Summary Key wordsTotal intravenous unaesthesia. Anaesthetics, intravenous; propofol, ketamine.Ketamine is a powerful analgesic, even in doses insufficient to induce anaesthesia.' It has many of the attributes of the 'ideal' analgesic agent for longer routine operations: a very high margin of safety, no irritation of the veins and no negative influence on ventilation or circulation. Its main disadvantages are that it produces hypertension and precipitates psychomimetic emergence phenomena. These effects can be mitigated by judicious medication, particularly by administering benzodiazepines. The combination of midazolam with ketamine has been recommended previously ror total intravenous anaesthesia in military surgery, general civilian practice and cardiac s~r g e r y .~-~ In this study, the combination of propofol/ketamine was compared to the combination propofol/fentanyl in a double-blind, prospective trial in patients undergolng general anaesthesia for elective surgery. Haemodynamic variables, the time to recovery and patient acceptability were compared. MethodsA prospective study of 18 patients who underwent noncardiac surgery was performed. Patients gave informed consent to a protocol approved by the medical ethics committee of our hospital. All patients were ASA grade 1 or 2 and scheduled for operations longer than 15 minutes (Table I). Patients received oral oxazepam (0.25-0.3 mg/kg) as premedication 2 hours before surgery and were allocated randomly to one of two groups to receive propofol with ketamine (n = 9) or propofol with fentanyl (n = 9) for total intravenous anaesthesia.Standard lead I1 of the electrocardiogram was monitored and an intravenous cannula inserted on arrival of the patient in the operating theatre. Heart rate was detected by electrocardiogram and calculated electronically on a beatto-beat basis. The pulse rate was timed for at least 30 seconds. One person recorded all arterial pressure measurements by auscultation (diastolic reading as Korotkoff phase V) using an anaeroid sphygomanometer previously calibrated at zero and 150 mmHg against a mercury column. Anaesthesia was induced with propofol (2 mg/kg) and either fentanyl (3 pg/kg) or ketamine (I mg/kg). Vecuronium (0.15 mg/kg) was administered. Anaesthesia was maintained with propofol 12 mg/kg/hour during the first 30 minutes, followed by 9 mg/kg/hour for 30 minutes and then 6 mg/kg/hour combined with fentanyf 1.5 pg/kg/hour or with ketamine 2 mg/kg hour. The patient's lungs were ventilated with oxygen-enriched air with an no2 of 0.35.The postinduction arterial pressure and heart rate were recorded one minute after induction, and direct laryngoscopy with a curved blade was initiated 2 minutes after induction. None of the patients received topical or intravenous lignocaine before laryngoscopy, and tracheal intubation was always accomplished within 20 seconds.

Correspondence 1085was excessive in 12.5% overall. Three patients exhibited abnormal movements after premedication, one following the smaller dose. There were no significant changes in heart rate, blood pressure or arterial oxygen saturation. Ketamine alone has a foul taste, but it was well accepted when disguised in orange juice, although one vomited 45 minutes later.The pharmacokinetics of oral ketamine and its metabolite norketamine are interesting. Although the bioavailability of oral ketamine is low at 16.5%, the peak plasma norketamine levels are much higher than those after the same dose intramuscularly.' Norketamine probably has anaesthetic and analgesic properties and this possibly accounted for the remarkable lack of response to venepuncture or arterial cannulation.We were attracted by the analgesic and sedative effects of oral ketamine, but the salivation, hypocarbia and dreaming were undesirable side effects in our patients. Eventually, we returned to more conventional premedicant drugs because of disquiet due to the passivity and abnormal movements, but not before combining oral trimeprazine and oral ketamine which rectified the salivation and hyperEdinburgh Royal Infirmary, Ketamine and propofol for TIVAWe would like to comment on the paper by Guit et al.(Anaesthesia 1991: 4 6 24-7). The infusion rates of propofol and ketamine are similar to the rates that we have used in the spontaneously breathing patient for peripheral surgery, such as debridement of wounds, below knee amputation and formation of arteriovenous fistulae. Because of the low infusion rate of ketamine and the interpatient variability to the effects of most drugs, how did they know that they did not have inadequate analgesia or hypnosis in a paralysed patient? Following premedication with oral lorazepam, our patients were given glycopyrroninm to reduce secretions and anaesthesia was induced with ketamine 1 mg/kg and initially propofol 2 mg/kg, which we have now reduced to 1 mg/kg. Ketamine was then infused at 3 mg/kg/hour, and the propofol as per the regimen of Richards et a!. ' An air/oxygen mixture was breathed through a medium concentration oxygen mask.A transient apnoea occurred at induction but on recovery from this the patients were able to maintain their airway. Heart rate increased slightly after the glycopyrronium (a mean of 7 beats/minute), but increased by a mean of 18 beats/minute after induction of anaesthesia. Subsequently the heart rate remained stable.Mean arterial blood pressure fell on average by about 5 mmHg. In two patients nonspecific limb movements occurred, unrelated to any surgical stimulus, and could not be controlled by additional propofol. The time taken to full recovery was very variable and independent of patient factors or duration of surgery. All patients were comfortable in the immediate recovery period and some felt intoxicated. Two patients experienced dreams, but they were not unpleasant. We have further refined this technique using other nonopioid analgesics (nefopam and tenoxicam) and this has i...