The effects of hyperbaric oxygen dosing on the delayed neuropsychiatric sequalae of carbon monoxide poisoning

Introduction: A paucity of evidence exists as to the comparative effectiveness of various hyperbaric oxygen regimens in preventing delayed neuropsychiatric sequelae. We aimed to compare the effect of four such treatment regimens: (1) PO₂ = 2.0_ATA for 90 min once; (2) PO₂ = 2.0_ATA for 90 min thrice; (3) PO₂ = 2.8_ATA for 90 min once; and (4) PO₂ = 2.8_ATA for 90 min once followed by two sessions of 2.0_ATA. Methods: We retrospectively reviewed the records of all patients treated in a large regional hyperbaric referral facility over the past 30 years, and identified patients displaying any new cognitive, motor or psychiatric symptoms within 2–40 days from exposure to carbon monoxide–any of which was defined as delayed neuropsychiatric sequelae. Excluded were patients not complying with the full prescribed treatment course or those lacking a full medical record for at least a year following exposure. Results: Of 312 patients included in the final analysis, the incidence of delayed neuropsychiatric sequelae was 31/87 (36%), 20/54 (37%), 20/102 (20%) and 13/69 (19%) in the 2.0_ATA once, 2.0_ATA thrice, 2.8_ATA once and 2.8_ATA thrice treatment groups, respectively (P = 0.011). Patients treated with the lower-pressure regimen (2.0_ATA once and thrice) had a significantly higher delayed neuropsychiatric sequelae rate compared to those treated with higher PO₂ (2.8_ATA once and thrice) − 36.2% versus 19.3% respectively (P = 0.0013). In a multivariate logistic regression model adjusting for age, carboxyhemoglobin levels, and presenting symptoms, the higher-pressure protocol was independently associated with a 55% reduction in the odds of developing delayed neuropsychiatric sequelae (adjusted odds ratio = 0.45; 95% confidence interval: 0.26–0.77; P = 0.004). Increasing the frequency of sessions from one to three was not associated with a statistically significant benefit (P = 0.9). The presence of seizures on admission was the strongest predictor of poor outcome (adjusted odds ratio = 4.27; P = 0.023). Discussion: The observed 55% reduction in the odds of developing delayed neuropsychiatric sequelae with the 2.8_ATA protocol suggests that achieving a higher initial partial pressure of oxygen is critical to effectively interrupting the inflammatory cascades associated with carbon monoxide toxicity. Because additional sessions did not confer a statistically significant benefit, meeting this initial therapeutic threshold appears more impactful than treatment frequency. Notably, the neuroprotective benefits of the higher-pressure regimen were sustained even among high-risk patients presenting with seizures. Conclusions: A PO₂ of 2.8_ATA rather than 2.0_ATA in the initial hyperbaric oxygen session, is associated with superior neuroprotective outcomes, while the overall number of sessions is not.