Abstract
Severe carbon monoxide (CO) poisoning causes fulminant deaths in common environment as well as neurological sequelae to survivors. Prevention of delayed neurological syndrome (DNS) after exposure to CO, the most important sequela, is based up to date on hyperbaric oxygen administration. Nevertheless, its use remains controversial due to the lack of evidence regarding its efficacy. The aim of this review is to report therapies under investigation for preventing or improving DNS, some of them with promising results in humans.
Keywords
Introduction
Carbon monoxide (CO) poisoning is among the most frequent intoxications with potential severity attended in hospital emergency departments. The CO comes from incomplete combustion of hydrocarbons that can be generated in fires, stoves, boilers, chimneys, motors, and so on. In the United States, approximately 50,000 visits are reported in the emergencies for this intoxication and a mortality rate of 1000–2000 patients per year is calculated. In Spain, of 1105 cases of chemical poisoning described in 2016, 49% were due to CO. The vast majority were produced by fires (215 cases), predominating those of domestic origin and boilers as a source of combustion (56 cases). 1
For the diagnosis of this intoxication, the antecedent of exposure (duration and source) and the level of carboxyhemoglobin (COHb) in blood must be taken into account. 2 The treatment of this poisoning is based on oxygen therapy, either normobaric oxygen therapy (NBOT) or hyperbaric oxygen therapy (HBOT) when available. These treatments, by increasing the partial pressure of oxygen in plasma, make possible to accelerate the dissociation of the COHb. 1,3,4
The most severe sequela in patients who survive CO poisoning is DNS, consisting of behavioral changes, depression, memory loss, language impairment, learning, urinary or fecal incontinence, gait disturbance, and Parkinson’s syndrome, 1,5 present from the first days after poisoning until several months later.
At present, there is no evidence of the best therapy to prevent the emergence of DNS. We have developed this narrative review to discuss the current knowledge situation given the uncertainty that offers the prevention of DNS in these patients.
Material and methods
A narrative review was carried out by consulting the PubMed database for the search of articles. This form of the review was chosen given the absence of adequate clinical trials to perform a systematic review. We searched for original articles and cases with new therapies with the terms “CO poisoning and delayed neurological syndrome or neuroprotective effect and CO poisoning.” We selected studies of experimental character in humans and animals about the treatment and prevention of DNS, published since 1989. This year was the publication date of the first important report about the prevention of DNS by the use of the hyperbaric chamber. 6
One hundred forty-six articles were initially agreed upon. It was taken into account that results of cognitive functions had been determined. The authors independently selected those articles in which new therapies were tested for the treatment and prevention of DNS. Finally, 17 reports were eligible to be evaluated in this review (Table 1).
Treatments tested in patients with CO poisoning, excluding HBOT alone, for the prevention of DNS.
HBOT: hyperbaric oxygen therapy; NBOT: normobaric oxygen therapy; DNS: delayed neurological syndrome; CO: carbon monoxide; COHb: carboxyhemoglobin; MDA: malondialdehyde; NBP: N-butylphthalide; B12r: reduced hydroxocobalamin; EPO: erythropoietin; MRI: magnetic resonance imaging; MSCT: mesenchymal stem cells transplantation; ECMO: extracorporeal membrane oxygenation.
In addition, the Cochrane Library was revised in which the search was introduced with the term “CO poisoning.” One review was found.
Results
NBOT and HBOT
The most widely used treatment in CO poisoning is NBOT with 100% oxygen concentration. HBOT 1,3 is performed when available. There is controversy in the bibliography on the ability of the treatment with HBOT to prevent late neurological sequelae. HBOT consists of 100% oxygen concentration management with pressure at least 50% above atmospheric pressure.
Studies that demonstrate the benefit of HBOT against NBOT in preventing DNS
One of the first prospective trials on the effectiveness of HBOT versus NBOT was performed in 1995 by Thom et al. The HBOT treatment was based on the use of a pressure of 2.8 absolute atmospheres (ATA) for 30 min, followed by 2 ATA for 90 min. They defined the DNS as a recurrence of the initial symptoms or the development of new symptoms considered typical of the DNS, plus a deterioration in one or more scores of the subtests of the neuropsychological evaluation. They concluded that the treatment with HBOT significantly reduced the incidence of DNS, as long as it started in the first 6 h of CO exposure. 3
Weaver et al. 24 performed a double-blind randomized clinical trial comparing both therapies. They included patients with documented CO exposure plus some of the symptoms, such as loss of consciousness, confusion, or headache. The HBOT treatment consisted of three sessions at intervals of 6–12 h. The first session was initiated before the first 24 h from exposure to CO. They were exposed to 3 ATA for 60 min followed by 2 ATA for another 60 min in the first session. In two other 120-min sessions, they used 2 ATA to reduce possible oxygen toxicity. Intubated patients were ventilated with oxygen at 100% with the same pressure as nonintubated patients. The trial was stopped after the third interim analysis, as hyperbaric oxygen was considered effective. At 6 weeks, the neurological sequelae were less frequent in the hyperbaric treatment group (24% patients) compared to the normobaric (43.1%). On the other hand, the scores of the neuropsychological tests did not show statistically significant differences. Neurological sequelae at 6 and 12 months were less frequent in the hyperbaric treatment group.
The same authors 25 published a new article aimed at determining risk factors to develop cognitive sequelae at 6 weeks of intoxication and modification of that risk with the use of HBOT. The study was performed with the patients included in the previous study 14 and patients who were not randomized. Two hundred thirty-eight patients, 147 from the randomized trial and 91 nonrandomized patients, were evaluated. Of the 146 patients treated with NBOT, 41% presented neurological sequelae at 6 weeks versus 24% of patients treated with HBOT.
Risk factors for the development of DNS in patients who had not been treated with HBOT were older than 36 years, CO exposure interval greater than 24 h before HBOT, initial symptoms of memory loss, loss of consciousness greater than 60 min, initial COHb level of 25% or higher, and initial base excess less than 22 mmol/L.
Chan et al. 26 conducted a historical cohort study recruiting patients with CO poisoning for 10 years. Indications for hyperbaric therapy in CO poisoning were loss of consciousness, myocardial ischemia evidenced by thoracic symptoms or changes in electrocardiogram, COHb levels greater than 25% on arrival to the emergency room, and pregnant women. Patients received three 30-min sessions of hyperbaric therapy for three consecutive days, with a maximum pressure of 2.8 ATA. Of the total number of patients, 24 received treatment with HBOT and 69 remaining with NBOT. None of the patients treated with HBOT presented DNS. On the other hand, seven patients from the group that had not received HBOT presented symptoms compatible with DNS, confirmed by image tests. However, no statistically significant difference was demonstrated between the two groups.
Jurič et al. 27 conducted an experiment with cell cultures of cortical astrocytes of neonatal rats exposed to CO. A significant affectation was observed in the astrocytic neurotrophic activity. In addition, neurotrophins extraction was carried out. Neurotrophins are capable of mediating the differentiation and growth of neuronal populations, the formation of synapses, and plasticity. The use of HBOT produced a significant increase in the production of the neurotrophins. This benefit was observed when the cells were exposed to hyperbaric oxygen between 1 h and5 h after exposure to CO, with a maximum effect after 3 or 5 h.
Studies that find no benefit from HBOT versus NBOT to prevent DNS
A review of the Cochrane Library has been published with the aim of examining available randomized trials on the effectiveness of HBOT compared to NBOT for the prevention of DNS at 4–6 weeks in patients who have suffered acute CO poisoning. 28
The studies selected for this review met the following criteria: randomized controlled trials, with or without blinding, participants were to be nonpregnant adults, and patients with intoxication were randomized to treatment with NBOT or HBOT. The main outcome measure was the presence or absence of symptoms compatible with late neurological sequelae at 4–6 weeks of the treatment. Six clinical trials were selected.
All the included trials presented a considerable risk of bias. Four studies had minimal or no blinding caused by not using a hyperbaric chamber for the treatment with NBOT. In addition, they presented a high possibility of bias performance, attrition, and detection especially strong, since the results were reports of the patient of new symptoms. Only two trials kept blinding patients and evaluators, using the hyperbaric chamber in the two treatment groups.
All six trials included a total of 1997 patients, of whom 1335 were randomly allocated to HBOT or NBOT. The prevalence of symptoms after treatment at 4 and 6 weeks was 29% for patients treated with HBOT and 34% for the group submitted to NBOT. The joint analysis using a random-effects model showed that there was no statistically significant reduction in neurological sequelae with the HBOT treatment.
Although the clustered analysis of these six studies suggested a possible statistically significant difference between HBOT and NBOT (odds ratio 0.78; confidence interval 95% 0.54–1.12), the wide methodological and statistical heterogeneity of the six trials make this analysis difficult to interpret.
The authors concluded that existing randomized controlled trials provide contradictory results on the effectiveness of HBOT. According to the results obtained, the routine use of HBOT for the treatment of acute CO poisoning cannot be recommended. Patients with greater severity may be able to benefit from the HBOT treatment. Additional research is needed.
New therapies in preclinical studies
Hemin
Guan et al. 7 used hemin to examine the effect it produces on the induction of hemoxigenase 1 (HO-1) in a rat model with hippocampal injury following CO exposure. HO-1 is a stress protein inducible by chemical inductors, such as hydrogen peroxide, which can modulate the levels of blood proteins by degrading heme in free iron, biliverdin, and CO. HO-1 is also a cellular defense protein that limits the injury caused by oxidative stress, having anti-inflammatory, antiapoptotic, and antiproliferative actions.
In rats, exposure to hemin produced a higher elevation of HO-1 in the hippocampus. As a result, the rate of survival of the rats at 24 h was significantly higher with the pretreatment with hemin. The authors concluded that the induction of HO-1 through pretreatment with hemin can prevent hippocampal injury after CO poisoning and increase survival rates within 24 h of exposure. Although the mechanisms underlying the induction and neuroprotection of HO-1 are not accurately known, they have an antioxidant effect by lowering the level of malondialdehyde (MDA).
Another study conducted a preclinical trial using mice exposed to CO. 8 Hemin administration significantly reduced the mortality rate, improved symptoms of acute CO poisoning, and reduced the level of COHb and serum concentration of MDA. Histologically, in the group without treatment, it was observed that there was inflammation of the cells in the CA1 region of the hippocampus and that the number of pyramidal cells was lower. These findings improved in the groups treated with hemin.
N-butylphthalide
Some trials have suggested that N-butylphthalide (NBP) could increase brain flow and metabolism in an area of ischemia, inhibit platelet aggregation, protect mitochondrial function, and reduce oxidative and apoptotic effects.
Li et al. 9 performed a study in rats aimed at investigating the neuroprotection with NBP after exposure to CO. After the administration of NBP, apoptotic cells were significantly reduced and a significant difference was observed between this group and another group that did not receive NBP. Data suggest that NBP may inhibit CO-induced early neuronal apoptosis. Another preclinical trial with rats studied the underlying mechanisms in cognitive dysfunction following CO exposure and the viability of NBP treatment on the structural and functional deterioration of the hippocampus. 10 Spatial and memory learning tests were evaluated. The results suggested that the NBP can improve learning and memory in rats and that the neuroprotective effect could last at least 1 month after poisoning. The histological studies showed that the treatment with NBP may significantly alleviate the damage of the hippocampus.
Reduced hydroxocobalamin
Roderique et al. 13 proposed an antidote for CO poisoning as a solution of hydroxocobalamin and ascorbic acid. Reduced hydroxocobalamin (B12r) could facilitate CO conversion to carbon dioxide (CO2) in the blood, allowing a rapid decrease in CO level. The objective of his first experiment in vitro was to demonstrate the reduction of CO in human blood following the administration of B12r. The average life of the COHb was reduced from 33 min to 15 min when using the combination of oxygen (O2) and B12r.
The study was completed with laboratory animals. B12 r improved cerebral oxygen tension and reversal of loss of total cell count, demyelination, and reduction of microglial activity produced by CO exposure.
Erythropoietin
Erythropoietin (EPO) has an important role in the development of the nervous system, acting as a neurotrophic factor with the ability to induce neurogenesis. There are EPO receptors in neurons, astrocytes, and oligodendrocytes. In addition, EPO has the ability to cross the brain barrier at low speed.
Moallem et al. 14 assessed the mechanisms by which EPO antagonizes CO neurotoxicity in rats. EPO reduced the level of glial fibrillary acidic protein at low doses, of MDA at any dose, and of myeloperoxidase at high doses and protected the integrity of the brain barrier, although it did not demonstrate reducing cerebral edema.
Methane-rich saline
Methane is a simple aliphatic hydrocarbon with antioxidant, antiapoptotic, and anti-inflammatory properties. Therefore, methane is a potential treatment in pathologies with oxidative stress, as is the case of CO poisoning.
Fan et al. 16 demonstrated the hypothesis of the neuroprotective effect of the methane-rich saline solution after long-term CO exposure in rats with an intraperitoneal injection of methane. They performed several experiments assessing, learning, histological analysis, and inflammatory factors. Methane was able to attenuate DNS, improve learning ability, and reduce neuronal injury. The same group confirmed the neuronal protection exerted by methane-rich saline at histological levels. 17
Dexamethasone
The use of dexamethasone has been investigated based on the autoimmune mechanisms related to the myelin basic protein (MBP) and the demyelination that CO originates. 18 A study with rats exposed to CO using dexamethasone and hyperbaric chamber showed that groups treated with dexamethasone suffered less magnetic resonance imaging (MRI) disturbances. Early treatment with dexamethasone prevented the emergence of DNS by avoiding lipid oxidation and the action of MBP. Inhibition of the inflammatory response was observed, decreasing the permeability and edema of the hypoxic cells.
Hydrogen-rich saline
Hydrogen-rich saline solutions have antioxidant effects, decrease necrosis and neuronal apoptosis, and improve behavior disorders after CO poisoning in experimental studies performed with rats. This effect appears to be mediated by the reduction of reactive oxygen and antioxidant regulation of several enzymes. 19,20
New therapies tested in patients
N-butylphthalide
Wang et al. 11 conducted a prospective study with the aim of investigating the efficacy of the combined treatment of NBP and mesenchymal stem cell transplantation (MSCT) to patients with DNS.
Three treatment groups were designed: a therapy group combined with NBP and MSCT, an NSCT transplant group, and a control group, under hyperbaric oxygen therapy and symptomatic treatment. At 1, 3, and 6 months, the patients treated with the double therapy had a higher score in the cognitive tests with respect to the MSCT group and the control group. The response rate in MRI in the combined therapy group was 92.9%, compared to 60% of the MSCT group and to 38.5% of the control group.
Xiang et al. 12 performed a randomized, double-blind prospective study to assess the effect of NBP and HBOT on cognitive dysfunction of DNS in 301 patients. The authors showed that this combined treatment could significantly increase the average score in the cognitive tests and in the functional situation of the patients.
Erythropoietin
Pang et al. 15 developed the first prospective and randomized study in patients with CO poisoning treated with EPO. This treatment had only been previously applied in two patients, with favorable outcome. 29
They designed a randomized prospective study of 103 patients with CO poisoning, divided into two groups: a treatment group with EPO and a control group treated with saline. Patients treated with EPO obtained better scores in functional tests, preserving this improvement on day 30. Regarding the serum levels of the S-100 β brain damage marker, the EPO treatment group had a lower maximum level than the placebo group patients and took less time to return to the normal range.
Extracorporeal membrane oxygenation
Oxygenation by extracorporeal membrane oxygenation (ECMO) is a form of vital cardiopulmonary support. The blood drains from the organism and circulates through the system thanks to the action of a mechanical pump, passing through the membrane oxygenator, where the gaseous exchange occurs, eliminating the CO2 and incorporating O2 to bloodstream. 21
A case 30 has been reported in which venous arterial ECMO was used in a woman of 38 years poisoned by CO, with good outcome. The patient was followed for 6 months and did not show symptoms compatible with DNS. The ability of this therapy to eliminate CO should be assessed in studies designed for this purpose.
Acetylcholinesterase inhibitors
Yanagiha et al. 22 treated, with acetylcholinesterase inhibitors, two patients with DNS, who did not respond to HBOT. In both cases, there was a favorable clinical response and a notable improvement in brain imaging studies in the prolonged follow-up.
We do not know the mechanism of action of acetylcholinesterase inhibitors on the neuropathological effects caused by CO. They could improve the acetylcholinergic neuronal function of the hippocampus, decreased after CO poisoning, as well as improve the antiapoptotic effect, the stimulation of nicotinic acetylcholine receptors, and the elevation of the expression of Bcl-2 protein, related to apoptosis. The authors suggested that galantamine hydrobromide has greater efficiency due to its allosteric and protective effects against nicotinic acetylcholinergic neuronal damage.
Methylprednisolone and memantine hydrochloride
Iwamoto et al. 23 treated a patient with DNS after CO poisoning administering methylprednisolone pulses and memantine hydrochloride (NMDA receptor inhibitor). This treatment produced a significant improvement of DNS, image tests, and cerebral blood flow determined by 99mTc-ECD SPECT.
Conclusions
There is currently no effective therapy for the prevention of DNS, one of the most serious complications following CO exposure. Conventional therapy is based on the use of oxygen therapy, normobaric or hyperbaric when available. After reviewing the available bibliography, it has not been demonstrated with adequate clinical trials the superiority of HBOT on NBOT in the prevention of DNS, although it should be considered HBOT in seriously poisoned patients. It remains to be established through clinical trials, the benefit of other available oxygenation therapies, such as ECMO in CO poisoning.
Preclinical trials with different substances show encouraging results for the development of effective therapy for DNS in the future, but clinical trials designed to do so are needed to demonstrate its clinical efficacy.
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) received no financial support for the research, authorship, and/or publication of this article.
