Rectal paraldehyde has good efficacy for terminating tonic-clonic convulsions in children with a low risk of respiratory depression, according to a prospective case audit review.

The authors of this paper note that paraldehyde is accepted as an anticonvulsant for tonic-clonic convulsions where other drugs have failed, however most of the limited published data relate to intramuscular not rectal administration. The aim of this study was to collect data on the efficacy and safety of this route of administration in a paediatric population. Four UK hospitals participated, one a specialist children's hospital and the others district general hospitals. Each provided data on every dose of rectal paraldehyde given for tonic-clonic convulsions in a child: the data collected included clinical circumstances, efficacy, and whether or not respiratory depression occurred. The same preparation was used in each hospital (pre-mixed paraldehyde 50% in olive oil), at the same dose (0.8ml/kg of the mixture; i.e. 0.4ml/kg paraldehyde). Data were collected for each administration.

Evaluable data were available for 53 episodes in 30 patients (mean age 6.12 years, range 5 months to 16 years). Rectal paraldehyde was used first-line in 19 episodes (7 patients), usually because of significant adverse effects after previous use of other agents; in the other 34 episodes it was used after a benzodiazepine (31) or phenytoin (3) had failed. Rectal paraldehyde terminated the convulsion in just under two-thirds of episodes (33, 62.3%); in those where there was a pre-existing diagnosis of epilepsy, it was effective in three-quarters (74.3%). In two episodes there was no comment on respiratory function and in the remainder no significant respiratory depression occurred.

The authors conclude that rectal paraldehyde is effective in treating acute prolonged tonic–clonic convulsions in children. No episodes of respiratory depression were reported in their case series. They therefore recommend that it should remain an option for management in this situation.

[Editor's note: based on the '3/n rule' (if none of n patients showed the event about which we are concerned, we can be 95% confident that the chance of this event is at most 3 in n, i.e. 3/n.  BMJ 1995; 311: 619-20), no episodes of respiratory depression in 53 instances suggests a maximum rate of no more than once in about 18 administrations].