Summary

Background and objectives: A recent survey in the British Medical Journal reported the attitudes of orthopaedic surgeons towards the intraoperative death of a patient. Several replies to this article were from anaesthetists, who pointed out that other staff might be affected by ‘death on the table’. We designed a questionnaire survey to assess the attitudes of anaesthetists, concerning intraoperative death. Methods: Three hundred anonymized questionnaires were distributed to 12 anaesthetic departments throughout England. Results: Two hundred and fifty-one replies were received (84% response rate); 92% of respondents had experienced an intraoperative death, the majority of deaths being expected (60%) and non-preventable (77%), occurring mainly during emergency surgery (80%), particularly involving vascular surgery (41% of cases); 87% had administered another general anaesthetic in the following 24 h, most without their professional ability being compromised (77%). Conclusions: This survey shows that anaesthetists are highly likely to experience intraoperative death, the consequences of which can be extremely stressful. Although the majority of anaesthetists (71%) agreed that it was reasonable for medical staff not to take part in operations for 24 h after an intraoperative death, fewer (25%) thought the proposal practicable. Nevertheless, all departments should provide for the discontinuation of further operations, if the circumstances require it. Consideration should be given by all departments of anaesthesia towards the prevention of intraoperative death, and the management of its aftermath, including the provision of support for psychologically traumatized staff.

Summary

Background and objective: The aim of this study was to investigate any possible protective effect of ketamine in acute muscular ischaemia and reperfusion injury by measuring malondialdehyde using thiobarbituric acid assay in rats. Methods: Twelve female Wistar albino rats were anaesthetized with chloral hydrate and randomly assigned into two groups to receive ketamine 1 mg kg−1 min−1 or saline infusion. Blood and gastrocnemius muscle samples were obtained 10 min after onset of infusion, before ischaemia. Then, femoral arteries were clamped for 30 min. Blood and muscle samples were obtained at the 30th minute of ischaemia and 10 min after reperfusion. Results: Muscle malondialdehyde concentrations were 27.88 ± 2.45, 27.62 ± 3.98 before ischaemia, 32.10 ± 4.19, 30.77 ± 2.73 in the 30th minute of ischaemia and 44.34 ± 2.45, 34.83 ± 2.78 after reperfusion in saline and ketamine-treated rats, respectively (nmol g−1, mean ± SD). The muscle malondialdehyde level after reperfusion was lower in ketamine-treated rats compared to saline group (P < 0.002). Plasma malondialdehyde levels were 3.77 ± 0.16, 3.78 ± 0.18 before ischaemia, 3.81 ± 0.25, 4.00 ± 0.86 at the 30th minute of ischaemia and 4.00 ± 0.53, 3.94 ± 0.95 after reperfusion, respectively, in saline and ketamine-treated rats (μmol L−1, mean ± SD). The effect of ketamine on muscular malondialdehyde was not observed in concurrent plasma malondialdehyde levels. Conclusion: Ketamine was found to attenuate acute ischaemia–reperfusion injury in muscle tissue in rats (muscular protective). Ketamine may attenuate lipid peroxidation in muscle tissue in tourniquet-requiring manoeuvres.

Summary

Background and objective: This study was conducted to determine the effect of ketamine on metabolic homoeostasis and particularly in lipoprotein lipase (LPL) activity in adipose tissue.

Methods: Sixty male Wistar rats were divided into six groups of 10 each. Group A served as controls, while Groups B–F received, respectively, ketamine 60, 80, 100, 120 and 140 mg kg−1 intraperitoneally. The animals were sacrificed 20 min after the administration of ketamine. Insulin concentrations in plasma and total cholesterol, triglyceride, high-density lipoprotein (HDL) and free fatty acid (FFA) concentrations in serum were measured. LPL activity in adipose tissue and medium-chain acyl-CoA dehydrogenase (MCAD) content in muscle were determined.

Results: FFA concentrations in serum significantly increased from the second lowest dose of ketamine. Insulin concentrations in plasma did not exhibit any significant difference between groups. MCAD levels were 0.5-fold more in Group F than in Group A, while there were no significant differences between control group and Groups B–E. Furthermore, high concentrations (120 and 140 mg kg−1) of ketamine interfered with in metabolic homoeostasis by significantly reducing LPL activity, thus elevating triglyceride concentrations in serum without affecting cholesterol and HDL metabolism.

Conclusions: Ketamine induces various metabolic effects due to changes in adipose LPL activity and MCAD levels in muscles. These findings seem to be significant only at high doses.

Summary

Background and objective: An anti-inflammatory effect of α2-adrenoreceptor agonists has been suggested. Phospholipase A2 is a key enzyme in the production of precursors of inflammatory lipid mediators. The aim of the present study was to investigate the effect of clonidine on phospholipase A2 activity in an established in vitro model. Methods: Human being platelet membranes containing active phospholipase A2 were exposed to buffer control or to three increasing concentrations of clonidine. Phospholipase A2 was measured by a radioisotope technique. Results: A massive increase in phospholipase A2 activity was measured after clonidine exposure leading to final values of 92.5 ± 3.1 pmol mg protein−1 min−1 (4.5-fold higher than control values; P ≤ 0.01 vs. control). After clonidine exposure the maximal reaction velocity increased, while the Michaelis–Menten constant did not change. The Lineweaver–Burk representation suggested an interaction of clonidine with the phospholipase A2–substrate complex as well as the phospholipase A2 molecule. Conclusion: We conclude that the putative anti-inflammatory effect of clonidine was not caused by inhibition of phospholipase A2.

Summary

Background and objective: An alteration in production of both interleukin-10 (IL-10) and nitric oxide (NO) has been found following surgical/anaesthesia trauma. It is also suggested that IL-10 could be an important factor in regulating NO metabolism during the postoperative period. Furthermore, NO seems to play a crucial role in the anaesthetic state. The purpose of this study was to investigate plasma levels of IL-10 and NO following surgery, any possible correlation between these two variables and whether anaesthesia technique could influence NO and IL-10 circulating concentrations.

Methods: Thirty-two patients scheduled to undergo elective major surgery were enrolled in the study and allocated into two groups to receive two different techniques of anaesthesia, total intravenous (i.v.) anaesthesia (Group I) and inhalational anaesthesia (Group II). Blood samples were drawn before (t0), at the end (t1) of operation and after 24 h (t2). Plasma IL-10 and NO levels were measured by using an enzyme-linked-immunosorbent assay (ELISA) and a total NO assay kit, respectively.

Results: In both patient groups there was a significant decrease of plasma NO levels at the end of surgery (30.35 ± 2.70 mmol L−1 at t0 to 13.76 ± 1.51 mmol L−1 at t1 in Group I, P < 0.0001; 28.23 ± 2.50 mmol L−1 at t0 to 11.38 ± 0.95 mmol L−1 at t1 in Group II, P < 0.0001). This reduction remained at 24 h postoperatively (14.33 ± 1.52 mmol L−1 in Group I, P < 0.0001; 12.52 ± 1.11 mmol L−1 in Group II, P < 0.0001, both vs. t0). There was an increase in IL-10 concentrations (26.35 ± 3.42 pg mL−1 and 75.39 ± 8.33 pg mL−1 at t1 and t2, respectively, vs. 4.93 ± 0.31 pg mL−1 at t0, P = 0.03 and P < 0.0001, respectively, in Group I; 26.18 ± 3.22 pg mL−1 and 69.91 ± 7.33 pg mL−1 at t1 and t2, respectively, vs. 5.50 ± 0.33 pg mL−1 at t0, P = 0.02 and P < 0.0001, respectively, in Group II). No relationship was found between circulating IL-10 and NO.

Conclusions: During the postoperative period, IL-10 overproduction does not correlate with the decrease in systemic NO concentration.

Summary

Background and objective: Volatile anaesthetics inhibit nicotinic acetylcholine receptors at clinically relevant concentrations with higher affinity for the neuronal nicotinic receptor. The inhibitory effects of propofol on nicotinic receptors have only been documented at supraclinical concentrations. The aim of this study was to determine recovery properties and train-of-four (TOF) fade of mivacurium during sevoflurane and propofol anaesthesia, in order to examine any differences both in the enhancement of the neuromuscular block (postjunctional effects) and in TOF fade (prejunctional effects).

Methods: Twenty ASA I–II adult patients were randomly allocated to maintenance of anaesthesia with sevoflurane (end-tidal concentration 2%) or propofol. Neuromuscular block was assessed by acceleromyography and a single dose of mivacurium (0.15 mg kg−1) was administered (in the sevoflurane group after 30 min of exposure to sevoflurane). We measured time for recovery of the first twitch of the TOF (T1) from 25–75%, time from 25% recovery of T1 to achieving a TOF ratio (TOFR) of 0.8, TOFR at 50%, 75% and 90% recovery of T1, and height of T1 at TOFR of 0.7 and 0.9. Data were tested using t-test for independent samples.

Results: Recovery times (mean (95% confidence interval, CI)) of mivacurium in the sevoflurane group (T1 25–75%, 11.3 (8.1–14.5) min; T1 25%-TOFR0.8, 19.1 (15.7–22.5) min) were significantly longer (P < 0.05) than in the propofol group (T1 25–75%, 6.5 (5.2–7.7) min; T1 25%-TOFR0.8, 11.3 (7.8–10.3) min). No differences were found in the relations between TOFR and T1 or vice versa, between the groups.

Conclusions: Recovery times after a single dose of mivacurium were prolonged by sevoflurane compared with propofol but no differences in TOF fade were observed between the two anaesthetics.

Summary

Background and objective: Hypotension, the commonest side-effect of spinal anaesthesia, results from sympathetic denervation. This study compared patient positioning (supine vs. decubitus) on haemodynamic variables during spinal anaesthesia. Methods: After intravenous crystalloid preloading with 5 mL kg−1, hyperbaric bupivacaine 0.5% 2.5 mL was injected intrathecally at the L2–3 or L3–4 interspace. Patients were then randomly assigned to be positioned immediately supine and horizontal for 30 min (Group SUP, n = 12), or remained in the lateral decubitus position (fractured hip dependent) for 30 min (Group LAT, n = 14). Systolic blood pressure, mean arterial pressure, and loss of sensation of pinprick sensation were recorded prior to induction of spinal anaesthesia (baseline) and at 1, 2, 3, 5, 10, 15, 30, 45, 60, 90 and 120 min after intrathecal injection. Results: In Group SUP, the percent maximum systolic blood pressure (36 ± 13%) and percent maximum mean arterial pressure decreases (27 ± 13%) were significantly greater (P < 0.05) than in Group LAT (30 ± 8% and 23 ± 11%, respectively). Additionally, there was a borderline significant delay in the time to maximum systolic blood pressure decrease in Group LAT (38 ± 30 min) when compared with Group SUP (20 ± 17 min, P = 0.06), while the total dose of ephedrine required in the SUP group (30 mg) was greater than that required in the LAT group (15 mg, P = 0.05). In Group LAT patients, the mean level of denervation on the operative side extended 2 dermatomes more cephalad than in Group SUP. Conclusions: Lateral positioning for spinal anaesthesia delays the onset of hypotension, while requiring smaller total doses of vasoconstrictors for blood pressure maintenance.

Summary

Background and objective: The degradation of the cytoskeletal protein microtubule-associated protein 2 (MAP2), by calpain has been known to occur following traumatic brain injury. We examined the therapeutic potential of calpain inhibitor 2, compared with that of moderate hypothermia in traumatic brain injury produced by weight drop in rats.

Methods: An inhibitor treated group (n = 8) received calpain inhibitor 2 intravenously (i.v.) for 5 min before and for 6 h after injury (total 2 μmol); a hypothermic (HT) group (n = 8) was maintained at 30°C (temporalis muscle temperature) for 45 min prior to and 60 min after injury; an untreated (UT) group (n = 8) received an infusion of inactive vehicle. Eight rats (sham group) underwent surgery without brain injury. Histopathological (haematoxylin and eosin staining) and MAP2 (immunohistchemistry and western blotting) evaluations were performed at 6 h after injury.

Results: Ipsilateral cortical damage was marked in the injured groups. In the hippocampus, marked pyramidal neuronal damage was observed in the UT and calpain inhibitor treated (CI) groups, while these neurons were better preserved in the HT group. The hippocampal MAP2 levels in the UT, CI and HT groups were significantly decreased to 13 ± 9%, 28 ± 33% and 62 ± 25% of the sham control, respectively. MAP2 concentration in the HT group was significantly higher than in UT and CI groups (P < 0.05).

Conclusion: The results suggest that moderate hypothermia, but not calpain inhibitor 2 with the tested regime, attenuates cytoskeletal damage in the ipsilateral hippocampus at 6 h after traumatic brain injury.

Summary

Background and objective: The aim of this study was to assess postoperative patients' self-evaluation after xenon anaesthesia compared to total intravenous anaesthesia with propofol. Methods: 160 patients aged 18–60 yr, ASA I–II undergoing elective surgery took part in this randomised-controlled trial. After approval by the local Ethics Committee and as soon as the patients had given their written informed consent, they were randomly allocated to either the xenon (n = 80) or propofol (n = 80) group. In both groups remifentanil was used as opioid. The postoperative patients' self-evaluation was assessed with a double-blind telephone poll. Early spatial orientation, patients' self-evaluation of anaesthesia, choice of the same anaesthesia for future operations and recall of uncomfortable feelings after anaesthesia were determined. Results: 116 Patients were analysed, 63 in the xenon and 53 in the propofol group. The two study groups were comparable with respect to age, weight, height, gender and ASA classification. The two groups indicated similar values in the early spatial orientation at the onset of recovery and thereafter. Patients' self-evaluation of anaesthesia with main emphasis at high marks and repetition of the same anaesthesia if necessary were similar in both groups. Recalls of uncomfortable feelings were comparable but not for postoperative pain and appetite/thirst which appeared with a significantly higher incidence in the xenon group. Conclusions: Patients' self-evaluation and memory of early spatial orientation following xenon anaesthesia are comparable to propofol.

Summary

Background and objective: The aim of this study was to evaluate the value of terminal complement complex (C5b-9) plasma levels as a marker for complement activation in septic shock with concomitant capillary leak syndrome. Methods: In a prospective animal study 10 fasted, anaesthetized, mechanically ventilated and multi-catheterized pigs (20.6 ± 1.3 kg) were investigated over a period of 8 h. Sepsis was induced by faecal peritonitis (1 g kg−1 body weight faeces, n = 5) and compared to controls (n = 5). The animals received 6% hydroxyethyl starch 200/0.5 to maintain a central venous pressure of 12 mmHg. To quantify capillary leak syndrome, albumin escape rate was measured using 99mTc-labelled human serum albumin. Plasma levels of terminal complement complex were measured in a double antibody immunoassay (neoepitope-specific MoAb aE11 as catching antibody). Immunohistological studies of renal specimens were performed to detect terminal complement complex deposition. Results: Albumen escape rate increased in septic animals (+52%) compared to controls (+3%, P < 0.05). Plasma levels of terminal complement complex decreased during the study period in both groups. In septic animals this finding was accompanied by a significant deposition of terminal complement complex in renal specimens (P < 0.05). Conclusion: We found an activation of the complement system proven by marked deposition of terminal complement complex in renal specimen, while its plasma levels decreased during the study period in septic and control animals. These results suggest that in septic shock with capillary leak syndrome plasma level of terminal complement complex may not be a reliable marker of complement activation.

Summary

Background and objective: Anaesthetic drugs and procedures interfere with secondary brain injury following severe head trauma, yet studies regarding the anaesthetic management of these patients are lacking. We investigated the behaviour of Italian anaesthetists regarding this topic.

Methods: A questionnaire investigating the approach to a patient with severe head trauma requiring an urgent splenectomy for ruptured spleen was sent to 250 Italian anaesthetists. Questions regarded the pre-, intra- and postoperative phases, and concerned the rationale use and availability of specific monitoring systems, and indications for invasive procedures and use of drugs, fluids and blood products.

Results: There were 162 (64.8%) responders. Seventy-five percent believed that early tracheal intubation within the emergency room was necessary, while 25% postponed it to the operating room. Basic monitoring was defined as essential by all responders, 147 (90.7%) considered invasive arterial pressure monitoring to be essential. Fifty-seven (84%) anaesthetists working in hospitals without neurosurgical facilities would have transferred the patient after splenectomy. Prophylactic hyperventilation was frequently used (36%). Sixty-eight percent of responders would have preferred in intracranial pressure monitoring inserted before laparotomy, but only 35% actually had this possibility. In case of acute intraoperative arterial hypotension after splenectomy, 54% of the responders advocated the use of blood or blood products to optimize peripheral oxygen transport.

Conclusions: More widespread knowledge of certain areas of severe head trauma management such as early tracheal intubation, avoidance of prophylactic hyperventilation, adequate invasive monitoring, appropriate use of blood products, and timing of transfer to hospitals with neurosurgical facilities is needed.

Summary

Background and objective: The combination of non-opioid analgesic drugs, though widely used, has been rarely evaluated. The aim of this study was to compare the efficacy of propacetamol and the non-steroidal analgesic drug ketoprofen, alone or in combination, on pain relief after thyroid surgery performed using remifentanil.

Methods: Ninety-seven patients were randomly allocated to one of the three groups: propacetamol 2 g (32), ketoprofen 100 mg (33) and propacetamol 2 g + ketoprofen 100 mg (32). Each regimen was administered intravenously (i.v.) 30 min before the end of surgery and then every 6 h. If pain was not relieved, patients received an i.v. bolus of tramadol 100 mg. Tramadol consumption and pain intensity using a visual analogue scale was recorded at 1, 2, 8 and 14 h after the end of surgery.

Results: Pain scores were significantly higher with propacetamol compared with ketoprofen 2 h after surgery (35 ± 3.7, 21 ± 2.6, respectively; P < 0.01). The number of patients receiving tramadol was higher with propacetamol alone compared with the two other groups, 1 h (14/32, 4/33, 2/32, respectively; P > 0.01) and 2 h (24/32, 6/33, 8/32, respectively; P < 0.01) after surgery. There was no difference between ketoprofen alone and ketoprofen plus propacetamol, and there was no difference between the three groups from the 8th hour onward.

Conclusions: In the immediate postoperative period after thyroid surgery performed using remifentanil, the concomitant use of propacetamol and ketoprofen does not improve analgesia compared with ketoprofen alone.

Summary

Background and objective: Isoflurane has been shown to mimic ischaemic preconditioning (IPC). The protective effect of IPC, or applying isoflurane or perfusion with the ‘push-pull’ complex zinc–desferrioxamine (Zn–DFO) in the canine heart, was investigated.

Methods: Thirty minutes after salicylate administration (100 mg kg−1) the heart was exposed. All dogs were subjected to a 10 min left anterior descending artery occlusion followed by 2 h of reperfusion. In Group I (n = 9) isoflurane (2.5%) was administered 10 min prior to and during ischaemia. In Group II (n = 8), IPC was elicited by 5 min coronary artery occlusion, followed by 5 min of reperfusion, prior to the 10 min ischaemia. In Group III (n = 9) Zn–DFO (2.5 mg kg−1) was given 10 min prior to ischaemia. The effects of these interventions were compared to control (n = 10). Coronary sinus blood concentrations of salicylate, 2,3-dihydroxybenzoic acid (DHBA), lactate, pH and oxygen content were monitored.

Results: In the control group, 2,3-DHBA increased by 32% above the pre-ischaemic value (P < 0.05). In contrast, in the IPC hearts, a significant decrease in the production of 2,3-DHBA was observed (40% lower than baseline, P < 0.01). In the isoflurane group only a 13% (and non-significant) decrease was noticed. In the Zn–DFO group a 33% decrease was found (P < 0.01). The increase in lactate concentrations in the IPC and Zn–DFO groups was significantly smaller than that of control and isoflurane groups.

Conclusions: IPC protected the heart against the deleterious effects of reperfusion, possibly by amelioration of the level of oxygen-derived reactive species, and the complete inhibition of reactive hydroxyl radical production. Isoflurane did not prove to be as effective in reducing the free radical damage.

Summary

Background and objective: Sore throat and hoarseness rank, besides pain and nausea, among the most frequent subjective complaints after tracheal intubation for general anaesthesia. Our intention was to determine the incidence of postoperative sore throat from a large sample of patients and thus to identify the most important associated factors.

Methods: We prospectively followed up 809 adult patients who underwent elective surgical interventions and examined their history, the applied anaesthetic techniques, perioperative course and the occurrence, intensity and duration of postoperative throat complaints. The assignment and professional experience of the involved intubators were also assessed. The influence of a multitude of variables on postoperative throat complaints was statistically analysed.

Results: Postoperative sore throat was present in 40% overall being significantly higher in female than in male (44% vs. 33%; P = 0.001). The mean pain intensity in the affected patients (n = 323) was 28 ± 12 mm on a visual analogue scale where 0 = no pain and 100 = extreme pain. The average duration was 16 ± 11 h. Main factors associated with throat complaints were female sex, history of smoking or lung disease, duration of anaesthesia, postoperative nausea, bloodstain on the endotracheal tube and natural teeth. We could find no influence on the occurrence or intensity of throat complaints by the professional assignment or the length of professional experience of the personnel involved.

Conclusions: Postoperative throat complaints frequently arise after tracheal intubation for general anaesthesia in the first 2 postoperative days, but they are of limited intensity and duration.

Summary

Background and objective: The tunnelling phase of ventriculoperitoneal shunt insertion is the most painful part but patients are often given inadequate opioid analgesic for fear of post operative delayed recovery and/or respiratory depression. This may result in an increase in intracranial pressure. Methods: Twenty adults scheduled to undergo ventriculoperitoneal shunt insertion were administered standard anaesthesia. Monitoring included heart rate, electrocardiogram, end-tidal carbon dioxide, invasive blood pressure, and oxygen saturation. Intracranial pressure was monitored by placing the ventricular end of shunt catheter in the dilated lateral ventricle. Five minutes before tunnelling, fentanyl 1 μg kg−1 was administered. Mean arterial pressure, heart rate and intracranial pressure were recorded during tunnelling and subsequently at 1-min interval for 5 min. Data were analysed using t-test and repeated measured test. Results: Tunnelling caused significant increase in mean arterial pressure (from 81.4 ± 11.0 to 110.9 ± 15.3 mmHg, P < 0.05), intracranial pressure (from 21.4 ± 8.1 to 29.2 ± 12.5 mmHg, P < 0.05) and heart rate (from 74.4 ± 13.8 to 94.1 ± 17.8 beats min−1, P < 0.05). Whereas, the increase in haemodynamic parameters persisted for 3 min post-tunnelling, elevated intracranial pressure lasted for 2 min. Conclusion: Tunnelling significantly increases intracranial pressure and blood pressure despite prior fentanyl administration. This may be deleterious in the presence of intracranial pathology.

Summary

Background and objective: Magnesium disorders are common in hospitalized patients. In patients with low or normal magnesium, the intravenous magnesium loading test has been demonstrated to be a sensitive test to assess magnesium deficiency in critically ill patients. However, it is more time consuming and more difficult than the measurement of intracellular or extracellular magnesium concentrations. This study evaluated whether erythrocyte, plasma and urinary magnesium concentrations predict renal magnesium retention measured by the magnesium loading test.

Methods: One-hundred-and-three intensive care patients (36 females, 67 males) in a tertiary care centre and 41 healthy subjects (13 females, 28 males) took part in this prospective study. Intracellular, total plasma, ionized extracellular and urinary magnesium concentrations were measured and also magnesium retention by intravenous magnesium loading test.

Results: Total plasma magnesium concentration was poorly correlated with magnesium retention (r = 0.36, r2 = 0.13) and was the only parameter that significantly predicted magnesium retention in intensive care patients (P < 0.01). However, only 10% of the magnesium retention data were linked to the total plasma magnesium concentration.

Conclusions: Total plasma magnesium concentration predicts magnesium retention in critically ill intensive care patients but not intracellular and urinary magnesium concentrations. Only a small proportion of the magnesium retention was due to the total plasma magnesium concentration.

Summary

Background and objective: This randomized, double-blinded clinical study was designed to compare the efficacy and safety of remifentanil and morphine administered using intravenous (i.v.) patient-controlled analgesia (PCA) for postoperative analgesia after major abdominal surgery during the first 24 postoperative hours.

Methods: Sixty-nine patients were randomly allocated into two groups, each receiving remifentanil or morphine. The first group received i.v. remifentanil PCA with a loading dose of 45 μg, a maintenance dose of 1 μg min−1, a bolus dose of 15 μg and a lockout interval of 10 min during the first 24 postoperative hours. The second group received i.v. morphine PCA with a loading dose of 5 mg, a maintenance dose of 0.3 mg h−1, a bolus dose of 1 mg and a lockout interval of 15 min. Age, weight, sex, history of general anaesthesia, duration of surgery and time spent in the post-anaesthesia care unit were recorded. Preoperative pulse rate, systolic and diastolic blood pressures (BP), respiration rate and arterial blood gases were collected. Pulmonary function was tested before induction of anaesthesia, as well as at 4 and 26 h after operation. Pulse rate, systolic and diastolic BP, respiration rate, arterial blood gases, sedation and visual analogue scores, and presence of side-effects in the recovery room and on the ward for 24 h were recorded at 0, 1, 2, 4, 6, 12, 18 and 24 h after operation. Total drug use, number of boluses delivered, number of boluses demanded and delivery/demand ratio were collected.

Results: Sixty patients were evaluated. The groups did not differ in age, weight, sex, history of general anaesthesia, duration of surgery or time spent in the recovery unit. There were also no clinically relevant differences between the groups with regard to haemodynamic and respiratory parameters as well as sedation and visual analogue scores (P > 0.05). More bolus doses were demanded and delivered and the delivery/demand ratio was significantly higher in the remifentanil group (P < 0.05). There was no finding suggesting acute opioid tolerance during remifentanil or morphine PCA.

Conclusion: The i.v. remifentanil PCA with the chosen dosage regimen after abdominal surgery produces postoperative analgesia and has cardiovascular side-effects similar to those achieved with i.v. morphine. Special attention must be given to respiratory depression during establishment of PCA with remifentanil.