Comatose patients in the non-traumatic emergency room : clinical findings, etiologies and prognosis

Abstract: A reduced level of consciousness is an acute and life threatening condition that requires a rapid and structured management in order to maintain life and brain function. Unconscious patients admitted to the non-traumatic emergency room thus present a major challenge to physicians. The scientific knowledge in the field is limited. The aims of this thesis were to improve the knowledge of underlying etiologies to coma and their short and long term prognosis, and to search for clinical tools to facilitate the diagnostic procedure. The findings reported in this thesis are mainly based on a cohort of prospectively included patients admitted unconscious to hospital during the years 2003-2005. A complementary cohort consists of poisoned patients consecutively admitted to hospital during the years 2009 through 2010. Poisoning was the most common cause of unconsciousness in the non-surgical emergency room (38%) and young age was a strong predictor of this condition (80% of the comatose patients with an age below 40 consisted of poisonings). Around one third of all hospitalized poisonings had a pronounced central nervous system depression on admission. The mortality rate among poisonings presenting unresponsive was found to be at least five times higher than the overall mortality from acute poisoning. The acute prognosis in patients presenting comatose to the emergency room was shown to be serious and dependent on both coma etiology and depth of coma. The overall hospital mortality was 26.5%. Long term prognosis among he hospital survivors was strongly correlated to the coma etiology, with 2-year mortality rates ranging from 11.5% for poisonings to 83% for malignancies, but was not influenced by the initial Glasgow coma scale score. Overall, the prognosis was much more favourable for the coma etiologies poisoning and epilepsy. A composite of age, systolic blood pressure and results of a routine neurological examination could be shown to validly discriminate between the two underlying causes of consciousness disturbances, namely those of metabolic or focal origin. From the data obtained, the following diagnostic algorithm may be formulated: If a patient is younger than 51 years of age, and his or hers systolic blood pressure on admission is below 151 mm Hg, and no neurological findings indicative of a discrete lesion within the central nervous system is present, then the statistical probability of an underlying metabolic coma is 96%. If the algorithm presented above were to be applied routinely in the emergency room, the numbers of emergency CT scans could be considerably reduced. Consequently, other potentially life saving procedures would achieve a higher priority in the emergency room.