To date, the relationship between structural changes and altered cognitive function in DE and non-DE patients has not been fully studied. In this first VBM study which compares these two groups, morphological differences were found between DE and non-DE patients, and many of the structural deficits of DE patients correlated with their lower level of cognition. The DE group demonstrated lower scores in the WAIS tests, with poor perceptual organization (Picture Completion) , and slow processing speed (Digit Symbol and Symbol Search) . Carbon monoxide poisoning-related cognitive impairment includes impaired memory and executive function , reduced mental processing speed, and decreased intellectual function , which are consistent with the NP results in this study. However, the memory function has not been assessed in this study that may be seen as a limitation. The results of this study further support previous findings indicating that delayed CO encephalopathy might impair the frontal executive function and persist even after the recovery of other neurological deficits . Increased structural alterations in DE patients corroborate the NP sequelae in patients with chronic status after CO intoxication. Long-term follow-up for this particular group is required.
Consistent with the hypothesis posed in this study, structural differences between the DE and non-DE groups were found. In the non-DE group, only one significant cluster of lower GMV than controls was observed, it’s location in the left post-central gyrus. In contrast, structural abnormalities in the DE group included lower volume of the left claustrum, right amygdala, right caudate body, left hippocampus, right inferior parietal lobule, left superior frontal gyrus, left medial frontal gyrus, right anterior cingulate, left post-central gyrus, and left mammillary body. Previous studies report generalized brain atrophy occurring in chronic CO intoxicated patients, with volume reductions in the fornix , hippocampus , corpus callosum , basal ganglia, and cerebral and cerebellar cortex. However, detailed lobar location of cortical involvement has never been studied . In this study, the impact of DE on the brain’s gray matter, which had not been documented before, was clearly demonstrated. This VBM study provides a highly objective method for clarifying significant brain atrophy in the DE group, compared to the non-DE counterpart.
The pathogenesis of DE after CO intoxication is likely multifactorial, involving mitochondrial damage combined with hypoxemia and hypoperfusion during the acute stage [39–41], as well as glutamatergic excitation  and lipid peroxidation after return to CO-free air . Although the exact mechanism in the development of DE is unclear, a COHb concentration level that is higher in the DE group than in the non-DE group in this study might increase the occurrence of delayed symptoms and suggest a more obvious effect on the brain [5–7]. The pathologic hallmark is extensive demyelination, and current theories for the pathogenic mechanism include direct toxicity effects of CO, cerebral blood vessel damage, cerebral edema, and a hypersensitive reaction . Individual self-protective factors like tolerance to hypoxia and hypoperfusion or resistance to CO cytotoxic action in WM [4, 8, 10], may be responsible for the presence and variable duration of clear periods in DE.
In this study, two locations display a lower GMV in the DE group compared to the non-DE group: the left anterior cingulate (BA32) and the right amygdala. Both are a part of the brain limbic system. Past neuroimaging research suggests the anterior cingulate cortex to be part of the circuit involved in a type of attention  that regulates both cognitive and emotional processing . The affective subdivision of the anterior cingulate connects to the amygdala , and modulation of the amygdala-anterior cingulate connections seems to be a key substrate of emotional attention bias [48, 49]. Lesions involving the anterior cingulate and amygdala is believed to disrupt and cause affective biases . In this study, the DE group presented lower levels of performance in digit symbol and symbol search, indicating a slower processing speed, which is thought to be connected with a poor attention function . This result is consistent with previous concept, even though there is no significant correlation between the cluster volume in the left anterior cingulate, right amygdala, and the reported NP tests.
Although this study offers valuable insights into the cortical involvement in CO intoxication issues, it nevertheless has some limitations. First, our sample size is too small for definite conclusion of DE and non-DE. Patients without DE were difficult to contact and often rejected invitations to join the non-invasive research because they have little concern about following-up results in the chronic stage. Therefore, the number of patients in the non-DE group was smaller than the number of patients in the DE group, even though the incidence of DE is approximately 10% in all CO-intoxicated patients . Second, varying treatments for CO-intoxicated patients exist, such as whether they receive hyperbaric oxygen therapy or not. Moreover, the initial laboratory data were not thoroughly collected for every subject in the emergency room. Therefore, initial disease severity could not be compared to long-term outcomes to define a clearer relationship.