Human and animal studies suggest a poorer outcome in the presence of abnormal blood glucose concentration during cerebral hypoxia-ischemia. It is unknown whether this is also the case in acute severe carbon monoxide poisoning. Using Levine-prepared rats, three groups were established and exposed to CO to answer this question: (1) hyperglycemics resulting from the administration of a 50% glucose solution, (2) hypoglycemics resulting from the administration of normal saline, and (3) untreated controls. The rats inhaled 2400 ppm CO for 90 min in the absence of anesthesia. Blood glucose was raised to a mean value of 402 mg/dL just prior to CO exposure in group 1. This resulted in an increased mortality rate (i.e., 54%), and during 4 h of room air recovery an impaired ability to regain body temperature, an increased plasma lactate dehydrogenase activity, and an increased neurologic deficit as compared with group 3. Hypoglycemia, which developed during CO exposure in group 2 (mean minimum glucose after 90 min, 44 mg/dL), resulted in an increased mortality rate (i.e., 46%), and during 4 h of room air recovery an impaired ability to regain body temperature and an increased neurologic deficit as compared with group 3. Blood glucose concentration in the rats in groups 2 and 3 that died during or shortly after CO exposure was significantly depressed relative to the survivors of those groups. Plasma insulin activity was elevated during CO exposure in group 1 as compared with group 3, but fell during recovery; insulin remained low throughout CO exposure and recovery in group 2. The results demonstrate the deleterious effects of both a very high and a very low blood glucose concentration during acute CO exposure.