Background and objective: Myocardial ischaemic preconditioning is characterized by a reduction in the rate of glycolysis. Brief myocardial ischaemia also reduces the glycogen content of the heart. The first objective was to determine whether augmenting glucose oxidation by activation of the pyruvate dehydrogenase complex would prevent the infarct limitation of ischaemic preconditioning. The second part of the study evaluates whether glycogen depletion before ischaemia mimics the infarct-limiting effect of ischaemic preconditioning.

Methods: Dichloroacetate (300 + 150 mg kg−1), an activator of the pyruvate dehydrogenase complex, was administered intravenously in the anaesthetized open-chest rabbit. All animals underwent 45 min of regional ischaemia and 3 h of reperfusion. Ischaemic preconditiong was elicited by 5 min of coronary occlusion. Control rabbits, those with ischaemic preconditioning with no dichloroacetate, received a saline vehicle. An isolated perfused rabbit heart model was employed to test the second hypothesis. Hearts were depleted of glycogen by perfusing them with a substrate-free buffer. Infarction was assessed by triphenyl tetrazolium chloride and area at risk determined with fluorescent particles.

Results: (a) Pyruvate dehydrogenase complex activation experiments. Treatment with dichloroacetate alone did not alter infarct size (58 ± 7% control vs. 60 ± 5% dichloroacetate). Addition of dichloroacetate did not attenuate the infarct-limiting effect of ischaemic preconditioning as evidenced by a similar reduction in infarct size in the ischaemic preconditioning group (22 ± 5%) and in the ischaemic preconditioning + dichloroacetate group (27 ± 7%). (b) Glycogen depletion experiments. Compared with control hearts with a normal glycogen content (4.84 ± 0.15 mg g−1 wet weight), glycogen depleted and ischaemic preconditioning hearts had reduced glycogen content before ischaemia (2.15 ± 0.26, 1.62 ± 0.17 mg g−1 wet weight, respectively; P < 0.01). Glycogen depletion did not reduce infarct size: 25.0 ± 4.5% cf. 27.9 ± 3.4% in the control group. However, ischaemic preconditioning resulted in a significant reduction of infarct size (11.5 ± 2.3% vs. 27.9 ± 3.4% control; P < 0.01).

Conclusions: Augmentation of oxidative glycolysis by dichloroacetate in in situ rabbit hearts does not alter the effect of ischaemic preconditioning, and glycogen depletion in the isolated rabbit heart does not influence infarct size after subsequent coronary occlusion.

Background and objective: This prospective, randomized, double-blinded study was designed to compare the effects of remifentanil or fentanyl on anaesthetic induction characteristics of propofol, thiopental or etomidate.

Methods: Seventy-two patients were enrolled in six groups of 12 individuals each. In three groups, fentanyl was given as a bolus dose of 1.5 μg kg−1, whereas the others received a remifentanil infusion at 0.5 μg kg−1 min−1. Five minutes later, propofol, thiopental or etomidate were titrated to a state of unresponsiveness. Assessment included the amounts of drug necessary for induction, haemodynamics and the times to apnoea, loss of eyelash reflex, and the release of a water-filled syringe held in the patient's hand.

Results: Induction times to loss of the eyelash reflex were significantly shorter in the remifentanil than in the fentanyl groups: with propofol 50.7 ± 13.6 s (mean ± SD) versus 74.9 ± 27.0 s (P < 0.01), with thiopental 42.9 ± 16.8 s versus 77.2 ± 27.8 s (P < 0.01) and with etomidate 54.7 ± 17.6 s versus 72.3 ± 24.0 s (P < 0.05). The times to respiratory arrest or for the syringe to fall were significantly shorter with remifentanil than with fentanyl for propofol and for thiopental, but not for etomidate. In terms of dosages per kg body weight necessary to achieve unresponsiveness, less propofol (−29%, P < 0.05), thiopental (−25%, P < 0.05) or etomidate (−32%, P < 0.01) was necessary with remifentanil than with fentanyl. Haemodynamic responses to tracheal intubation were controlled more effectively with remifentanil. However, within the remifentanil groups, mean arterial pressure significantly decreased during induction: −26% with propofol,−18% with thiopental and −14% with etomidate (all P < 0.01).

Conclusions: During anaesthetic induction, a remifentanil infusion of 0.5 μg kg−1 min−1 over 5 min is a suitable alternative to a 1.5 μg kg−1 bolus dose of fentanyl: induction times are shorter with reduced amounts of propofol, thiopental or etomidate.

Background and objective: Target-controlled infusions of anaesthetic agents have become increasingly available. They can involve the use of propofol in combination with an opioid or a benzodiazepine. The effect site concentration of propofol infusions has been advocated as a method of estimating drug distribution. We investigated the influence of co-induction with remifentanil and midazolam on effect site propofol requirements at induction of anaesthesia using target-controlled infusions.

Methods: Sixty-six consenting adult patients were randomly allocated to three treatment groups. Each group received induction of anaesthesia with a different total intravenous technique. One group was induced with target-controlled propofol alone; another received target-controlled propofol and target-controlled remifentanil (3 ng mL−1); and the last received midazolam (0.03 mg kg−1), target-controlled remifentanil (3 ng mL−1) and target-controlled propofol. Computer simulation was used to calculate effect site concentrations. We recorded propofol dose and effect site concentration at loss of verbal response.

Results: The effect site concentration (Ce50) of propofol alone was 2.19 µg mL−1. This was reduced to 1.55 µg mL−1 during co-induction with remifentanil and further reduced to 0.64 µg mL−1 with midazolam premedication (P < 0.001; ANOVA).

Conclusions: We conclude that co-induction with remifentanil alone or with midazolam can be used to reduce propofol doses at induction of anaesthesia using target-controlled infusions. We believe that using effect site concentration may prove a useful tool in routine clinical practice.

Background and objective: Children are vulnerable to regurgitation with a relatively high incidence of aspiration during general anaesthesia which is attributed to the high intragastric pressure, a short oesophagus, an immature laryngeal reflex and incomplete lower oesophageal sphincter muscle function. Subcutaneous administration of epinephrine is generally used in surgery to decrease bleeding due to local vasoconstriction. The effect of epinephrine on the sphincter muscle tone was investigated during general anaesthesia in children.

Methods: Ten children scheduled for skin graft or plastic surgery of the ear were studied. A gastrointestinal pressure sensor was inserted nasally, and the intraluminal pressures of the lower oesophagus, lower oesophageal sphincter and stomach were monitored under sevoflurane, nitrous oxide anaesthesia. The effect of epinephrine on the lower oesophageal sphincter muscle tone was measured.

Results: The resting pressure of the lower oesophageal sphincter muscle tone significantly decreased from 4.56 ± 1.85 to 3.79 ± 1.11 kPa after 3 μg kg−1 epinephrine for 4 min. The barrier pressure is the difference between the lower oesophageal sphincter and intragastric pressure, and that decreased to 1.23 ± 1.17 kPa from the 2.07 ± 1.77 kPa resting level. The observations implied that epinephrine had a long-lasting relaxing effect on lower oesophageal sphincter muscle in children.

Conclusions: The observations may give some explanation about the mechanism of gastro-oesophageal reflux during general anaesthesia, especially in the participation of the adrenergic receptors.

Background and objective: To evaluate preparation and discharge times as well as the anaesthesia-related costs of out-patient knee arthroscopy performed with a combined sciatic-femoral nerve block, or a propofol-remifentanil general anaesthetic.

Methods: With Ethics Committee approval and written informed consent, 40 healthy patients were premedicated with intravenous midazolam (0.05 mg kg−1) and ketoprofen (50 mg). They were then randomly allocated to receive either a combined sciatic-femoral nerve block with 25 mL mepivacaine 2% (15 mL for the femoral nerve, 10 mL for the sciatic nerve) (PNB group, n = 20), or a general anaesthetic with a continuous intravenous infusion of remifentanil (0.1–0.3 μg kg−1min−1) and propofol (target plasma concentration 2–4 μg mL−1) with a laryngeal mask airway (GA group, n = 20).

Results: The median (range) preparation time was 16 (10–28) min in the PNB group and 13 (8–22) min in the GA group (P = 0.015). Ten PNB patients were directly discharged to the day-surgery unit after the procedure as compared with one GA patient (P = 0.003). Discharge from the postanaesthesia care unit (PACU) required 5 (5–20) min in the PNB group and 23 (7–95) min in the GA group (P = 0.001). Home discharge criteria were fulfilled after 277 (150–485) min in the PNB group and 170 (100–400)min in the GA group (P = 0.005). Costs related to the time spent in the PACU were lower for the PNB group (€1.10, range 0–22) compared with the GA group (€30, range €0–176) (P = 0.0005). There were no differences in total costs: PNB group €158 (range €105–194) versus GA group €160 (range €101–238) (P = 0.61).

Conclusions: In patients undergoing out-patient knee arthroscopy, the length of stay in the PACU can be shorter after a sciatic-femoral nerve block with a small volume of mepivacaine 2% compared with a propofol-remifentanil anaesthetic, and there is an increased likelihood that they will bypass the first phase of the postoperative recovery.

Background and objective: The effect of anaesthesia and surgery on gastric emptying is not constant, and this has not been previously studied in patients undergoing thoracotomy for lung resection with a bupivacaine-fentanyl epidural infusion for analgesia. There are important implications in this group of patients with regards to the recommencement of important oral medication as well as the risks of aspiration of gastric contents. The study examined gastric emptying in these patients until the second postoperative day.

Methods: In a prospective repeated measures study, the effect of fentanyl-bupivacaine epidural analgesia at the mid-thoracic level on gastric emptying was assessed in 11 patients undergoing thoracotomy for lung resection. Gastric emptying was measured using a paracetamol absorption technique. Patients acted as their preoperative controls and were assessed 4 h postoperatively and on the second postoperative day.

Results: The mean (SEM) maximum plasma paracetamol concentration was 204.6 (20.4) μmol L−1 before operation, 61 (9.5) μmol L−1 4 h postoperatively and 114.3 (22.6) μmol L−l on the second postoperative day. Mean (SEM) paracetamol absorption at 120 min was 15 638 (1441) μmol min L−l preoperatively, 5731 (821) μmol min L−l4 h postoperatively and 9325 (1759) μmol min L−l on the second postoperative day. Postoperative values were significantly (P < 0.005) less than the preoperative values.

Conclusions: After thoracotomy, gastric emptying was delayed until at least the second postoperative day in patients receiving mid-thoracic fentanyl-bupivacaine epidural analgesia.

Background and objective: The role of sevoflurane has not been studied in relation to awareness during anaesthesia. We observed the effect of sevoflurane on the incidence of awareness during cardiopulmonary bypass for open-heart surgery.

Methods: Fifty-nine patients of age >17 yr undergoing open-heart surgery with cardiopulmonary bypass were randomly assigned to two groups. In both groups, induction was with etomidate, dehydrobenzperidol and fentanyl; anaesthesia was maintained with sevoflurane, fentanyl and N2O; vecuronium was used for muscular paralysis. Group 1 (30 patients) received dehydrobenzperidol and fentanyl during cardiopulmonary bypass; Group 2 (29 patients) received sevoflurane and fentanyl. Patients were given different auditory inputs during different phases of surgery. All patients were interviewed with standard questions 8 and 24h postoperatively for evidence of explicit awareness.

Results: Five patients in the dehydrobenzperidol group gave a history of awareness (16.67%) as opposed to none in the sevoflurane group. The difference in the incidences of awareness was significant (P < 0.05), but no differences were found between the interviews conducted at 8 and 24h. Sevoflurane and opioid combination reduced the incidence of awareness in open-heart surgery.

Background and objective: Multimodality treatment for patients with locally advanced primary or locally recurrent rectal cancer, including high-dose preoperative external beam radiotherapy, extensive surgery and intraoperative radiation therapy, decreases the local recurrence rates and improves survival. During this aggressive operation, the anaesthesiologist is faced with potential problems such as major transfusion requirements, hypothermia, intraoperative position changes, the need to transport the patient to the intraoperative radiation therapy applicator, and the risks associated with remote monitoring of the patient during the 10 min intraoperative radiation therapy application. The anaesthetic management and perioperative results were evaluated for the anaesthetic results and the complications.

Methods: One-hundred-and-six patients undergoing the multimodality treatment between February 1994 and March 2000 for locally advanced primary (n = 50) and locally recurrent rectal cancer (n = 56) were retrospectively evaluated for their anaesthetic results and complications.

Results: All patients were operated upon using a combination of general and epidural anaesthesia. The average duration of anaesthesia was 6 (range 3–10.5) h and the mean blood loss 3.6 (range 0.4–14) L. All patients recovered well from anaesthesia. Two patients (2%) died in the intensive care unit (34 and 48 days postoperatively) because of adult respiratory distress syndrome following postoperative haemorrhage. Severe haemorrhage during or after the operation was significantly related with the development of adult respiratory distress syndrome (P < 0.0001).

Conclusions: With adequate preoperative assessment and optimalization of the patient's condition, maintaining peroperative haemodynamic stability with the help of adequate remote monitoring, early and fast transfusion, and multidisciplinary communication, anaesthetic complications can be minimized.

Background and objective: The effects of altering the concentration of a local anaesthetic on the development of epidural anaesthesia in pregnant females are unclear. We compared the anaesthetic effects of a constant dose of two different concentrations of epidural lidocaine for Caesarean section.

Methods: After Institutional Review Board approval and informed consent, patients undergoing elective Caesarean section were randomized to receive either lidocaine 1% 30 mL (+epinephrine 5 μg mL−1) or lidocaine 2% 15 mL (+epinephrine 5 μg mL−1) (n = 20 each) for epidural anaesthesia at the L1-L2 interspace. The spread of the sensory block to pinprick and the degree of motor block (modified Bromage scale) were measured at 5, 10, 15, 20 and 30 min after injection.

Results: No significant differences in the progression of analgesia and motor block were observed at any time between 1 and 2% lidocaine. The maximum cephalad spread was observed 30 min after injection; the median was at T4 (range T3–T5) and at T4 (range T3–T6) for lidocaine 1 and 2%, respectively.

Conclusions: The same doses but different volumes of lidocaine 1 and 2% produced comparable anaesthetic effects in pregnant females. The effects of epidural anaesthesia depend primarily on the total dose of the local anaesthetic.

Background and objective: To assess the publication rate of full papers presented as abstracts at the 1995 meeting of the European Society of Anaesthesiologists, and to assess factors that might predict subsequent full publication.

Methods: All abstracts presented at the meeting and published in the British Journal of Anaesthesia (Suppl 1, 1995) were included. To verify subsequent full publication, a MEDLINE search was performed and validated. We studied the average time from the meeting to publication, the first author's country, the subspeciality, the publishing journal of the full report, the type of presentation (oral or poster), the object of investigation, and the quality of research design and of statistical reporting in the abstract.

Results: Of 472 meeting abstracts, 199 (42.2%) were eventually published. The average (± SD) delay between meeting and publication was 16.8 (15.6) months (range 24–60 months). Most papers (79.4%) had been published within 3 yr of the meeting. Circulation, pharmacology and intensive care papers had the highest rates of publication. Sixty-three journals attracted papers, with the British Journal of Anaesthesia publishing most (n = 29). No difference in subsequent publication was found between oral and poster presentations. Randomized trials and animal research were more likely to be published. The number of authors or their positions differed between the abstract and the full publication in 145 cases (72.9%); the first author was changed in 43 cases.

Conclusions: Less than half of the abstracts accepted at the 1995 European Society of Anesthesiologists' meeting were subsequently published in journals indexed by MEDLINE in the 3 yr following the meeting. Many changes in authorship occurred between the abstract and the full publication. The study architecture and the object of investigation predicted full publication.

Background and objective: The aim was to examine the immediate effects of short-term anaesthesia on the different components of psychomotor performance of the upper extremity and cognitive functions, and to find out if there were any differences in the sensitivities of the different tests. The measured psychomotor aspects were simple reaction time, choice reaction time, speed of movement, index finger-tapping speed, co-ordination, visual spatial memory capacity, digit-symbol substitution and the Maddox Wing test.

Methods: The subjects were 30 female patients aged 24–50 yr who had been through a minor gynaecological operation. Anaesthesia had been induced with propofol and alfentanil. The measurements were mainly made with the HPM/BEP system, and the tests were performed 1 h before the anaesthesia and immediately after the wake-up.

Results: Short-term anaesthesia prolonged the simple reaction time by 7% and the choice reaction times by 25% (one-choice) and 7% (two-choice) and decreased the speed of movement by 10% (one-choice) and 19% (two-choice), index finger-tapping speed by 7% and co-ordination by 7%. In addition, visual spatial memory capacity decreased by 21%, digit-symbol substitution increased by 5% and the Maddox Wing test increased by 68%.

Conclusions: Based on the results, it seems that short-term anaesthesia reduces both signal processing at the central level, and motor control and co-ordination of movements at the peripheral level, and has a decreasing effect on motor performance in the above-mentioned aspects measured immediately after wake-up.

Background and objective: The removal of the non-functioning liver in cases of fulminant liver failure has been advocated by some authors as a means of improving haemodynamic instability and acid-base disturbances associated with acute liver failure.

Methods: The aim of the present experimental study was to investigate whether maintaining a non-functioning liver is preferable over removing it in terms of haemodynamic variables, after acute hepatic failure has been surgically induced. Twenty Landrace pigs were used in the study. All of them underwent portocaval anastomosis and ligation of the hepatic artery. After an 18-h period and with biochemical indices of fulminant hepatic failure clearly demonstrated, the animals were randomly assigned to one of two groups: in 10 pigs (Group A) the ischaemic liver was left in situ and no further surgical intervention was undertaken. The other 10 (Group B) underwent total hepatectomy. Haemodynamic monitoring was the same in both groups. No inotropes were administered throughout the whole period of observation.

Results: Haemodynamic deterioration was observed in the hepatectomized pigs (Group B) whereas the group with the ischaemic liver in situ (Group A) remained stable in terms of the haemodynamic variables evaluated until the end of the experiment. (Cardiac index in Group A 7.59 ± 1.25 Lmin−1 m-2 vs. 2.92 ± 0.68 Lmin−1 m−2 in Group B, P < 0.05.)

Conclusions: The concept of salvage hepatectomy in cases of acute liver failure should be redefined since there seems to be some experimental evidence that it may not be as beneficial as originally thought.

Background and objective: We measured the physical performance error of commercial syringe pumps incorporating the Diprifusor® module to deliver propofol by target-controlled infusion in a laboratory experiment.

Methods: Three infusion devices (Graseby 3500®; Fresenius Vial Master TCI®; Alaris TIVA-TCI®) were studied when infusing propofol at six different target concentrations (1, 1.5, 2, 3, 5, 8 µg mL−1) over 10 min. This procedure was repeated five times to average each target. Precision measurements were based on the volume delivered and a mean performance error was calculated.

Results: The mean performance error bandwidth for all pumps was from −2.07 to 4.22% with a 95% confidence interval always <4.5%. There was a higher variability at lower target concentrations associated with lower infusion rates. The Alaris TIVA-TCI® pump showed the largest overall bias whereas the Fresenius Vial Master TCI® pump had the smallest divergence and the lowest variation across all target concentrations. The Graseby 3500® pump had the smallest range in mean performance error with an acceptable variation (95% CI <2%).

Conclusions: Despite the differences between the pumps, which may be related to the synchronization of the hardware components, the absolute inaccuracies in the physical performance were low and presumably negligible from a clinical point of view.

Background and objective: Epidural analgesia with bupivacaine plus either sufentanil or fentanyl is widely used during labour, but it is not clear which opioid is to be preferred. The study compared these opioids at equianalgesic doses in terms of analgesia, onset time and side-effects.

Methods: Ninety females in active labour were entered into the randomized, double-blind trial. A test dose of bupivacaine was given into the epidural space. Parturients in Group S received sufentanil 8 mL as a bolus dose, followed by an infusion at a rate of 5 mL h−1 of a mixture containing sufentanil 1 µg mL−1 and bupivacaine 1 mg mL−1. Patients in Group F received fentanyl 8 mL as a bolus, followed by an infusion at 5 mL h−1 of a solution containing fentanyl 3.5 µg mL−1 and bupivacaine 1 mg mL−1. Additional boluses of 5 mL were of the relevant solution were given if necessary.

Results: In a ratio of 1.0 : 3.5 (sufentanil 1 µg versus fentanyl 3.5 µg), both groups reported the same analgesia with the same onset time. The onset time to obtain 50% of the initial visual analogue score was 10 and 11 min for Groups S and F, respectively. Mean visual analogue scores in Groups S and F respectively declined from 77 and 82 before epidural blockade, to 29 and 27 during the first stage of labour, and to 69 and 59 respectively during the second stage. Overall satisfaction among parturients was high (98 and 96%), particularly during the first stage (98 and 98%), and also to a large degree during the second stage of labour (74 and 79%). Furthermore, only a few extra bolus doses were required (mean 0.9 and 1.2, Groups S and F, respectively). All the females could stand on their own, and almost all (81% Group S; 79% Group F) could walk 20 m without help. There were no serious adverse effects. Moderate side-effects occurred equally often with the possible exception of less nausea and vomiting in the fentanyl group.

Conclusions: Epidural analgesia for ambulatory parturients with bupivacaine plus either sufentanil or fentanyl (ratio 1.0 : 3.5) provides good analgesia with a low frequency of modest side-effects. No clinical differences were found between the opioids.

Background and objective: Excessive alcohol consumption is a major cause for premature death and preventable ill health in the whole population. We set out to audit admissions to the intensive care unit (ICU) of our large teaching hospital with respect to alcohol-related admissions.

Methods: A prospective audit was conducted into alcohol-associated admissions to our university hospital adult ICU over 12 months. The following data were collected for each patient admitted: age, gender, diagnosis and amount of alcohol consumption, APACHE II score, and the probability of death.

Results:Patients with definite alcohol consumption constituted 39.1% (124 patients) of all admissions (317 patients) to the ICU. In Group 1 (89 patients), the admission was directly associated with alcohol consumption, which included heavy alcohol consumption. Group 2 (35 patients) consisted of ‘social drinkers’ and Group 3 was composed of patients who denied any alcohol intake. Group 1 had a high median APACHE II score of 19 (range 4–35) and a significantly higher mortality rate (41.6%) than Group 3 (18.4%, P < 0.001) or the entire sample studied (23.7%,P < 0.001). The results also demonstrated an incidence of pneumonia of 29.2% (26 patients) in alcohol-associated admissions compared with 22.8% (eight patients) in Group 2 and 21.8% (19 patients) in Group 3. Trauma admissions amounted to 11.4% of the patients studied (24/211 patients), 17 (70.8%) of whom were directly associated with alcohol and it included 11 pedestrians.

Conclusions: Alcohol may play a major role in the admission and mortality of ICU patients. Further investigations are warranted because our sample size is so small.

Background and objective: To compare the efficacy of tramadol and ondansetron in minimizing the pain due to injection of propofol in 100 patients.

Methods: An intravenous cannula was inserted in the dorsum of the hand. After tourniquet application to the forearm, tramadol 50 mg (Group 1, n = 50) or ondansetron 4 mg (Group 2, n = 50) was injected. The tourniquet was released after 20 s, and propofol 5 mL was administered over 5 s. The patients were observed and asked if they had pain in the arm and the response was assessed. Nausea and vomiting and degree of sedation were recorded for the first postoperative 24 h.

Results: Twenty-one patients in Group 1 and 14 patients in Group 2 reported no pain. Slight pain was seen in 15 patients in Group 1 and in 18 patients in Group 2. Moderate pain was seen in 10 patients in Group 1 and 15 patients in Group 2. Severe pain was seen in four of the patients in Group 1 and three patients in Group 2. There was no significant difference of pain between Groups 1 and 2, but we found a significant reduction of nausea and vomiting in the ondansetron group compared with the tramadol group (P = 0.033).

Conclusions: Tramadol or ondansetron are equally effective in preventing pain from propofol injection. The added benefit of a reduction in nausea and vomiting after operation in the ondansetron group may be a reason to prefer this drug.

Background and objective: Thiopental prolongs the QT interval more than propofol, and the two induction agents were compared in patients with subarachnoid haemorrhage predisposed to electrocardiographic abnormalities and cardiac dysrhythmias.

Methods: Twenty-nine patients were studied randomly. Anaesthesia was induced with either thiopental or propofol and fentanyl; vecuronium was used as a neuromuscular blocking agent. The electrocardiogram and arterial blood pressure were monitored from before the induction of anaesthesia to 2 min after endotracheal intubation.

Results: The median QT interval was at baseline 423 ms in the thiopental group and at 432 ms in the propofol group, and it increased in the thiopental group to 446 ms and decreased in the propofol group to 425 ms (P < 0.01 between groups). After induction and endotracheal intubation, the number of patients with increased QT dispersion was greater in the propofol group (P < 0.05). The incidence of cardiac dysrhythmias was similar in the study groups.

Conclusions: Thiopental and propofol are equally suitable for the induction of anaesthesia in patients with subarachnoid haemorrhage.

Background and objective: Diabetes mellitus associated with hypertension often causes perioperative complications. The α1-adrenoceptor antagonist/5-hydroxytryptamine-1A receptor agonist urapidil is an approved drug used in hypertension and hypertensive emergencies. 5-Hydroxytryptamine-1A (5-HT1A) receptor agonists impair glucose metabolism. To evaluate a possible dose-dependent hyperglycaemic effect of urapidil due to its 5-HT1A receptor agonistic properties, the effect of three doses of urapidil on hyperglycaemia in the streptozotocin diabetic rat was investigated.

Methods: Male Wistar-Kyoto rats were made diabetic by streptozotocin and randomly allocated to the following daily treatments for 7 days (n = 6 each): urapidil 6 mg kg−1, urapidil 20 mg kg−1, urapidil 60 mg kg−1, insulin 4 IU kg−1 subcutaneously. One diabetic group and one non-diabetic healthy group served as controls.

Results: Treatment for 7 days with urapidil 20 mg kg−1 and urapidil 60 mg kg−01 reduced mean glucose concentrations significantly (urapidil-20: 15.6 ± 1.1 mmol L−1P = 0.023; urapidil-60: 15.8 ± 0.8 mmol L−1, P = 0.04) compared with diabetic controls (20.9 ± 0.8 mmol L−1), whereas those after urapidil 6 mg kg−1 were similar to diabetic controls. Insulin treatment normalized blood glucose concentrations.

Conclusions: The α1-adrenoceptor antagonist/5-HT1A receptor agonist urapidil has no hyperglycaemic effect on experimental diabetes mellitus, even in high doses, despite its 5-HT1A receptor agonistic properties.

Background and objective: To investigate the interactions of postoperative pain and endocrine stress response, three groups of 21 patients each with total knee arthroplasty were compared in a randomized, prospective design. For postoperative pain management, a three-in-one block, an epidural catheter analgesia or an intravenous patient-controlled analgesia was used.

Methods: After standardized balanced anaesthesia, the pain intensity was measured by a visual analogue scale (VAS). For detection of epinephrine, norepinephrine, antidiuretic hormone, adrenocorticotropic hormone and cortisol in the plasma, blood samples were taken at six time points before and up to 180 min after the start of pain therapy. In addition, systolic arterial pressure, heart rate, partial arterial oxygen saturation, nausea, vomiting and satisfaction of the patients were recorded.

Results: Within 15 min after the start of pain therapy, VAS in all groups was similarly reduced from >40 mm to a range <10 mm (P < 0.001). Initially, all endocrine stress variables exceeded the normal range. Epidural anaesthesia led to a significant decrease of epinephrine and norepinephrine concentrations, while an increase was observed in the group with patient-controlled analgesia, and the decrease in patients with the three-in-one block was less than in patients receiving epidural anaesthesia (P = 0.001). Differences in antidiuretic hormone, adrenocorticotropic hormone and cortisol were less pronounced. Systolic arterial pressure decreased significantly in all groups, particularly in patients with epidural anaesthesia. Partial arterial oxygen saturation and the incidence of nausea and vomiting were comparable. All patients were satisfied with the methods used.

Conclusions: All methods of pain management led to sufficient analgesia, but they were not accompanied by an adequate reduction in endocrine stress response. Thus, postoperative pain is only a secondary stressor and sufficient analgesia with subjective well-being does not prove a stress-free state. With regard to the reduction of sympathoadrenergic stress response, epidural anaesthesia is superior to the three-in-one block and patientcontrolled analgesia. Epidural anaesthesia is recommended particularly for high-risk patients with hypertension, coronary heart disease and diabetes mellitus. In these patients, the reduction of a ‘hidden’ endocrine stress response in addition to prevention of pain is of special interest.