a b s t r a c t

Backgrounds: Delayed neurological sequelae (DNS) are a severe complication of carbon monoxide poisoning (COP) and high predisposing rates of disability and mortality, yet the relationship between exposure factors and DNS remains unknown. The aim was to investigate the association between domestic sources of COP and DNS.

Methods: Patients diagnosed with COP between December 2016 and November 2021 were included and divided into two groups according to their sources of poisoning and the endpoint outcome was analyzed by logistic regression before and after propensity score matching (PSM).

Results: Overall, medical data from 314 patients were analyzed. In multivariate logistic regression, advanced age (adjusted odds ratio (AOR): 1.028, 95% CI: 1.008-1.049, P = 0.007), longer duration of exposure to the first treat- ment of Hyperbaric oxygen (HBO) (AOR: 1.081, 95% CI: 1.036-1.127, P = 0.001), and intoxication by charcoal burning (AOR: 3.24, 95% CI: 1.208-8.69, P = 0.019) were associated with a higher risk of developing DNS. After 1:1 PSM, the outcomes also revealed that charcoal burning intoxication (odds ratio (OR): 8.396, 95% CI: 3.342-21.095, P<0.001) was associated with greater odds of DNS. Conclusions: This study indicates that domestic COP caused by charcoal burning is more likely to trigger DNS than gas-emitting heaters.

(C) 2022 Published by Elsevier Inc.

1. Introduction

Carbon monoxide poisoning (COP), a global Public health problem, is a major cause of death and disability due to improper management [1-3]. Hemoglobin (Hb) has a 200-250 fold greater affinity for carbon monox- ide (CO) CO than for oxygen, developing carboxyhemoglobin (COHb). The pleiotropic effects of CO on mitochondrial respiration, cellular energy utilization, free radical generation, and inflammation, particularly in the brain and heart, is the fundamental to the pathophysiology of COP [2-4]. Delayed neurological sequelae (DNS) are a leading disabling complication of COP, with associated symptoms often manifesting within 2 days to 6 weeks [3,5]. The pathophysiology of DNS remains elusive with some evi- dence suggesting that it may be caused by hypoxia of brain tissue, in- creased oxidative stress, and lipid peroxidation [3,6]. However, there are only scant data on the liability of this complication.

* Corresponding author at: The Second Affiliated Hospital, Department of Emergency, Hengyang Medical School, University of South China, Jiefang Avenue, Zhengxiang District, 421001 Hengyang, Hunan, China.

E-mail address: [email protected] (X. Zhang).

Historically, research investigating the high-risk of exposure factors associated with DNS has focused on hyperbaric oxygen (HBO) therapy with few other factors addressed [7,8]. In previous studies on DNS, advanced age, longer duration of CO exposure, and duration of CO expo- sure to HBO therapy were relevant to a higher possibility of developing DNS [6-10]. To date, few studies have investigated the association be- tween different sources of COP and DNS. Whereas domestic sources of CO are the leading cause of COP in large parts of the world [1,11-13]. In south China, charcoal burning and gas-emitting heaters are the major sources of CO. Recent evidence suggests that charcoal burning suicide behavior and DNS are closely related [14,15]. Therefore, this study aims to explore the relationship between COP and DNS from domestic sources.

1. Methods
   1. Study design

Prior to commencing the study, ethical permission was granted from the Medical Ethics Board of our hospital, informed consent was waived

https://doi.org/10.1016/j.ajem.2022.06.003 0735-6757/(C) 2022 Published by Elsevier Inc.

Table 1

Symptoms related to delay neurological sequelae.

Neurological sequelae Cognitive and psychological sequelae

Motor and gait deficits Memory loss

Dysphagia Cognitive damage

Cortical blindness Mood disorder

Urinary or fecal incontinence Psychosis

Bradykinesia Personality changes

Peripheral neuropathies Dementia

Dysarthria Concentration decline

Parkinson-like Syndromes Dyssynergia

Dizzy Headache

Hearing Impairment and tinnitus

given that it was a retrospective study design. The medical data in this study were obtained from patients with COP diagnosed in our hospital within five years (December 2016 to November 2021). Primary inclu- sion criteria for the COP were: 1. History of CO exposure; 2. clinical man- ifestations related to COP; 3. COHb levels were >=3% and >=10% in nonsmokers and smokers, respectively [2,3,12]. Exclusion criteria were: 1. Age below 14 years; 2. Death on arrival at the Emergency De- partment (ED); 3. Not treated with HBO; 4. Sources of intoxication not related to domestic; 5. Persistent neurological symptoms.

• 1. Data collection and medical management

After COP patients arrived at the ED, we collected the following data based on their first medical records: gender, age, medical comorbidities, smoking history, sources of intoxication, clinical symptoms, initial Glasgow Coma Scale (GCS), duration of CO exposure to ED, duration of exposure to the first treatment of HBO, laboratory and imaging data, and disease com- plications. New-onset symptoms (Table 1) within 6 months after COP in- dicated DNS. At our hospital, all patients considered to have COP received immediate high-flow (10 L/min) oxygen by face mask, and the fraction of inspired oxygen (FiO2) was adjusted to 100% for patients who required intubation for mechanical ventilation. HBO therapy would be pro- vided within 6 h of the patient’s arrival in the ED at a target pressure of 2.2 standard atmospheres for 90 min/session, with a minimum of three ses- sions over a 48-h period. Blood tests would be taken admittedly and MRI would be performed within 3 days of intoxication if conditions permit.

• 1. Statistical analysis

Continuous and categories variables were summarized descriptively as median (interquartile range) and percentages, respectively. A total of 314 patients were divided into two groups (charcoal burning and gas emitting heaters) and the Mann-Whitney non-parametric test and chi-square test were used for comparison between groups, respectively, while propensity score matching (PSM) was applied to reduce con- founding factors and multivariable logistical regression analysis was conducted before and after PSM. Covariates with p<0.1 in univariate analysis were contained as candidate variables for multivariable analy- sis. Covariates with matched baseline characteristics were generated by PSM (1:1) with matching tolerance of 0.05; matched covariates contained gender, age, history of coma, smoking history, medical co- morbidities, duration of CO exposure to ED, and time from exposure to the first treatment of HBO. Matched covariates remained unbalanced and multivariable analysis was performed. All statistical tests were two-sided, P<0.05 was defined as statistically significant, and data were analyzed by SPSS (version 26.0, IBM).

1. Results

A total of 417 COP inpatients were screened and 103 were excluded for the following reasons:43 were not treated with HBO, 31 were <14

years old, 19 had a non-domestic related source of intoxication, 4 died on arrival in the ED, and 6 had persistent neurological sequelae, thus leaving 314 patients. (Fig. 1).

The baseline characteristics of the two groups (210 patients in char- coal burning and 104 patients in gas-emitting heaters, respectively) were summarized in Table 1. The median age (years) was greater in the charcoal-burning group than in the gas-emitting heaters group (5.4 years vs. 47 years, P = 0.002). The median time from exposure to ED and exposure to the first treatment of HBO was longer in the char- coal burning group than in the gas-emitting heaters (6 h vs. 4 h, P = 0.042; 9 h vs 6 h, P = 0.008). The charcoal-burning group also had higher levels of WBC (*10^9/L), Mb ((ug/L)) and Bun (mmol/L) at

10.24 *10^9/L [IQR,7.98-13.76*10^9/L], 104.7 ug/L [IQR,37.55-369.4

ug/L], and 5.94 mmol/L [IQR,4.81-8.18 mmol/L], while the correspond- ing data for the gas-emitting heater group were 9.75*10^9/L [IQR,7.51-11.77*10^9/L] (P = 0.032), 53.75 ug/L [IQR,22.92,-165.75 ug/L] (P = 0.001), and 5.21 [IQR,4.26-6.38 mmol/L] (P = 0.001). The

charcoal burning group had a higher proportion of the coma history than the other groups (81.9% vs 65.4%, P = 0.002), while the charcoal burning group also had a higher proportion of developed DNS than the other groups (19% vs 6.7%, P = 0.007). Interestingly, the concentra- tion of COHb was higher in the gas-emitting heaters group (24.2% [IQR,10.8-32.3%] vs 18.5% [IQR,9.9-28.1%], P = 0.026) than in the

other group (Table 1).

Exposure factors such as gender, age, history of smoking, history of coma, medical comorbidities, duration of CO exposure to ED, and expo- sure to the first treatment of HBO were included. Univariate logistic re- gression analysis indicated that age, current smoking, history of stroke, duration of CO exposure, and exposure to the first treatment of HBO were associated with DNS. Multivariable logistic regression models were then performed, adjusting for potential covariates, with only ad- vanced age (adjusted odds ratio (AOR)): 1.028, 95% CI: 1.008-1.049, P = 0.07), longer time from exposure to the first treatment of hyper- baric oxygen (HBO) (AOR: 1.081, 95% CI: 1.036-1.127, P = 0.001), and

charcoal burning intoxication (AOR: 3.24, 95% CI: 1.208-8.69, P = 0.019) remained (Table 3).

After PSM,101 pairs of patients were included with other exposure factors well balanced between the two groups (Table 2). In logistic regres- sion, charcoal burning intoxication (odds ratio (OR): 8.396, 95% CI: 3.342-21.095, P<0.001), remained statistically significant (Table 4). In summary, the above results suggest a higher risk of DNS from charcoal burning intoxication.

1. Discussion

Of the 314 cases in this study, a total of 47 (14.9%) had been devel- oped DNS, 40 (19%) in the charcoal burning group, and 7 (6.7%) in the gas emitting group. In other studies, the incidence of DNS was similar with COP patients who underwent HBO therapy [7,10]. New-onset symptoms within 6 months after COP may indicate DNS but lacks spec- ificity. Given that a history of COP is not readily available in some cases, there were difficulties in diagnosing DNS [16]. MRI is critical in assessing DNS, as lesions of globus pallidus, cerebellum, cerebral cortex, hippo- campus, and basal ganglia are typically seen in DNS patients [5]. There- fore, the combination of both clinical findings and MRI may improve the diagnosis of DNS.

In this study, charcoal burning was more susceptible to developing into DNS (AOR: 3.24, 95% CI: 1.208-8.69, P = 0.019). To reduce the in-

fluence of confounding factors, PSM was performed. Nevertheless, the matched result was identical, indicating a high risk in the charcoal burn- ing group (OR: 8.396, 95% CI: 3.342-21.095, P<0.001). Surprisingly, the concentrations of COHb were higher in the gas-emitting group. The higher COHb level was not associated with adverse outcomes, which was supported by the previous observations [7,10,17], and there are several possible explanations based on our findings.

	Propensity Score Matching

Fig. 1. Flowchart of patient selection.

Gas-emitting heaters (n=101)

Charcoal burning

(n=101)

First, the previous studies have identified the different characteris- tics of CO emissions between charcoal burning and gas-emitting heaters [1,18,19]. At the beginning of charcoal burning, the concentration of CO is low, after which there is a slow increase that remains high for some time and eventually remains at a low concentration level. In contrast to the former, when using gas-emitting heaters, a short-term peak in CO emissions is common when the environment is not ventilated, a phenomenon that will lead to severe symptoms [20], facilitating rapid detection of COP and prompt treatment. Several Animal experiments showed the exposure of rats to relatively low levels of CO (1000 ppm) for 40 min followed by high levels of CO (3000 ppm) for 20 min caused delayed cognitive impairment in rats [21,22]. However, the CO levels were higher in rats than in humans as rats’ hemoglobin binding affinity was lower [23]. And these rat models of DNS showed CO exposure trig- gers a cascade of biochemical reactions that generate delayed hippo- campal cell death in the dentate gyrus and induces Cognitive function

Excluded(n=103)

• 43 for whom not received HBO

therapy

• 31 were under 14-year-old
• 19 were poisoned not related to domestic
• 4 died when arrived ED
• 6 had persistent neurological sequelae

314 patients included

Gas-emitting heaters

(n=104)

Charcoal burning (n=210)

417 patients screened

disorder, which may be the pathophysiologic manifestations of DNS [21-24]. These studies may interpret why the COP outcomes were in- consistent with CO levels and also implied that the identification and prevention of associated biochemical reactions, rather than merely the focus on reducing COHb levels, is crucial in treating DNS.

Second, charcoal burning is the primary form of heating in South China, especially in rural areas. People are more likely to use charcoal for extended periods in cold weather in their living area, especially when sleeping. Repeated and prolonged exposure to CO can cause chronic hypoxia, which is a cause of neurological complications [25,26]. Previous studies have found that brain lesions on MRI are asso- ciated with DNS, and such lesions have been observed in patients with chronic COP [17,27], which suggests a weak link between chronic COP and DNS.

Third, charcoal burning releases many gases, such as carbon dioxide, CO, sulfur dioxide (SO2), and other harmless gases. However,

Table 2

Baseline characteristics before and after propensity score matching.

Characteristics	Before matching (n =	314)			After matching (n = 200)
	Charcoal burning	Gas-emitting heaters	p		Charcoal burning	Gas-emitting heaters	p
N	210	104			101	101
Sex Male (%)	90 (42.9)	39 (37.5)	0.842		39 (38.6)	40 (39.6)	1
Age (years) (median [IQR])	54 [38, 68.75]	47 [29, 61]	0.002		52.00 [30.00, 65.00]	50.00 [29.00, 62.00]	0.552
Current smoker (%)	43 (20.5)	12 (11.5)	0.151		7 (6.9)	13 (12.9)	0.239
History of coma (%)	172 (81.9)	68 (65.4)	0.002		76 (75.2)	67 (66.3)	0.216
Comorbidities (%) Stroke	9 (4.3)	2 (1.9)	0.456		2 (2.0)	1 (1.0)	1
Hypertension	15 (7.1)	8 (7.7)	1		8 (7.9)	7 (6.9)	1
COPD	5 (2.4)	1 (1.0)	0.67		2 (2.0)	2 (2.0)	1
Diabetes	13 (6.2)	8 (7.7)	0.794		8 (7.9)	8 (7.9)	1
CAD	5 (2.4)	3 (2.9)	1		5 (5)	3 (3)	0.718
Initial Glasgow Coma Scale<9 (%)	59 (28.1)	21 (20.2)	0.169		33 (32.7)	20 (19.8)	0.112
Time from CO exposure to ED (hours) (median [IQR])	6 [4, 10]	4 [3, 8]	0.042		5.00 [4.00, 8.00]	4.00 [3.00, 8.00]	0.359

Time from exposure to the first treatment of HBO (hours) (median [IQR])

9 [6, 14] 6 [4.75, 13] 0.008 7.00 [6.00, 14.00] 6.00 [5.00, 12.00] 0.083

COHb (%) (median [IQR])	18.5 [9.9, 28.1]	24.2 [10.8, 32.3]	0.026	17.70 [10.30, 23.90]	24.20 [10.80, 31.90]	0.019
Troponin I positive (%)	101 (51.8)	43 (45.7)	0.402	45 (46.9)	41 (45.6)	0.973
WBC (*10^9/L) (median [IQR])	10.24 [7.98, 13.76]	9.75 [7.51, 11.77]	0.032	10.41 [8.49, 14.00]	9.76 [7.55, 11.75]	0.035
Hb (g/L) (median [IQR])	132 [118.75, 149]	132.00 [114.5, 142]	0.275	130.00 [118.00, 150.00]	133.00 [117.00, 143.00]	0.655
CRP (mg/L) (median [IQR])	3.23 [2, 14.3]	3.13 [1.48, 11.97]	0.344	3.40 [3.05, 13.40]	3.13 [0.35, 9.70]	0.034
Mb (ug/L) (median [IQR])	104.7 [37.55, 369.4]	53.75 [22.92, 165.75]	0.001	87.00 [29.58, 305.80]	54.00 [23.00, 156.10]	0.036
AST (U/L) (median [IQR])	27.1 [20.5, 51.4]	25.2 [19.15, 36.3]	0.151	27.10 [20.80, 50.70]	25.30 [19.82, 36.25]	0.283
Bun (mmol/L) (median [IQR])	5.94 [4.81, 8.18]	5.21 [4.26, 6.38]	0.001	5.60 [4.44, 8.19]	5.22 [4.41, 6.40]	0.04
CK (U/L) (median [IQR])	162 [88.93, 714.5]	124.5 [83.75, 297]	0.065	169.5 [80.5, 623.85]	120.6 [83.75, 225.1]	0.086
Acute brain lesions on MRI (%)	55 (36.4)	25 (33.8)	0.81	28 (40.6)	23 (31.1)	0.312
DNS (%)	40 (19)	7 (6.7)	0.007	35 (34.7)	6 (5.9)	<0.001

Abbreviations: COPD, chronic obstructive pulmonary disease; CAD, coronary artery disease; HBO, hyperbaric oxygen; CO, carbon monoxide; ED, emergency department; COHb, carboxyl hemoglobin; WBC, white blood cell; Hb, hemoglobin; CRP, C-creative protein; Mb, myoglobin; AST, aspartate aminotransferase; Bun: Blood urea nitrogen; CK, c-creatine kinase; MRI, mag-

netic resonance imaging.

gas-emitting heaters mainly emit carbon dioxide and CO, while other harmful gases are emitted at much lower levels than charcoal burning [19,28]. Some studies have found that SO2 exposure could accelerate hypoxia and induce brain impairment [14,29]. Therefore, a combination of CO and SO2 exposure may be a possible factor in increasing the risk of DNS.

A vital strength of the present study is that it is the first to investigate the association between domestic sources of COP and DNS, and our find- ings may provide new ideas for clinical guidance. COP intoxication by charcoal burning should be treated more aggressively and earlier, such as with HBO therapy to prevent DNS. We found a subtle connection be- tween DNS and the intoxication of mixed harmless gases, and further research could be conducted to determine the feasibility of such a con- nection. In addition, our study attempts to provide ideas for preventing DNS at the source. Given that over 2.7 billion population worldwide were using solid fuels for cooking and heating per year, mostly in devel- oping countries [30], predictably, data from a survey in China has shown that the prevalence of solid fuel (including charcoal) for domestic

purposes varied from 36% to 95% in some rural areas [30]. In rural China, indoor combustion of solid fuels in outdoor boilers is common, generating more polluted air than ventilated cookstoves [18,30], and predisposes individuals to COP. Additionally, those patients do not seek immediate medical attention due to poverty and ignorance, thus increasing the odds of DNS. Therefore, relevant interventions such as health education and home ventilation may be the most effective way to stem DNS in this population.

Some limitations of this study should be mentioned as follows. First, given that this was retrospective research, several essential pieces of in- formation were incomplete. Duration of CO exposure is an acknowl- edged risk factor for DNS, the vital information we were unable to gather the available data. And thus, we used the duration of CO expo- sure to ED instead, which may impinge on results. Second, given that this was a single-center study with regional discrepancies, selection bias was not easily eliminated. Futher studies with a larger sample size and multicentric samples are needed to provide a higher degree of evidential feasibility for this finding.

Table 3

Exposure factors of delayed neurological sequelae by univariate and multivariate logistic regression analysis

Variables	Primitive cases (n = 314)
	Univariate analysis	Multivariate analysis
	P	AOR (95%CI)	P
Age	0.001	1.028 (1.008-1.049)	0.007
Current smoker	0.088	0.224 (0.068-0.740)	0.014
Stroke	0.001	3.126 (0.730-13.377)	0.124
Duration of CO exposure to ED	0.001	1.044 (0.953-1.145)	0.348
Duration of exposure to the first treatment of HBO	0.001	1.081 (1.036-1.127)	0.001
Intoxication by charcoal burning	0.003	3.24 (1.208-8.690)	0.019

Abbreviations: CO, carbon monoxide; ED, Emergency Department. AOR, adjusted odds ratio.

Table 4 Exposure factors of delayed neurological sequelae by multivariate logistic regression anal- ysis after propensity score matching

Variables Univariate analysis (n = 202) OR (95% CI) P

Intoxication by charcoal burning 8.396 (3.342-21.095) <0.001 Abbreviations: OR, odds ratio.

1. Conclusion

This study shows that we need to focus on charcoal burning as a source of intoxication in domestic COPs, and given the higher risk of de- veloping DNS from charcoal burning compared to gas-emitting heaters, it may hint at a distinction for patients with an increased risk of DNS. However, additional studies are essential to confirm our findings.

Credit authorship contribution statement

Shan Liu: Writing - review & editing, Writing - original draft, Soft- ware, Data curation, Conceptualization. Yan Liu: Methodology, Investi- gation, Formal analysis, Data curation. Dedong Xie: Validation,

Supervision. Chanjuan Yang: Data collection. Xia Zhang: Funding

acquisition, Conceptualization.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influ- ence the work reported in this paper.

References

1. Casey JG, Ortega J, Coffey E, Hannigan M. Low-cost measurement techniques to char- acterize the influence of home heating fuel on carbon monoxide in Navajo homes. Sci Total Environ. 2018;625:608-18. https://doi.org/10.1016/j.scitotenv.2017.12. 312.
2. Rose JJ, et al. Carbon monoxide poisoning: pathogenesis, management, and future directions of therapy. Am J Respir Crit Care Med. 2017;195:596-606. https://doi. org/10.1164/rccm.201606-1275CI.
3. Weaver LK. Clinical practice. Carbon monoxide poisoning. N Engl J Med. 2009;360: 1217-25. https://doi.org/10.1056/NEJMcp0808891.
4. Piantadosi CA. Carbon monoxide poisoning. N Engl J Med. 2002;347:1054-5. https:// doi.org/10.1056/NEJMp020104.
5. Jeon SB, et al. Acute brain lesions on magnetic resonance imaging and delayed neu- rological sequelae in carbon monoxide poisoning. JAMA Neurol. 2018;75:436-43. https://doi.org/10.1001/jamaneurol.2017.4618.
6. Betterman K, Patel S. neurologic complications of carbon monoxide intoxication. Handb Clin Neurol. 2014;120:971-9. https://doi.org/10.1016/B978-0-7020-4087-0. 00064-4.
7. Thom SR, et al. Delayed neuropsychologic sequelae after carbon monoxide poison- ing: prevention by treatment with hyperbaric oxygen. Ann Emerg Med. 1995;25: 474-80. https://doi.org/10.1016/s0196-0644(95)70261-x.
8. Huang CC, et al. Impact of hyperbaric oxygen therapy on subsequent neurological sequelae following carbon monoxide poisoning. J Clin Med. 2018;7. https://doi. org/10.3390/jcm7100349.
9. Liao SC, et al. Targeting optimal time for hyperbaric oxygen therapy following car- bon monoxide poisoning for prevention of Delayed neuropsychiatric sequelae: a

retrospective study. J Neurol Sci. 2019;396:187-92. https://doi.org/10.1016/j.jns. 2018.11.025.

1. Pepe G, et al. Delayed neuropsychological sequelae after carbon monoxide poison- ing: predictive risk factors in the Emergency Department. A retrospective study. Scand J Trauma Resusc Emerg Med. 2011;19:16. https://doi.org/10.1186/1757- 7241-19-16.
2. Annest JHT, Clower J, Yip F, Stock A, Lucas M, Iqbal S Nonfatal. Unintentional, non- fire-related carbon monoxide exposures-United States, 2004-2006 (Reprinted from MMWR, vol 57, pg 896-899, 2008). JAMA J Am Med Assoc. 2008;300:

2362-3. url. https://pubmed.ncbi.nlm.nih.gov/18716581.

1. Chiew AL, Buckley NA. Carbon monoxide poisoning in the 21st century. Crit Care. 2014;18. https://doi.org/10.1186/cc13846.
2. Centers for Disease, C. & Prevention. Carbon monoxide exposures-United States, 2000-2009. MMWR Morb Mortal Wkly Rep. 2011;60:1014-7. url. https://www. ncbi.nlm.nih.gov/pubmed/21814164.
3. Chang CC, et al. Clinical significance of the pallidoreticular pathway in patients with carbon monoxide intoxication. Brain. 2011;134:3632-46. https://doi.org/10.1093/ brain/awr287.
4. Adam J, Baulac M, Hauw JJ, Laplane D, Duyckaerts C. Behavioral symptoms after pallido-nigral lesions: a clinico-pathological case. Neurocase. 2008;14:125-30. https://doi.org/10.1080/13554790802032200.
5. Tabuas-Pereira M, Santana I, Santo GC. Clinical reasoning: a 55-year-old man with rapidly progressive dementia and parkinsonism. Neurology. 2017;89:e182-7. https://doi.org/10.1212/WNL.0000000000004495.
6. Liu Z, et al. acute carbon monoxide poisoning with low saturation of carboxyhaemoglobin: a forensic retrospective study in Shanghai. China Sci Rep. 2021;11:18554. https://doi.org/10.1038/s41598-021-97436-8.
7. Deng M, et al. Real-time combustion rate of wood charcoal in the heating fire basin: direct measurement and its correlation to CO emissions. Environ Pollut. 2019;245: 38-45. https://doi.org/10.1016/j.envpol.2018.10.099.
8. Belanger K, Triche EW. Indoor combustion and asthma. Immunol Allergy Clin North Am. 2008;28:507-19. vii. https://doi.org/10.1016/j.iac.2008.03.011.
9. Raub JA, Mathieu-Nolf M, Hampson NB, Thom SR. Carbon monoxide poisoning-a public health perspective. Toxicology. 2000;145:1-14. https://doi.org/10.1016/ s0300-483x(99)00217-6.
10. Ochi S, et al. Neural precursor cells are decreased in the hippocampus of the delayed carbon monoxide encephalopathy rat model. Sci Rep. 2021;11:6244. https://doi.org/ 10.1038/s41598-021-85860-9.
11. Thom SR, Bhopale VM, Fisher D, Zhang J, Gimotty P. Delayed neuropathology after carbon monoxide poisoning is immune-mediated. Proc Natl Acad Sci U S A. 2004; 101:13660-5. https://doi.org/10.1073/pnas.0405642101.
12. Ochi S, et al. The nicotinic cholinergic system is affected in rats with delayed carbon monoxide encephalopathy. Neurosci Lett. 2014;569:33-7. https://doi.org/10.1016/j. neulet.2014.03.054.
13. Sekiya K, et al. Carbon monoxide poisoning-induced Delayed encephalopathy ac- companies decreased microglial cell numbers: distinctive pathophysiological fea- tures from hypoxemia-induced brain damage. Brain Res. 2019;1710:22-32. https://doi.org/10.1016/j.brainres.2018.12.027.
14. Kao LW, Nanagas KA. Carbon monoxide poisoning. Emerg Med Clin North Am. 2004;

22:985-1018. https://doi.org/10.1016/j.emc.2004.05.003.

1. Bol O, Koyuncu S, Gunay N. Prevalence of hidden carbon monoxide poisoning in auto service workers; a prospective cohort study. J Occup Med Toxicol. 2018;13:35. https://doi.org/10.1186/s12995-018-0214-9.
2. Uchino A, et al. MRI of the brain in chronic carbon monoxide poisoning. Neuroradi- ology. 1994;36:399-401. https://doi.org/10.1007/BF00612127.
3. Raluy RG, Dias AC. Life cycle assessment of a domestic gas-fired water heater: influ- ence of fuel used and its origin. J Environ Manage. 2020;254:109786. https://doi.org/ 10.1016/j.jenvman.2019.109786.
4. Wang C, et al. Co-exposure to fluoride and sulfur dioxide on histological alteration and DNA damage in rat brain. J Biochem Mol Toxicol. 2018;32. https://doi.org/10. 1002/jbt.22023.
5. Yu K, et al. Association of Solid Fuel use with Risk of cardiovascular and all-cause mortality in rural China. JAMA. 2018;319:1351-61. https://doi.org/10.1001/jama. 2018.2151.