Codeine is frequently used for postoperative analgesia in children. Intramu
scular injections are not ideal and the rectal route may be preferable. We
compared rectal and intramuscular codeine administered following neurosurge
ry. 20 children (over 3 months) undergoing elective neurosurgical procedure
s, were randomized to receive either rectal or intramuscular codeine phospa
te (1 mg.kg(-1)) at the end of the procedure. Serum levels of codeine and m
orphine were assayed at intervals following administration (0, 30, 60, 120,
240 min). Fentanyl was the intraoperative analgesic and postoperative resc
ue analgesia was paracetamol, diclofenac and intramuscular codeine. The Chi
ldren's Hospital of Eastern Ontario Pain Scale was used to assess analgesia
. Peak codeine levels in both groups were observed at 30 min and morphine l
evels were consistently low. The plasma codeine levels were significantly g
reater at 30 and 60 min following intramuscular injection, and were associa
ted with slightly better analgesia scores, but did not reach statistical si
gnificance. However, the peak plasma level occurred at similar times in bot
h groups. Codeine is absorbed as rapidly via the rectal route compared with
the intramuscular route but the peak levels are lower.