All post-neonatal children with acute non-traumatic coma admitted over an 8-month period were analysed and followed up for 18-24 months to determine the aetiology and outcome of their coma. One hundred and sixteen children, 72 boys and 44 girls, were recruited. Half the children were under 1 year of age and only 16 (14%) were more than 6 years of age. Eighty cases (69%) were due to infection, 15 (13%) to toxic metabolic causes, six (5%) to hypoxic ischaemic insults, four (3.5%) had intracranial haemorrhage, nine (7.8%) were due to miscellaneous causes and in two (1.7%) the cause was unknown. Seven cases were lost to follow-up. Of the remainder, 39 (35.7%) died, 32 (29.3%) developed permanent neurological deficit, and 38 (35%) were discharged well. The outcome was worst in the infectious group. Age of onset and sex did not significantly affect outcome. Our findings are similar to experience in Japan, where infection accounts for 74% of non-traumatic coma, but differ considerably from Western data on childhood coma where only a third of cases are due to infection.
Ventriculostomy or external ventricular drain (EVD) placement by free-hand technique has a high malplacement rate. It is a blind procedure that often requires multiple attempts and revisions. To date, no neurological complication due to EVD malplacement has been reported in the literature. In this report, we present the first case of coma induced by a malplaced EVD and the patient regained consciousness after the drain was adjusted. Our discussion focused on various techniques that can improve the accuracy of EVD insertion. EVD insertion under image guidance provides better accuracy with limited disadvantages. We hypothesized that the patient's coma was due to the mass effect and irritation of the malplaced EVD exerted onto the ventral periaqueductal grey matter and the ascending neurons from upper brainstem.