Twenty patients presenting for submucous resection of the nasal septum under general anaesthesia were randomly allocated to four groups to receive either 1.0 ml 25% cocaine HCI in paraffin paste, 1.0 m125% cocaine HCI combined with 0.1% adrenaline in paraffin paste, 4.0 ml aqueous 4% cocaine HCI combined with 0.050/0 adrenaline or 4.0 ml aqueous 4% cocaine HCI on ribbon gauze applied to the nasal mucosa. Mean intraoperative blood loss was significantly decreased when the 25% cocaine 0.1% adrenaline combination in paraffin paste was used (11 (SD 8) ml, 60 (SD 30) ml, P<0.05, for adrenaline and plain paste respectively). Combination of adrenaline with cocaine in the aqueous formulation was not associated with a significant decrease in blood loss compared with aqueous cocaine alone (75 (SD 51), 96 (SD 66) ml respectively). Cocaine adrenaline paste and plain cocaine paste were associated with higher mean maximum cocaine blood concentrations (1.6 (SD 1.4), 2.0 (SD 1.5) JAg/ml respectively) when compared with aqueous cocaine adrenaline and aqueous cocaine alone (0.03 (SD 0.003), 0.5 (SD 0.3) JAg/ml respectively). Heart rate and blood pressure changes were similar in all four groups and cardiovascular toxicity was not observed. One ml of topical intranasal 25 % cocaine HCI with 0.1 % adrenaline in paraffin paste provided the best haemostasis for nasal septal surgery.

In a prospective double-blind study, single-dose lumbar epidural blockade was administered to 60 healthy patients undergoing lower abdominal surgery, the patients lying in the lateral position only during the time of injection of the local anaesthetic. Solutions used were bupivacaine HCI 0.5%, lignocaine HCI 2% and Iignocaine-bupivacaine mixtures in the ratios of 1:3, 1:1 and 3:1 by volume. Data were pooled and analysed for the effects of posture on epidural blockade. Using skin temperature as a criterion of sympathetic blockade, onset of blockade was more rapid and there was more prolonged blockade on the dependent side. Initial onset of sensory blockade was faster on the dependent side by 1 minute and 3.1 minutes for partial and complete blockade, respectively. Mean duration of sensory blockade was longer in the dependent dermatomes for partial (14, SD 7. minutes, T6-L4) and complete blockade (20, SD 6, minutes, TS-L3) respectively. Initial onset of complete motor blockade was 5 minutes more rapid on the dependent side, with mean myotome score consistently greater at all time intervals on the dependent side. Our study therefore suggests that a more efficacious sensory and motor blockade could be achieved by lying the patient on the operative side during the administration of the epidural and injection of the local anaesthestic solution close to the operative dermatomes.