-hospital stage. In the case of patients with clear evidence of death at the site, CPR was not performed and the cases were immediately handed over to the police. Hence, there is a possibility that the mortality rate or severity of the study may be lower than the actual result due to the occurrence of selection bias. In addition, patients who were transferred to non-emergency medical institutions that did not transmit data to NEDIS might have not been included in the study subject and therefore might have affected the study outcomes. Second, in the case of patients with smoke inhalation, there is the possibility of concurrent damage such as airway inhalation injury and superficial burn. However, due to the nature of the data available to us, it was not possible to identify the prognostic impact of these additional factors. Third, due to the nature of NEDIS data in this study, the data for evaluating patient prognosis included only the results of ED discharge, hospital discharge, and patient diagnosis. The patient’s COHb level, other blood tests, radiological tests, treatment prescriptions, and consciousness status at hospital discharge were not included; thus, there were limitations in the evaluation of the patient’s treatment process and prognosis. Fourth, in the analysis of the transferred patients based on the emergency department results, it was virtually impossible to match the prognosis after transfer to another hospital due to data limitations. Therefore, the group of transferred patients was primarily excluded from the analysis. Fifth, due to the nature of the data, it was not possible to analyze the presence and role of other possibly toxic gases. In emergency medicine, there is a general lack of awareness of the role of other toxic gases that are widely recognized in environmental medicine [7]. This lack of awareness greatly hinders the progress of understanding the mechanisms of toxicity of these poisonous substances, likely resulting in unsatisfactory treatment and enhanced mortality. Therefore, further studies on other toxic gaseous substances are needed. In the future, large-scale prospective and comprehensive studies are needed to better assess the prognostic significance of comorbid injuries, laboratory results, radiologic findings, specific HBO2-therapy parameters, state of consciousness at discharge, and possible complications during hospital-stay.

5. Conclusions

Our study demonstrated that age, mental change, SBP, HR, SpO2, O2 supply, use of nasal prongs, and intentional inhalation were associated with ICU admission in patients with CO poisoning by smoke inhalation. Considering these results, it is necessary to improve the pre-hospital EMS guidelines for CO poisoning patients by smoke inhalation.

Author contributions

Conceptualization—TH. Data curation—JHH. Formal analysis—JHH. Funding acquisition—SML, TH. Investigation—SML, DHL. Methodology—DHL, HYL. Software—DHL. Validation—TH. Writing - original draft—SML, HYL. Writing - review & editing—SML, DHL, JHH, TH.

Ethics approval and consent to participate

Approval for data abstraction was obtained from the Institutional Review Board of Chonnam National University Hospital (CNUH-EXP-2019-147). This study is a database study on anonymous patient visits; therefore, no informed consent from patients was required.

Acknowledgment

The authors would like to thank Editage (www.editage.co.kr) for providing English language editing assistance.

Funding

This study was supported by a grant (HCRI 19 001-1* HCRI21002) from the Chonnam National University Hwasun Hospital Research Institute of Clinical Medicine.

Conflict of interest

The authors declare no conflict of interest.

References

Huzar TF, George T, Cross JM. Carbon monoxide and cyanide toxicity: etiology, pathophysiology and treatment in inhalation injury. Expert Review of Respiratory Medicine. 2013; 7: 159–170.

[Google Scholar]

O’Brien DJ, Walsh DW, Terriff CM, Hall AH. Empiric Management of Cyanide Toxicity Associated with Smoke Inhalation. Prehospital and Disaster Medicine. 2011; 26: 374–382.

[Google Scholar]

Sircar K, Clower J, Shin MK, Bailey C, King M, Yip F. Carbon monoxide poisoning deaths in the United States, 1999 to 2012. The American Journal of Emergency Medicine. 2015; 33: 1140–1145.

[Google Scholar]

Ball LB, Macdonald SC, Mott JA, Etzel RA. Carbon monoxide-related injury estimation using ICD-coded data: methodologic implications for public health surveillance. Archives of Environmental and Occupational Health. 2005; 60: 119–127.

[Google Scholar]

Smith DL, Cairns BA, Ramadan F, Dalston JS, Fakhry SM, Rutledge R, et al. Effect of inhalation injury, burn size, and age on mortality: a study of 1447 consecutive burn patients. The Journal of Trauma. 1994; 37: 655–659.

[Google Scholar]

Kimmel EC, Still KR. Acute lung injury, acute respiratory distress syndrome and inhalation injury: an overview. Drug and Chemical Toxicology. 1999; 22: 91–128.

[Google Scholar]

Baud FJ. Acute poisoning with carbon monoxide (CO) and cyanide (CN). Therapeutische Umschau. Revue therapeutique. 2009; 66: 387–397.

[Google Scholar]

Weaver LK, Hopkins RO, Chan KJ, Churchill S, Elliott CG, Clemmer TP, et al. Hyperbaric oxygen for acute carbon monoxide poisoning. The New England Journal of Medicine. 2002; 347: 1057–1067.

[Google Scholar]

Jasper BW, Hopkins RO, Duker HV, Weaver LK. Affective Outcome Following Carbon Monoxide Poisoning: a prospective longitudinal study. Cognitive and Behavioral Neurology. 2005; 18: 127–134.

[Google Scholar]

Weaver LK. Hyperbaric oxygen therapy for carbon monoxide poisoning. Undersea and Hyperbaric Medicine. 2014; 41: 339–354.

[Google Scholar]

Gorman D, Drewry A, Huang YL, Sames C. The clinical toxicology of carbon monoxide. Toxicology. 2003; 187: 25–38.

[Google Scholar]

Hampson NB, Weaver LK. Carbon monoxide poisoning: a new incidence for an old disease. Undersea and Hyperbaric Medicine. 2007; 34: 163–168.

[Google Scholar]

Weaver LK, Deru K, Churchill S, Legler J, Snow G, Grey T. Carbon monoxide poisoning in Utah: 1996-2013. Undersea and Hyperbaric Medicine. 2016; 43: 747–758.

[Google Scholar]

Braubach M, Algoet A, Beaton M, Lauriou S, Héroux ME, Krzyzanowski M. Mortality associated with exposure to carbon monoxide in who European Member States. Indoor Air. 2013; 23: 115–125.

[Google Scholar]

Huang CC, Ho CH, Chen YC, Lin HJ, Hsu CC, Wang JJ, et al. Demographic and clinical characteristics of carbon monoxide poisoning: nationwide data between 1999 and 2012 in Taiwan. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine. 2017; 25: 70.

[Google Scholar]

Ministry of Health and Welfare. Study of the optimal allocation of hyperbaric oxygen therapy center in South Korea 2019. 2020 Available at: http://www.prism.go.kr/homepage/entire/retrieveEntireDetail.do;jsessionid=0C66CCBBFFF9A65AEAF8B824A75338C3.node02?cond_research_name=&cond_research_start_date=&cond_research_end_date=&research_id=1351000-201900321&pageIndex=24&leftMenuLevel=160 (Accessed: 20 June 2020).

[Google Scholar]

Peterson JE, Stewart RD. Absorption and Elimination of Carbon Monoxide by Inactive Young Men. Archives of Environmental Health. 1970; 21: 165–171.

[Google Scholar]

Weaver LK, Howe S, Hopkins R, Chan KJ. Carboxyhemoglobin half-life in carbon monoxide-poisoned patients treated with 100% oxygen at atmospheric pressure. Chest. 2000; 117: 801–808.

[Google Scholar]

Pace N, Strajman E, Walker EL. Acceleration of carbon monoxide elimination in man by high pressure oxygen. Science. 1950; 111: 652–654.

[Google Scholar]

Centers for Disease Control and Prevention (CDC). Carbon monoxide exposures – United States, 2000–2009. Morbidity and Mortality Weekly Report. 2011; 60: 1014–1017.

[Google Scholar]

National fire agency 119. The standard protocols for 119 emergency medical services providers 2019. 2019. Available at: http://www.nfa.go.kr/nfa/publicrelations/legalinformation/0017/0003/?boardId=bbs_0000000000001097&mode=view&cntId=4&category=&pageIdx=&searchCondition=&searchKeyword= (Accessed: 25 November 2019).

[Google Scholar]

Hwang JS, Choi WJ, Kim SH, Choi BH, Lee HJ, Ahn R, et al. Early Predictors of Critical Cases for the Patients Who Visited Emergency Department due to Gas Inhalation: Early Predictors of Severity in Gas Inhalation. Journal of the Korean Society of Integrative Medicine. 2017; 28: 475–483.

[Google Scholar]

Choi BH, Jeon J, Ryoo SM, Seo DW, Kim WY, Oh BJ, et al. Recent Epidemiologic Features of Carbon Monoxide Poisoning in Korea: A Single Center Retrospective Cohort Study. Journal of the Korean Society of Clinical Toxicology. 2012; 10: 80–85.

[Google Scholar]

Bae SH, Lee JS, Kim KH, Park JS, Shin DW, Kim HJ, et al. Epidemiologic Characteristics of Carbon Monoxide Poisoning: Emergency Department Based Injury In-depth Surveillance of Twenty Hospital. Journal of the Korean Society of Clinical Toxicology. 2016; 14: 122–128.

[Google Scholar]

Kim YJ, Sohn CH, Seo DW, Oh BJ, Lim KS, Kim WY. Clinical Predictors of Acute Brain Injury in Carbon Monoxide Poisoning Patients with Altered Mental Status at Admission to Emergency Department. Academic Emergency Medicine. 2019; 26: 60–67.

[Google Scholar]

Hampson NB, Dunford RG, Kramer CC, Norkool DM. Selection criteria utilized for hyperbaric oxygen treatment of carbon monoxide poisoning. The Journal of Emergency Medicine. 1995; 13: 227–231.

[Google Scholar]

Weaver LK. Hyperbaric oxygen in the critically ill. Critical Care Medicine. 2011; 39: 1784–1791.

[Google Scholar]

Rose JJ, Wang L, Xu Q, McTiernan CF, Shiva S, Tejero J, et al. Carbon Monoxide Poisoning: Pathogenesis, Management, and Future Directions of Therapy. American Journal of Respiratory and Critical Care Medicine. 2017; 195: 596–606.

[Google Scholar]

Wolf SJ, Lavonas EJ, Sloan EP, Jagoda AS. Clinical policy: Critical issues in the management of adult patients presenting to the emergency department with acute carbon monoxide poisoning. Annals of Emergency Medicine. 2008; 51: 138–152.

[Google Scholar]

Myers RA, Linberg SE, Cowley RA. Carbon monoxide poisoning: the injury and its treatment. Journal of the American College of Emergency Physicians. 1979; 8: 479–484.

[Google Scholar]

Eichhorn L, Thudium M, Jüttner B. The Diagnosis and Treatment of Carbon Monoxide Poisoning. Deutsches Arzteblatt International. 2018; 115: 863–870.

[Google Scholar]

Rastelli G, Callegari S, Locatelli C, Vezzani G. Myocardial injury in carbon monoxide poisoning. Giornale Italiano di Cardiologia. 2009; 10: 227–233. (In Italian)

[Google Scholar]

Choi IS. Delayed Neurologic Sequelae in Carbon Monoxide Intoxication. Archives of Neurology. 1983; 40: 433–435.

[Google Scholar]

Min SK. A brain syndrome associated with delayed neuropsychiatric sequelae following acute carbon monoxide intoxication. Acta Psychiatrica Scandinavica. 1986; 73: 80–86.

[Google Scholar]

Aubard Y, Magne I. Carbon monoxide poisoning in pregnancy. BJOG: An International Journal of Obstetrics and Gynaecology. 2000; 107: 833–838.

[Google Scholar]

Roderique EJD, Gebre-Giorgis AA, Stewart DH, Feldman MJ, Pozez AL. Smoke inhalation injury in a pregnant patient: a literature review of the evidence and current best practices in the setting of a classic case. Journal of Burn Care and Research. 2012; 33: 624–633.

[Google Scholar]

Sohn CH, Huh JW, Seo DW, Oh BJ, Lim KS, Kim WY. Aspiration Pneumonia in Carbon Monoxide Poisoning Patients with Loss of Consciousness: Prevalence, Outcomes, and Risk Factors. The American Journal of Medicine. 2017; 130: 1465.e21–1465.e26.

[Google Scholar]

Miller M, Azrael D, Hemenway D. The epidemiology of case fatality rates for suicide in the northeast. Annals of Emergency Medicine. 2004; 43: 723–730.

[Google Scholar]

Heo IY, Choi SC, Lee CA, Ahn JH, Min YG, Jung YS, et al. Influence of the Werther effect: an increase of intentional carbon monoxide poisoning. Journal of the Korean Society of Clinical Toxicology. 2009; 7: 143–149.

[Google Scholar]

Kim WK, Kim KH, Shin DW, Park JS, Kim H, Jeon WC, et al. Characteristics of Korean poisoning patients: retrospective analysis by National Emergency Department Information System. Journal of the Korean Society of Clinical Toxicology. 2019; 17: 108–117.

[Google Scholar]

Ministry of health & welfare Korea suicide prevention center. 2020 Suicide prevention white paper. 2020. Available at: https://spckorea-stat.or.kr/boadpublishview.do (Accessed: 29 May 2020).

[Google Scholar]

Weaver LK, Churchill S, Deru K. Critical care of patients treated in monoplace hyper baric chambers, past 20 years. Undersea and Hyperbaric Medicine. 2006; 33: 350–351.

[Google Scholar]

National fire agency 119. 119 emergency medical services annual report in 2018. 2018. Available at: https://opengov.seoul.go.kr/sanction/15777062 (Accessed: 30 July 2018).

[Google Scholar]

Shannon MW, Borron SW, Burns MJ. Haddad and Winchester’s Clinical Management of Poisoning and Drug Overdose (pp. 1309–1316). 4th edn. Elsevier: Philadelphia. 2007.

[Google Scholar]

Cianci P, Slade JB Jr, Sato RM, Faulkner J. Adjunctive hyperbaric oxygen therapy in the treatment of thermal burns. Undersea and Hyperbaric Medicine. 2013; 40: 89–108.

[Google Scholar]

Mathieu D, Marroni A, Kot J. Tenth European Consensus Conference on Hyperbaric Medicine: recommendations for accepted and non-accepted clinical indications and practice of hyperbaric oxygen treatment. Diving and Hyperbaric Medicine. 2017; 47: 24–32.

[Google Scholar]