Prediction of delayed neuropsychiatric sequelae after carbon monoxide poisoning via serial determination of serum neuron-specific enolase levels

Abstract

Background

Neuron-specific enolase (NSE) is released into serum when nerve cells are damaged, and the levels thereof are used to determine neurological prognosis in patients who have suffered cardiac arrest or stroke. Delayed neuropsychiatric sequelae (DNS), a major complication of carbon monoxide poisoning (COP), can be caused by inflammatory response which is a mechanism of neuronal injury in cardiac arrest and stroke. NSE is known as a predictor of neurological prognosis in ischemic brain injury after cardiac arrest, and it is also reported as a predictor of DNS in acute COP. When serum NSE is measured serially in cardiac arrest patients, the best time to predict neurological prognosis is known at 48–72 h, but there are no studies analyzing serial serum NSE in acute COP. Thus, we explored whether serum NSE levels measured three times at 24 h intervals after COP predicted the development of DNS.

Methods

This prospective observational study was conducted on patients treated for COP from May 2018 to April 2020 in a tertiary care hospital in Korea. Neuron-specific enolase levels were assessed 24, 48, and 72 h after presentation at hospital. We used logistic regression to explore the association between NSE levels and DNS development.

Results

The NSE level was highest at 48 h, and the difference between the DNS group and the non-DNS group was greatest on the same time point. On multivariable logistic regression analysis, the NSE level at 48 h of >20.98 ng/mL (odds ratio [OR], 3.570; 95% confidence interval [CI], 1.412–9.026; P = .007) and the initial Glasgow Coma Scale (GCS) score of <9 (OR, 4.559; 95% CI, 1.658–0.12.540; P = .003) was statistically significant for DNS development.

Conclusion

Early identification of those who will experience DNS in acute COP patients is clinically important for deciding treatment. In this study, we revealed that NSE level of >20.98 ng/mL at 48 h time point can be used as an independent predictor of DNS (OR, 3.570; 95% CI, 1.412–9.026; P = .007; AUC, 0.648).

Keywords

Neuron-specific enolase carbon monoxide neuropsychological sequelae poisoning neurotoxicity

Introduction

Carbon monoxide (CO) is a colorless, tasteless, odorless toxic gas produced from incomplete combustion of carbon-based fuels and materials.¹ Carbon monoxide binds to hemoglobin with an affinity approximately 250-fold greater than that of oxygen. Even a low amount of inhaled CO can trigger hypoxia via the formation of carboxyhemoglobin (COHb), thus shifting the oxyhemoglobin dissociation curve to the left.² Due to these characteristics, carbon monoxide poisoning (COP) is one of the leading causes of mortality and morbidity from poisoning, with incidence and mortality rates of 137 and 4.6 per million worldwide, respectively.^3,4 Even a small amount of CO can easily damage the brain, which is particularly vulnerable to hypoxia if oxygen-carrying capacity is reduced.⁵ Also, CO can cause inflammation through multiple pathways which are independent to hypoxia, resulting in cardiac and neurologic injury.⁶

Delayed neuropsychiatric sequelae (DNS) caused by COP are defined as a set of brain injuries that develop after a lucid interval of several days to 6 weeks after recovery from hypoxic injury.⁷ Signs of DNS may include memory and movement disorders, psychosis, and Parkinsonism.^6,8 The etiology of DNS remains incompletely understood, but it is assumed that the brain is injured by hypoxia with inflammatory reactions mediated by CO.⁶ Because of these lethal properties of DNS, previous studies suggested factors such as Glasgow Coma Scale (GCS) score, S100B protein, lactic acid, and acute brain lesion on magnetic resonance imaging (MRI), as predictors.^9–12

Neuron-specific enolase (NSE), one of the several isoenzymes of enolase, is a glycolytic enzyme; an isoform is found in the neuronal cytoplasm of the central nervous system. Also, NSE has attracted attention as an adjunct test for the small cell lung carcinoma, Creutzfeldt–Jakob disease and neuroendocrine tumors.¹³ The enzyme is thought to be released from neuronal and glial tissues into the blood when cell membrane integrity is lost, so diagnosis or prognosis of brain damage can be predicted by measuring the serum NSE level.¹⁴ Thus, in patients with ischemic brain injury developing after cardiac arrest, or ischemic stroke caused by ischemic–reperfusion injury and inflammation, the enzyme level serves as a marker for neurological damage.^13–15 Delayed neuropsychiatric sequelae may occur not only by similar mechanism of cardiac arrest and stroke, but also by intracellular oxidative stress, impaired mitochondrial function, lipid peroxidation, and apoptosis which are caused by CO.⁶ We thus hypothesized that the NSE level would predict DNS development. Previous studies on the relationship between the NSE level and DNS checked the enzyme levels only on initial presentation.^10,16 When NSE levels were measured at 24 h intervals to predict neurological prognosis after cardiac arrest, the most useful value was measured at 48–72 h time point.^17,18 Therefore, we explored whether any serum NSE level measured three times at 24 h intervals to 72 h after presentation at hospital, or any other clinical parameter, was prognostic of DNS.

Methods

Study design and setting

This prospective observational study was conducted using a CO registry that we established for all COP patients who visited the emergency department (ED) of a tertiary care hospital located in Gyeonggi-do, Korea; our hospital treats approximately 70,000 patients each year. This study was approved by the Institutional Review Board of Soonchunhyang University Hospital and permission was also granted by the Medical Ethics Committee.

Participant selection

All COP patients visiting our ED were enrolled in the CO registry. Patients who described appropriate histories of CO exposure, exhibited the expected signs on physical examination, and exhibited COHb levels >5% for non-smokers (COHb levels >10% for smokers) on presentation to the ED were thought to be suffering from COP. We enrolled all COP patients treated from May 2018 to April 2020. The exclusion criteria were age <18 years, the lack of serum NSE data due to unintentionally omitted blood sampling (more than one time), discharge against medical advice, loss to follow-up, persistent neurological symptoms on discharge, and presentation more than 24 h after CO exposure.

Clinical and laboratory assessments

Data on vital signs, demographic data, medical comorbidities, GCS scores, whether COP was accidental or intentional, accompanying symptoms, and laboratory results were collected. Venous blood was collected on presentation at our ED. And for determining changes in serial serum NSE levels, venous blood was also drawn three times at 24 h intervals from 24 h after presentation at the hospital. Serum NSE levels (reference range, 0–16.33 ng/mL) were determined using a Cobas 8000 analyzer (Roche Diagnosis, Indianapolis, IN, USA). All patients were scheduled for MRI within 2 days of entry into the ED.

DNS were defined as neurological abnormalities developing within 3 months after discharge and included symptoms such as depression, psychosis, urinary or fecal incontinence, disturbed consciousness, mutism, gait disturbance, memory disorder, and Parkinson-like syndrome.^8,19 To evaluate cognitive ability, neurologist-administered Mini-mental state examination was performed and a score lower than 24 was defined as cognitive disorder.²⁰ And patient who have problem in recalling a word list was defined to have a memory disorder.²⁰ Psychiatric disease was defined by the criteria of the Diagnostic and Statistical Manual of Mental Disorders-5. And Parkinsonism was defined as having two or more of the symptoms like bradykinesia, resting tremor, impairment of postural reflexes, and rigidity. Neurological symptoms and signs were prospectively estimated at discharge. At 1 month after discharge, patients visited the ED and saw emergency physicians. Telephone interviews exploring DNS development were also conducted with all patients through cross-checking with their families at 2 weeks, 6 weeks, and 3 months after discharge. Patients and their families were also educated to revisit hospital if DNS symptoms were suspected to develop within 3 months after discharge. Re-evaluating DNS was based on additional examinations, including a mini-mental state examination, brain MRI, and consultations with neurologists and psychiatrists. Since some of the symptoms of DNS can be mistaken for psychiatric diseases, we checked if there was any history of psychiatric disease such as a depression by consulting a psychiatrist at the time of arrival at the ED or early in hospital admission. And, if patients complained of DNS suspicious symptoms within 3 months, the development of DNS was decided under the assessment of the same psychiatrist who had previously met the patient.

Management of carbon monoxide poisoning

All COP patients received 15 L of oxygen/min through a non-rebreather mask in the ED prior to hyperbaric oxygen (HBO) therapy. According to the Undersea and hyperbaric medical society indications, HBO therapy was delivered if the initial COHb was ≥25% (COHb ≥15% for pregnant women), if there was any history of loss of consciousness or neurological abnormalities (including decreased consciousness) regardless of COHb levels, and/or if acute cardiac injury was obvious.²¹ Hyperbaric oxygen therapy was delivered within 30 min of arrival at the ED considering contraindications (three sessions at intervals of 6–12 h within a single day). The first session lasted for 150 min, with a target pressure of 3atmosphere absolute (ATA). The subsequent sessions lasted for 120 min at 2 ATA.²² Compression was conducted over the first 30 min of each session and decompression from 30 min before the end of each session.

Statistical analysis

Data are expressed as medians with interquartile ranges (IQRs) for continuous variables and as absolute numbers or relative frequencies for categorical variables. Patients were categorized into two groups (non-DNS and DNS). Continuous variables that were normally distributed were compared using Student’s t test and those that were not normally distributed using the Mann–Whitney U test. Pearson’s χ² test and Fisher’s exact test were used to compare categorical variables. To obtain the optimal cutoff value of the serum NSE level to predict DNS, the receiver operating characteristic (ROC) curve was determined for each 24 h serial serum NSE levels, respectively, and the results were obtained using the Youden index. The results showed area under curve (AUC), sensitivity (Se) and specificity (Sp). Multivariable logistic regression analysis was performed to identify predictive factors and a statistical significance of P-value <.05 was set for variables between the two groups in the univariable analysis and a factor divided by the cutoff value of serum NSE. Adjusted odds ratios (ORs) with 95% confidence intervals (CIs) were calculated using logistic regression analysis. And P-value <.05 was taken to indicate statistical significance. All statistical analyses were performed with the aid of SPSS for Windows software ver. 26.0 (IBM Corp., Armonk, NY, USA).

Results

From May 2018 to April 2020, a total of 236 patients visited our ED with COP. Of these, 49 were excluded for the following reasons: five were <18 years of age, seven lacked serum NSE data due to unintentionally omitted blood sampling, 12 were discharged against medical advice, 14 were lost to follow-up, seven exhibited persistent neurological symptoms at discharge, and four presented more than 24 h after COP (Figure 1). Therefore, 187 patients were finally included.

Figure 1.

Flow chart of the study subjects. Note: COP: carbon monoxide poisoning; NSE: neuron-specific enolase; DNS: delayed neuropsychiatric sequelae.

Table 1 describes the general characteristics of the patients. The median age was 42 years and males comprised 68.5% of the patients. The rates of ischemic stroke and depression were 2.2% and 42.8%, respectively; 54.7% of the patients were smokers. The initial GCS score was 14 (IQR, 11–14). Of all patients, 136 (72.7%) had intentionally exposed themselves to CO. DNS developed in 25 patients (13.4%) at a median time to onset of 30 days (IQR, 12.5–30 days).

Table 1.

General characteristics of the study subjects.

	Total (n = 187)
Age, years	42 [30–54]
No. of males, n (%)	128 (68.5)
BMI, kg/m²	23.0 [20.9–25.4]
Comorbidities, n (%)
Hypertension	17 (9.1)
Diabetes mellitus	10 (5.4)
Myocardial infarction	5 (2.7)
Ischemic stroke	4 (2.2)
Depression	80 (42.8)
Current smoker, n (%)	94 (54.7)
Initial GCS score	14 [11–14]
Duration of CO exposure, min	120 [43.8–240]
Intentional exposure, n (%)	136 (72.7)
HBO therapy, n (%)	171 (91.4)
Development of DNS, n (%)	25 (13.4)
Time to onset of DNS, days	30 [12.5–30]

Note: Values are expressed as median [interquartile range], or number (proportion).

Note: BMI: body mass index; GCS: Glasgow Coma Scale; CO: carbon monoxide; HBO: hyperbaric oxygen; DNS: delayed neuropsychiatric sequelae.

The differences between the non-DNS and DNS groups are listed in Table 2. The median GCS scores of the DNS and non-DNS groups were 10 and 14, respectively; thus, lower in the DNS group. The CO exposure time was longer in the DNS group at 195 min (IQR, 90–480 min) versus 120 min (IQR, 42.5–240 min) in the non-DNS group. The C-reactive protein level was also higher in the DNS group at 0.39 mg/dL (IQR, 0.08–0.76 mg/dL) versus 0.1 mg/dL (IQR, 0.04–0.29 mg/dL) in the non-DNS group. The trends in serum NSE levels over time in the non-DNS and DNS groups are shown in Figure 2. From 24 h to 48 h time point, the NSE levels increased, and then fell on 72 h time point. The NSE levels at 48 h and 72 h differed significantly between the non-DNS and DNS groups (P = .02 and .03, respectively).

Table 2.

Comparison of clinical variables between the non-DNS and DNS groups.

	Non-DNS (n = 162)	DNS (n = 25)	P-value
Age, years	41.5 [29–52.75]	43 [33–55]	.58
No. Of males, n (%)	113 (69.8)	15 (60)	.46
BMI, kg/m²	23 [20.9–25.3]	22.95 [20.1–25.4]	.74
Initial GCS score	14 [12–14]	10 [7–14]	<.01
Vital signs
Systolic BP, mmHg	130 [115.5–140]	130 [115–145]	.68
Diastolic BP, mmHg	80 [79–90]	80 [70–95]	.69
Heart rate, beats/min	94 [80–103]	94 [84–107]	.43
Respiratory rate, breaths/min	20 [19–20]	20 [18–20]	.48
Oxygen saturation, %	98 [96–99]	97 [94–98]	.36
Comorbidities, n (%)
Hypertension	14 (8.6)	3 (12)	.71
Diabetes mellitus	10 (6.2)	0 (0)	.36
Myocardial infarction	3 (1.9)	2 (8.0)	.13
Ischemic stroke	3 (1.9)	1 (4.2)	.43
Depression	66 (40.7)	14 (56)	.22
Current smoker, n (%)	80 (54.4)	14 (56)	>.99
Duration of CO exposure, min	120 [42.5–240]	195 [90–480]	.27
Intentional exposure, n (%)	117 (72.2)	19 (76)	.88
Symptoms, n (%)
Headache	20 (12.4)	1 (4)	.32
Loss of consciousness	45 (27.8)	11 (44)	.16
Dizziness	19 (11.7)	3 (12)	>.99
Chest pain	5 (3.1)	0 (0)	>.99
Laboratory findings
COHb, %	11.1 [4.65–17.7]	12.2 [9.5–15.6]	.14
White blood cells, ×10³/mm³	11.86 [8.17–15.37]	12.75 [9.81–19.71]	.14
Creatinine, mg/dL	1 [0.9–1.2]	1.1 [0.9–1.2]	.47
Arterial pH	7.41 [7.38–7.44]	7.41 [7.38–7.45]	.91
C-reactive protein, mg/dL	0.1 [0.04–0.29]	0.39 [0.08–0.76]	.02
Lactate, mmol/L	2.35 [1.5–4.4]	2.6 [1–4]	.60
Troponin I, ng/mL	0.1 [0.1–0.2]	0.1 [0.1–0.6]	.21
NSE level, ng/mL
24 h NSE	18.18 [13.72–22.39]	19.82 [16.07–24.72]	.24
48 h NSE	18.19 [14.46–23.87]	23.02 [17.5–27.96]	.02
72 h NSE	15.11 [11.8–19.5]	18.83 [12.92–24.16]	.03
HBO therapy, n (%)	148 (91.4)	23 (92)	>.99
Abnormal finding on brain MRI, n (%)	27 (27.6)	7 (36.8)	.59

Note: Values are expressed as median [interquartile range], or number (proportion).

Note: DNS: delayed neuropsychiatric sequelae; BMI: body mass index; GCS: Glasgow Coma Scale; BP: blood pressure; CO: carbon monoxide; COHb: carboxyhemoglobin; NSE: neuron-specific enolase; HBO: hyperbaric oxygen; MRI: magnetic resonance imaging.

Figure 2.

A line plot of neuron-specific enolase levels by neurological outcome. Note: DNS: delayed neuropsychiatric sequelae. Dots indicate median values.

The AUC were 0.574, 0.648 and 0.634 for each 24 h serial serum NSE levels, respectively. And the optimal cutoff values were 33.92 (Se 0.240, Sp 0.963), 20.98 (Se 0.680, Sp 0.660) and 16.79 ng/mL (Se 0.720, Sp 0.623) for each 24 h serial serum NSE levels, respectively (Figure 3).

Figure 3.

ROC curves for the serum NSE level and development of DNS following acute COP. (a) ROC curve for the 24 h NSE level and development of DNS. The AUC is 0.574. The best cutoff value in 33.92 ng/mL with a sensitivity of 24.0% and a specificity of 96.3%. (b) ROC curve for the 48 h NSE level and development of DNS. The AUC is 0.648. The best cutoff value in 20.98 ng/mL with a sensitivity of 68.0% and a specificity of 66.0%. (c) ROC curve for the 72 h NSE level and development of DNS. The AUC is 0.634. The best cutoff value in 16.79 ng/mL with a sensitivity of 72.0% and a specificity of 62.3%. Note: ROC: receiver operating characteristic; NSE: neuron-specific enolase; DNS: delayed neuropsychiatric sequelae; COP: carbon monoxide poisoning; AUC: area under curve.

On multivariable logistic regression analysis, the adjusted OR of 48 h serum NSE level (>20.98 ng/mL) for development of DNS was 3.570 (95% CI, 1.412–9.026; P = .007). And adjusted OR of initial GCS score (<9) for development of DNS was 4.559 (95% CI, 1.658–12.540; P = .003) (Table 3).

Table 3.

Multivariable logistic regression analysis of potential risk factors for DNS in patients with acute COP.

	Multivariable analysis
	Multivariable analysis
	OR (95% CI)	P-value
Initial GCS score (<9)	4.559 (1.658–12.540)	.003
48 h NSE (>20.98 ng/mL)	3.570 (1.412–9.026)	.007

Note: OR: odds ratio; CI: confidence interval; GCS: Glasgow Coma Scale; NSE: neuron-specific enolase.

Discussion

We explored whether the serum NSE level, measured three times serially at 24 h intervals from 24 h after presentation, could predict DNS in patients with COP. From 24 h to 72 h time point, the NSE levels changed from 19.82 to 23.02 and 18.83 ng/mL in the DNS group and from 18.18 to 18.19 and 15.11 ng/mL in the non-DNS group. The levels and difference between the non-DNS and DNS groups were greatest at 48 h time point. Our study revealed that serum NSE level of >20.98 ng/mL at 48 h time point which was verified by ROC analysis can be used as an independent predictor of DNS (OR, 3.070). To the best of our knowledge, this is the first study to explore the relationship between 24 h serial serum NSE levels and DNS development.

Zhang et al.²³ found that both the GCS scores on site and in the ED independently predicted DNS. Ku et al.²⁴ also reported similar result that lower GCS scores could be a meaningful predictor for developing DNS. We found that the initial hospital GCS score was lower in the DNS group and predictive of DNS. A low GCS score reflects CO exposure for a rather long time, which causes severe brain injury. The cerebral hypoxia and inflammatory response induced by CO can injure brain cells and trigger decreased consciousness.⁶ In our study, the rate of intentional exposure was 72.7%, which is more than the majority. Although the toxic co-ingestion rate was not obtained in our study, according to previous studies, toxic co-ingestion was relatively common in patients with intentional COP.²⁵ Also, patients who attempted suicide are usually uncooperative to the medical staff.²⁶ This can lead to clinical exaggeration of GCS scores which, while being statistically meaningful, could lead to bias in the study results.

Several studies have explored the relationship between COP and serum NSE levels. Yarden et al. found that unconscious patients with acute COP had higher serum NSE levels than did conscious patients.⁹ We found that the GCS score was lower in the group with a high serum NSE level, suggesting that the neuronal injury caused by hypoxia and inflammation was sufficiently severe to cause a change in consciousness. Cha et al.¹⁰ reported that the initial serum NSE measured in the ED served as an early predictor of DNS development. Cha et al. measured the serum NSE level only once at the initial presentation. However, according to previous studies, the best time to measure serum NSE for predicting neurological prognosis in ischemic brain injury after cardiac arrest was 48–72 h.^17,18 Therefore, we measured the serial serum NSE level in acute COP patients to confirm which time point is the best to check the neurological prognosis, reflecting previous results. The etiology of DNS remains unclear, but brain hypoxia, lipid peroxidation, increased levels of reactive oxygen species, and the inflammatory response to COP can injure neuronal cells.^6,27 In previous studies, the DNS group experienced more severe neuronal injury compared to the non-DNS group as evidenced by the higher serum NSE levels.^10,28 In another study, the serum NSE level was elevated in children with hypoxic ischemic encephalopathy caused by COP.²⁹

In our study, the NSE level at 48 h after initial presentation was higher than that measured 24 h after presentation, and the difference between the DNS and non-DNS groups was also greatest at 48 h time point. An earlier study investigated the association between the NSE level and neurological prognosis after cardiac arrest; an elevated serum NSE level at 48 h after arrest predicted a poor outcome.¹⁷ Although the pathophysiological mechanisms underlying NSE release when cerebral ischemia develops after COP or cardiac arrest remain unclear, several studies on focal cerebral ischemia indicated that the NSE release profile may reflect the several steps of ischemic brain injury evolution. The first peak may reflect rapid NSE release from initially injured tissue, followed by a second peak reflecting secondary brain injury caused by the inflammatory response that triggers neuronal cell death or persistent disturbance of the blood–brain barrier.^30–32 The NSE release pattern after COP is thought to be similar; the fact that the NSE level was highest at 48 h time point might indicate that brain injury caused by inflammation (via multiple pathways) developed after initial ischemia.⁶

Moon et al. reported that the serum NSE level at presentation was not associated with the risk of DNS in 236 patients with acute COP. However, only the initial NSE level was measured.¹⁶ We sought to identify a time point at which serum NSE levels might differ markedly; we thus took three serial 24 h measurements. The 48 h time point values exhibited the greatest difference. Rasmussen et al. measured serum NSE levels at 12-h intervals in 20 COP patients. The level was approximately 10 ng/mL regardless of the COHb level and did not change over time.³³ However, the sample size was very small, perhaps explaining the absence of a significant relationship between the NSE level and DNS development.

Our work had several limitations. First, this was a single-center study; representativeness might be insufficient due to small sample size, so a larger multicenter study is required to obtain more reliable results. Second, although we investigated the relationship between serial NSE levels and DNS, we did not consider any biomarker other than NSE (such as the s100b protein, the level of which is predictsive of brain injury). Third, the cerebrospinal fluid NSE level is also elevated after neuronal cell injury.⁹ However, we did not analyze cerebrospinal fluid. Fourth, the AUC values range between 0.574 and 0.648, which are of inadequate predictive value of DNS occurrence. Fifth, GCS score was also found to be a predictor of DNS in our study. However, most of the patients enrolled in our study were intentionally exposed to CO for suicide, which could lead to bias in clinical examination or GCS score related to sedating drugs, alcohol, or patients’ uncooperative attitude. But, unlike clinical examination, concentration of biomarker such as NSE is unlikely to be affected by other factors and is easy to assess blindly. A well-designed large-scale study will be helpful in validating the findings of this study.

Conclusion

In patient with acute COP, serum NSE level of >20.98 ng/mL at 48 h time point can be used as an independent predictor of DNS (OR, 3.570; 95% CI, 1.412–9.026; P = .007; AUC, 0.648). The serum NSE levels, measured three times at 24 h intervals from 24 h after presentation at the hospital, were highest at 48 h time point, and the difference in levels between the DNS and non-DNS groups was greatest at the same time point.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Soonchunhyang University Research Fund (grant no. 10210035).

Ethical approval

The study protocol was approved by the Institutional Review Board of Soonchunhyang university hospital and conducted in accordance with the provisions of the Declaration of Helsinki.

ORCID iDs

Sangun Nah .

Sungwoo Choi .

Sangsoo Han .

References

Ernst

Zibrak

. Carbon monoxide poisoning. New Engl J Med 1998; 339: 1603–1608.

Rodkey

O'Neal

Collison

, et al. Relative affinity of hemoglobin S and hemoglobin A for carbon monoxide and oxygen. Clin Chem 1974; 20: 83–84.

Hampson

. U.S. mortality due to carbon monoxide poisoning, 1999–2014. accidental and intentional deaths. Ann Am Thorac Soc 2016; 13: 1768–1774.

Mattiuzzi

Lippi

. Worldwide epidemiology of carbon monoxide poisoning. Hum Exp Toxicol 2020; 39: 387–392.

Guzman

. Carbon monoxide poisoning. Crit Care Clin 2012; 28: 537–548.

Weaver

. Carbon monoxide poisoning. New Engl J Med 2009; 360: 1217–1225.

Sonmez

İşcanlı

Parlak

, et al. Delayed neurologic sequelae of carbon monoxide intoxication. Turk J Emerg Med 2018; 18: 167–169.

Choi

. Acute carbon monoxide poisoning and delayed neurological sequelae: a potential neuroprotection bundle therapy. Neural Regen Res 2015; 10: 36–38.

Yardan

Cevik

Donderici

, et al. Elevated serum S100B protein and neuron-specific enolase levels in carbon monoxide poisoning. Am J Emerg Med 2009; 27: 838–842.

10.

Cha

Kim

, et al. Serum neuron-specific enolase as an early predictor of delayed neuropsychiatric sequelae in patients with acute carbon monoxide poisoning. Hum Exp Toxicol 2018; 37: 240–246.

11.

Jeon

Sohn

Seo

, et al. Acute brain lesions on magnetic resonance imaging and delayed neurological sequelae in carbon monoxide poisoning. JAMA Neurol 2018; 75: 436–443.

12.

Kim

Choi

Park

, et al. Serum markers and development of delayed neuropsychological sequelae after acute carbon monoxide poisoning: anion gap, lactate, osmolarity, S100B protein, and interleukin-6. Clin Exp Emerg Med 2018; 5: 185–191.

13.

Zaheer

Beg

Rizvi

, et al. Correlation between serum neuron specific enolase and functional neurological outcome in patients of acute ischemic stroke. Ann Indian Acad Neurol 2013; 16: 504.

14.

Wiberg

Hassager

Stammet

, et al. Single versus serial measurements of neuron-specific enolase and prediction of poor neurological outcome in persistently unconscious patients after out-of-hospital cardiac arrest–a TTM-trial substudy. PLoS One 2017; 12: e0168894.

15.

Bai

Lyden

. Revisiting cerebral postischemic reperfusion injury: new insights in understanding reperfusion failure, hemorrhage, and edema. Int J Stroke 2015; 10: 143–152.

16.

Moon

Chun

Lee

, et al. Serum neuron-specific enolase levels at presentation and long-term neurological sequelae after acute charcoal burning-induced carbon monoxide poisoning. Clin Toxicol 2018; 56: 751–758.

17.

Storm

Nee

Jorres

, et al. Serial measurement of neuron specific enolase improves prognostication in cardiac arrest patients treated with hypothermia: a prospective study. Scand J Trauma Resusc Emerg Med 2012; 20: 1–6.

18.

Chung-Esaki

Mui

Mlynash

, et al. The neuron specific enolase (NSE) ratio offers benefits over absolute value thresholds in post-cardiac arrest coma prognosis. J Clin Neurosci 2018; 57: 99–104.

19.

Ide

Kamijo

Ide

, et al. Elevated S100B level in cerebrospinal fluid could predict poor outcome of carbon monoxide poisoning. Am J Emerg Med 2012; 30: 222–225.

20.

Galea

Woodward

. Mini-mental state examination (MMSE). Aust J Physiother 2005; 51: 198.

21.

Undersea HMSHOC Gesell

. Hyperbaric oxygen therapy indications: the hyperbaric oxygen therapy committee report. North Carolina, USA: Undersea and Hyperbaric Medical Society, 2008.

22.

Weaver

Hopkins

Chan

, et al. Hyperbaric oxygen for acute carbon monoxide poisoning. N Engl J Med 2002; 347: 1057–1067.

23.

Zhang

Jia

, et al. Multicenter retrospective analysis of the risk factors for delayed neurological sequelae after acute carbon monoxide poisoning. Am J Emerg Med 2021; 46: 165–169.

24.

Yang

Lee

, et al. Predictors of carbon monoxide poisoning-induced delayed neuropsychological sequelae. Gen Hosp Psychiatry 2010; 32: 310–314.

25.

Hampson

Bodwin

. Toxic CO-ingestions in intentional carbon monoxide poisoning. J Emerg Med 2013; 44: 625–630.

26.

Lee

Lin

Yeh

, et al. Etiology and outcome of patients presenting for poisoning to the emergency department in Taiwan: a prospective study. Hum Exp Toxicol 2008; 27: 373–379.

27.

Rose

Wang

, et al. Carbon monoxide poisoning: pathogenesis, management, and future directions of therapy. Am J Respir Crit Care Med 2017; 195: 596–606.

28.

Hassan

Abdelaleem

Hamdy

. A prospective comparative study between three chemical markers for predicting delayed neurological sequelae in patients with acute carbon monoxide poisoning of poison control center in Minia University Hospital. Ain Shams J Forensic Med Clin Toxicol 2018; 31: 23–32.

29.

Akelma

Celik

Ozdemir

, et al. Neuron-specific enolase and S100B protein in children with carbon monoxide poisoning: children are not just small adults. Am J Emerg Med 2013; 31: 524–528.

30.

Wunderlich

Ebert

Kratz

, et al. Early neurobehavioral outcome after stroke is related to release of neurobiochemical markers of brain damage. Stroke 1999; 30: 1190–1195.

31.

Wunderlich

Wallesch

Goertler

. Release of neurobiochemical markers of brain damage is related to the neurovascular status on admission and the site of arterial occlusion in acute ischemic stroke. J Neurol Sci 2004; 227: 49–53.

32.

Wunderlich

Lins

Skalej

, et al. Neuron-specific enolase and tau protein as neurobiochemical markers of neuronal damage are related to early clinical course and long-term outcome in acute ischemic stroke. Clin Neurol Neurosurg 2006; 108: 558–563.

33.

Rasmussen

Poulsen

Christiansen

, et al. Biochemical markers for brain damage after carbon monoxide poisoning. Acta Anaesthesiol Scand 2004; 48: 469–473.