Increased long-term mortality among survivors of acute carbon monoxide poisoning *

Abstract

Recent data suggest that patients surviving acute carbon monoxide (CO) poisoning (COP) may have increased risk for long-term mortality. The objective of this study was to analyze long-term mortality of a large population of CO-poisoned patients treated at one medical center over three decades. Retrospective cohort study of patients treated with hyperbaric oxygen and surviving the acute poisoning episode. Long-term mortality was compared to a standard population. Comparison of mortality within the cohort by clinical indicators of poisoning severity was assessed using Cox proportional hazards regression analysis. Regional referral center for hyperbaric treatment of COP. One thousand seventy-three patients aged >or=18 years treated from 1978 to 2005. All patients received hyperbaric oxygen treatment. During 11,741 person-years of follow-up, 162 subjects died. The expected number of deaths was 87 (standardized mortality ratio [SMR]), 1.9; 95% confidence interval [CI], 1.6-2.2). Most of the excess mortality was in the group treated initially for intentional COP (58 excess deaths; SMR, 3.7; 95% CI, 2.9-4.6) vs. those treated for accidental COP (17 excess deaths; SMR, 1.3; 95% CI, 1.01-1.6). For the entire cohort, the major causes of death with significantly raised mortality were mental and psychiatric disorders, injuries, and violence. More specific causes of death with significantly raised mortality were alcoholism, motor vehicle accidents with pedestrians, motor vehicle accidents of unspecified type, accidental poisonings, and intentional self-harm. Within cohort comparisons showed that no difference in survival was observed by measure of CO poisoning severity, after controlling for age at poisoning, sex, race, and intent of CO poisoning. Adult survivors of acute CO poisoning treated with hyperbaric oxygen were at increased risk for long-term mortality. Such patients should be followed closely after discharge with consideration given to psychiatric and/or neurocognitive evaluation, as appropriate.