Consequences of fluid absorption during transurethral resection of the prostate using distilled water or glycine 1.5 per cent

This prospective and randomized study compared the consequences of two irrigating fluids, distilled water and glycine for transurethral prostatectomy. Forty-nine consecutive unselected patients undergoing transurethral resection of the prostate with spinal anaesthesia were investigated. The irrigating fluid was either distilled water (group A, 24 patients) or glycine 1.5 per cent (group B, 25 patients). The absorption of irrigating fluid was measured, all surgical events and any clinical signs of TURP syndrome during and after surgery were recorded. Early signs of TURP syndrome were observed in one patient in group A and in four in group B without further consequence. From all the biological variables, only plasma protein concentration, haematocrit, free plasma haemoglobin and free bilirubin concentrations were found to have changed. Plasma protein concentration and haematocrit decreased significantly during and after surgery in the two groups. Free plasma haemoglobin increased significantly with time: a significantly higher concentration was observed in group A than group B. Free bilirubin concentration increased with time in the two groups and was statistically greater in group A. With the two irrigating fluids, we observed a significant amount of haemolysis and haemodilution without clinical consequences. A low irrigating fluid pressure, a short resection time, and the use of spinal anaesthesia seems to us to be essential. Close observation of patients following transurethral prostatectomy is needed but the choice of the irrigating fluid does not seem to be important. Cette étude prospective et randomisée a comparé les conséquences de deux liquides d’irrigation, l’eau distillée et le glycocolle au cours de la chirurgie prostatique par voie basse. Quarante-neuf patients consécutifs, non sélectionnés, bénéficiant d’une résection transuréthrale de prostate sous rachianesthésie, sont étudiés. Le liquide d’irrigation est soil l’eau distillée (groupe A, 24 patients), soit le glycocolle à 1.5 pour cent (groupe B, 25 patients). L’absorption des liquides d’irrigation, les incidents chirurgicaux et les signes cliniques du syndrome de réabsorption sont systématiquement notés. Des signes caractéristiques du syndrome de réabsorption sont observés dans un cas dans le groupe A et dans quatre cos dans le groupe B sans conséquence ultérieure. Seuls la protidémie, l’hématocrite, l’hémoglobine plasmatique libre et la bilirubine libre changent significativement. L’hémoglobine plasmatique libre augmente significativement dans les deux groupes, avec un taux plasmatique significativement plus élevé dans le groupe A. Avec les deux liquides d’irrigation, on observe une hémolyse et une hémodilution sans conséquence clinique. Un temps de résection court, un faible niveau de pression d’irrigation, l’utilisation de l’anesthésie rachidienne nous semblent trés importants. Une surveillance clinique et biologique attentive pendant et après la résection transuréthrale de prostate sont nécessaires; en revanche, le choix du liquide d’irrigation ne semble pas l’élément essentiel.     

Delayed seizures following sedation with propofol

Seizures occurred in two adolescents approximately six hours after sedation with propofol for bone marrow biopsy. Case #1 was a patient with chronic renal failure, hypertension, and anaemia. Case #2 had just been diagnosed with acute lymphocytic leukaemia. Neither child had experienced seizures before, and both recovered without neurological sequelae. Although other factors may have caused the seizures, the episodes have raised concerns about the safety of propofol for patients travelling home after outpatient surgery. Further study is required to explain the cause of this complication or, at least, to identify risk factors.     

Epidural fentanyl and caesarean section: when should fentanyl be given?

Epidural fentanyl is often added to epidural local anaesthetic agents to improve the quality of anaesthesia obtained during Caesarean section. Fentanyl may be given either before or after delivery of the infant. When given before delivery, fentanyl has not been reported to cause neonatal depression, although this remains a concern. A prospective, randomized, double-blind study was undertaken to determine if fentanyl was more effective if given before or after delivery of the baby in 64 women undergoing Caesarean section under lidocaine epidural anaesthesia. Maternal outcome was determined by time to achieve T4 neural blockade, the dose of lidocaine necessary to achieve this block and intraoperative scores for pain, nausea, vomiting, shivering, and sedation. Neonates were assessed by umbilical arterial blood pH and Apgar scores. No differences were detected in either group with respect to maternal or neonatal outcome. We recommend using only epidural local anaesthetic agents before delivery, and giving epidural fentanyl following delivery of the infant.     

Oxygen embolism due to hydrogen peroxide irrigation during cervical spinal surgery

Hydrogen peroxide is used to cleanse and irrigate wounds. As it decomposes immediately into water and oxygen on contact with organic tissue, it is usually regarded as a safe agent. We report a case of oxygen embolism associated with hydrogen peroxide irrigation of the surgical field during anterior fusion of the cervical vertebrae. It was accompanied by precipitous hypotension and decrease in pulse oximetry oxygen saturation and end-tidal CO₂ tension. Semi-closed spaces formed under the apatite dowel and between the apatite dowel and vertebral bodies may have precipitated the absorption of oxygen bubbles into the vasculature. Although this case was associated with a rapid recovery and uneventful sequelae, it discourages the use of hydrogen peroxide in this procedure because of the potential hazards including cardiovascular collapse. On utilise le peroxyde d’hydrogène pour nettoyer et irriguer les plaies. Comme il se décompose instantanément en eau et en oxygène au contact d’un tissus organique, on le considère inoffensif. Nous rapportons un cas d’embolie associée à l’irrigation d’une plaie au peroxyde d’hydrogène pendant une fusion cervicale antérieure. L’embolie s’est manifestée par une hypotension subite et une chute de la saturation en oxygène et du CO₂ téléexpiratoire. La présence d’espaces semi-clos sous le pivot d’apatite et entre le pivot d’apatite et le corps vertébral peut avoir précipité l’absorption des bulles d’oxygène dans l’arbre vasculaire. Bien que ce malade ait récupéré rapidement et sans séquelles, il est préférable de ne pas utiliser le peroxyde d’hydrogène pendant cette intervention à cause du danger inhérent de collapsus vasculaire.     

Early respiratory depression during Caesarean section following epidural meperidine

A case of respiratory depression which occurred following administration of epidural meperidine during Caesarean section is described. Epidural meperidine, 75 mg (10 mg.ml-1) was given after delivery of the infant to provide postoperative analgesia. Oxygen desaturation (SaO2 90%) and a decrease in respiratory rate (4.min-1) were noted 30 min after epidural meperidine was administered. Naloxone, 0.1 mg, was given iv which resulted in prompt improvement in both respiratory rate and oxygen saturation. Vascular absorption of meperidine from the epidural venous plexus is the most probable explanation for this case of early respiratory depression. We recommend a maximum bolus dose of 50 mg of epidural meperidine for pain management after Caesarean section. It is also important to monitor oxygen saturation continuously during the intraoperative period, and to monitor the patient closely in the recovery room for at least one hour for evidence of respiratory depression.     

Anaesthetic management of phaeochromo-cytoma associated with tricuspid atresia

The anaesthetic management of a patient with phaeochromocytoma, tricuspid atresia and pulmonary vascular stenosis is reported. The patient received no preoperative preparation with adrenergic blockers. Anaesthesia was induced and maintained with fentanyl, diazepam and sevoflurane. Intraoperative blood pressure was controlled with sodium nitroprusside, sevoflurane, phentolamine, and propranolol. For hypotension after resection of the tumour norepinephrine was required. This patient did not have a systemic to pulmonary shunt procedure performed, so the maintenance of pulmonary blood flow in the presence of haemodynamic instability during operation for phaeochromocytoma was a major concern. Monitoring of oxyhaemoglobin saturation (SpO2) with a pulse oximeter was considered to be useful because SpO2 may reflect pulmonary flow. During serious haemodynamic disturbances due to the manipulation of the tumour, the heart rate was inversely correlated with SpO2, but the relationship between mean arterial pressure and SpO2 was weak. Therefore, control of heart rate appeared to be more important than control of blood pressure in this case.     

Fluid through the epidural needle does not reduce complications of epidural catheter insertion

Epidural catheter insertion may be associated with blood vessel trauma or nerve root irritation. The purpose of this study was to assess whether the injection of small amounts of fluid through the Tuohy needle prior to catheter insertion reduced the incidence of these and other minor complications. Two hundred patients in labour, requesting epidural analgesia, were randomly assigned to one of three groups: Group I--the catheter was inserted without previous injection of fluid; Group II--3 ml, 1.5 per cent lidocaine hydrochloride was injected through the needle prior to catheter insertion; Group III--3 ml, saline was injected prior to catheter insertion. There were no differences among the groups in the incidence of blood vessel trauma or paraesthesiae. We conclude that there is no advantage in injecting of fluid routinely into the epidural space prior to catheter insertion.     

Apnoea and unconsciousness after apparent recovery from alfentanil-supplemented anaesthesia

Several cases of recurrent respiratory depression progressing to apnoea and unconsciousness after apparent recovery from sufentanil have been reported recently. Alfentanil has the shortest elimination half-time of the narcotics used in anaesthesia, suggesting that it should be the least likely to cause postoperative respiratory depression. A case of recurrent unconsciousness and respiratory arrest after apparent recovery from alfentanil-isoflurane-nitrous oxide anaesthesia is reported. A total dose of 137 micrograms.kg-1 alfentanil was given over a 3.25-hr period to a 45-year-old female undergoing partial gastrectomy. Naloxone, 0.16 mg IV, rapidly restored spontaneous ventilation and consciousness. This case demonstrates that apnoea and unconsciousness can also recur after apparent recovery from alfentanil. Recovery room personnel should be aware of this phenomenon. Earlier detection may permit treatment before apnoea occurs. Patients given narcotic-supplemented anaesthesia should be monitored by capnography and/or pulse oximetry in the early postoperative period.     

Lansoprazole reduces preoperative gastric fluid acidity and volume in children

The purpose of this study was to explore the efficacy of lansoprazole, a proton pump inhibitor, in reducing the acidity and volume of gastric aspirate in children immediately following the induction of anaesthesia. One hundred healthy in-patients aged 3–11 yr undergoing elective surgery were randomly allocated to four groups (n = 25 each): lansoprazole-lansoprazole, placebo-placebo, placebo-lansoprazole, and lansoprazole-placebo. For each treatment regimen, the first medication was administered at 9:00 pm on the night before surgery and the second at 5:30 am on the morning of the day of surgery (three hours preoperatively). The dose of lansoprazole was 30 mg (approximately 1.4 mg · kg^?1 mean). Children were offered 10 ml · kg^?1 apple juice three hours before induction of anaesthesia. After induction of anaesthesia and tracheal intubation, gastric fluid was aspirated through a large-bore, multiorifice orogastric tube and analyzed for pH and total fluid volume. Lansoprazole increased gastric fluid pH and decreased gastric fluid volume regardless of whether it was administered before or after placebo. Two consecutive doses of lansoprazole was the most effective means of increasing the pH and reducing the volume of gastric aspirate; in this group, there were no subjects with gastric aspirate volume >0.4 ml · kg^?1 and pH <2.5. Oral lansoprazole, at least 30 mg, given on the night before surgery or on the morning of surgery will improve the gastric environment at the time of induction of paediatric anaesthesia. The most effective regimen was two doses (at bedtime and on the morning) of lansoprazole.     

Perioperative stroke Part I: General surgery,carotid artery disease,and carotid endarterectomy

Although stroke, defined as a focal neurological deficit lasting more than 24 hr, is uncommon in the perioperative period, its associated mortality and long-term disability are high. No large-scale data are available to identify the importance of recognized risk factors for stroke in the perioperative period. A review of the literature shows that the incidence and mechanism of its occurrence are influenced by the presence of cardiovascular disease and the type of surgery. The most common cause of perioperative stroke is embolism. In non-cardiac surgery, the incidence of perioperative stroke is higher among the elderly. Properly administered, controlled hypotension is associated with minimal risk of stroke. Cerebral vasospasm may be the cause of focal cerebral ischaemia in eclamptic patients, and the aggressive treatment of hypertension may exacerbate the neurological damage. The risk of stroke associated with carotid endarterectomy is closely related to the preoperative neurological presentation, and the experience of the surgical/anaesthetic team. Symptomatic cerebrovascular disease, acute stroke, asymptomatic carotid lesions, preoperative assessment of risk, local and general anaesthesia, cerebral protection and monitoring during carotid endarterectomy are discussed with reference to reducing the risk of perioperative stroke. Adequate monitoring and protection have minimized the risk of ischaemia from carotid clamping, and the major mechanism of stroke is embolization.     

A comparative study of patient-controlled epidural fentanyl and single dose epidural morphine for post-caesarean analgesia

In a prospective, randomized, double-blinded study, 23 patients who had undergone Caesarean delivery under epidural anaesthesia were assessed to evaluate the effectiveness of patientcontrolled epidural analgesia (PCEA) with fentanyl compared with a single dose of epidural morphine for postoperative analgesia. Group A (n = 11) received epidural fentanyl 100 μg intraoperatively, then self-administered a maximum of two epidural fentanyl boluses 50 μg (10 μg · ml^?1) with a lockout period of five minutes for a maximum of two doses per hour. Group B (n = 11) received a single bolus of epidural morphine 3 mg (0.5 mg · ml^?1) intraoperatively and received the same instructions as Group A but had their PCA devices filled with 0.9% NaCl. Patients were assessed up to 24 hr for pain, satisfaction with pain relief, nausea and pruritus using visual analogue scales (VAS). The treatments for inadequate analgesia, nausea and pruritus as well as time to first independent ambulation were recorded. The ventilatory response to carbon dioxide challenge was measured at four and eight hours. Pain relief, satisfaction with pain relief, and the use of supplemental analgesics were similar in both groups. The mean 24 hr dose of epidural fentanyl used by group A patients was 680 μg. Pruritus was less common in Group A patients at the 8 and 24 hr observation periods (P < 0.0125). Both groups experienced the same degree of nausea and clinically unimportant respiratory depression. We conclude that PCEA with fentanyl provides analgesia equal to a single dose of epidural morphine and may be suitable for patients who have experienced considerable pruritus after epidural morphine adminstration.     

Preoperative naproxen sodium reduces postoperative pain following arthroscopic knee surgery

This study was undertaken to assess the efficacy of a single preoperative dose of naproxen sodium in reducing postoperative pain and length of day surgery stay in patients undergoing arthroscopic knee surgery. A randomized, double-blind clinical trial was carried out on 66 ASA I and ASA II patients scheduled for arthroscopic knee surgery. The treatment group (n = 26) received two capsules containing 275 mg of naproxen sodium each, and the control group (n = 40) received placebo. Preoperative and postoperative visual analogue pain scores, postoperative analgesic requirements in hospital as well as 24 hr after discharge, and length of day surgery stay were studied. There was a decrease in postoperative pain, both in hospital (naproxen 0.7 ±1.2 vs placebo 2.2 ±2.3) and at 24 hr after discharge (naproxen 0.8 ±1.9 vs placebo 3.8 ±3.2) (P = 0.0001). There was no difference in the need for in-hospital postoperative analgesics or in the time to discharge. However, there was a difference in the use of analgesics after discharge (naproxen group 30.4% vs placebo group 71.4%) (P < 0.01). The results of this study suggest that a single preoperative dose of 550 mg naproxen sodium is effective in reducing postoperative pain in arthroscopic knee surgery, both in the immediate postoperative period and for up to 24 hr after the completion of surgery.     

Haemodynamic and cerebral blood flow alterations after reduction of increased cerebrospinal fluid pressure in dogs

To clarify some of the mechanisms for the hypotension that may occur after cranial decompression, the authors examined alterations in cerebral blood flow (CBF) and systemic and pulmonary haemodynamic variables when cerebrospinal fluid (CSF) pressure was increased and then suddenly reduced in eight anaesthetized dogs. After CSF pressure was elevated to 50-85 mmHg for two hours, CBF decreased from 46.3 +/- 4.4 to 31.6 +/- 8.5 ml.100 g-1.min-1 (mean +/- SD, P less than 0.01). Mean systemic arterial pressure (MAP), mean pulmonary artery pressure (MPAP), pulmonary artery wedge pressure (PAWP), and systemic vascular resistance index (SVRI) increased by 20 +/- 11 mmHg, 3.9 +/- 2.5 mmHg, 5.2 +/- 3.3 mmHg, and 1448 +/- 1377 dynes.sec.cm-5.m2 from baseline values, respectively (P less than 0.01). Rapid reduction of increased CSF pressure caused CBF to increase to 61.5 +/- 19.1 ml.100 g-1.min-1, whereas MAP, MPAP, PAWP, and SVRI decreased by 22 +/- 11 mmHg, 2.4 +/- 0.9 mmHg, 2.3 +/- 2.0 mmHg, and 1289 +/- 1237 dynes.sec.cm-5.m2 from previous values (P less than 0.01) at 30 min following the decompression. However, cardiac index and pulmonary vascular resistance index remained unchanged during the study period. The present animal data indicate that the decrease in MAP after decompression is mainly a result of a reduction in systemic vascular resistance.     

Uterine and umbilical blood flow velocity during epidural anaesthesia for Caesarean section

The purpose of this study was to use colour Doppler to determine the effect of epidural anaesthesia on the uterine and umbilical blood flow velocities. After determining the precision of the technique, Doppler insonation of the uterine and umbilical arteries was performed in consenting non-labouring patients requesting epidural anaesthesia for Caesarean section. Patients in Group I were normal and those in Group II were at high risk for uteroplacental blood flow abnormalities. The pulsatility indexes (PI) of both uterine and umbilical arteries were compared at the following times: control, after fluid and after anaesthesia using repeated measure analysis of variance. In Group I (n = 30) the PI increased from 0.72 to 0.82 in the left uterine artery and from 0.71 to 0.85 in the right uterine artery (P < 0.05). In Group II (n = 10) the PI increased from 0.67 to 0.85 in the left uterine artery (NS) and from 0.98 to 1.38 in the right uterine artery (P < 0.05). There was no change in the PI in the umbilical artery. We conclude that the PI of the uterine arteries increases after epidural anaesthesia with lidocaine, epinephrine and fentanyl but there is no change in the umbilical PI. While these changes do not appear to be clinically important in the low-risk population, further studies are required to determine the impact on fetuses at high risk for in utero hypoxaemia.     

Diuretic-induced hypokalaemia inducing torsades de pointes

Torsades de pointes (TP), an unique polymorphous type of ventricular tachycardia, is associated with either an acquired or congenitally prolonged QT interval. Several reports have demonstrated TP to follow an acquired prolonged QT interval secondary to chronic hypocalcaemia, hypomagnesaemia, or hypokalaemia. We report a rapid onset, acute extracellular hypokalaemia not associated with other electrolyte disturbances inducing a prolonged QT interval followed by TP. This is the first case report of a rapid onset isolated acute extracellular hypokalaemia inducing TP. Since anaesthetists are involved in therapies that will rapidly reduce extracellular potassium (diuretic, catecholamine, and/or insulin administration, hyperventilation), this case report serves as a warning that such therapy may have the risk of arrhythmia induction.     

Malignant hyperthermia and calcium-induced heat production

The abnormal increase in intracellular Ca⁺⁺ in malignant hyperthermia (MH) is well documented, but the link between the increased Ca⁺⁺ concentration and high temperature remains speculative. We investigated the possibility that the Ca⁺⁺-induced change in the state of cell membranes may contribute to the temperature elevation. Calcium ion transforms phosopholipid membranes from the fluid to solid state. This is analogous to the freezing of water, and liberates latent heat. Differential titration calorimetry (DTC) measures heat production or absorption during ligand binding to macromolecules. When CaCl₂ solution was added to anionic dimyristoylphosphatidic acid (DMPA) and dimyristoylphosphatidylglycerol (DMPG) vesicle membranes in incremental doses, DTC showed that the heat production suddenly increased when the Ca⁺⁺ concentration exceeded about 120/μM. At this Ca⁺⁺ concentration range, these lipid membranes underwent phase transition. The latent heat of transition was measured by differential scanning calorimetry (DSC). The values were 7.1 ± 0.7 (SD, n = 4) kcal · mol^?1 of DMPA and 6.8 ± 0.7 (SD, n = 4) kcal · mol^?1 of DMPG. The study shows that Ca⁺⁺ produces heal when bound to lipid membranes. We are not proposing, however, that this is the sole source of heat. We contend that the lipid phase transition is one of the heat sources and it may trigger a hypermetabolic state by elevating the temperature of cell membranes. Because Ca⁺⁺ is implicated as the second messenger in signal transduction, multiple systems may be involved. More studies are needed to clarify how Ca⁺⁺ increases body temperature.     

Multicore myopathy in a patient with anhidrotic ectodermal dysplasia

We report a patient with multicore myopathy, a rare myopathy not previously reported in the anaesthetic literature. It is characterised by a myopathy of proximal muscles which tends to follow a benign course but may be associated with a severe form of cardiomyopathy. The myopathy is related to central core disease so these patients should be considered to have a potential for developing malignant hyperthermia. Complicating this case was an associated anhidrotic type of ectodermal dysplasia resulting in the absence of sweating, febrile episodes, recurrent pulmonary infections, conical and missing teeth, scaly skin and fine, sparse hair. The patient had a scoliosis repair which was uneventful but died three weeks later following a major pulmonary aspiration while on the ward. The cause of the aspiration is thought to have been unsuspected laryngeal incompetence associated with ectodermal dysplasia, the myopathy involving his bulbar muscles and analgesic medication.     

Randomized comparison of outcome after propofol-nitrous oxide or enflurane-nitrous oxide anaesthesia in operations of long duration

A randomized, prospective, comparative study was performed to evaluate induction characteristics, haemodynamic changes and recovery in 60 ASA I-II patients undergoing mainly gynaecological laparotomies with either propofol or thiopentone-enflurane anaesthesia. The propofol group (n = 30) received 2 mg.kg-1 propofol for induction of anaesthesia followed by propofol infusion. The thiopentone-enflurane group (n = 30) received thiopentone 4 mg.kg-1 for induction followed by enflurane (0.5-2 per cent). All patients received nitrous oxide (66 per cent] in oxygen begun one minute after tracheal intubation, and fentanyl (1.5 micrograms.kg-1) four minutes prior to induction. Other drugs administered during or after anaesthesia were similar among the groups. Haemodynamic measurements were similar between propofol and enflurane groups except after tracheal intubation when the mean arterial pressure was lower in the propofol group (P less than 0.05). The propofol group had significantly less (P less than 0.01) emesis in the recovery room than the enflurane group. The propofol group experienced significantly less (P less than 0.05) dizziness, depression/sadness and hunger than the enflurane group in the postoperative period as assessed with a visual analogue questionnaire. We conclude that propofol provided better outcome than enflurane in terms of these nonvital but annoying outcome measures after relatively long intra-abdominal operations.     

Interpleural analgesia improves pulmonary function after cholecystectomy

The purpose of this study was to examine the effects of interpleural bupivacaine on analgesia and ventilatory capacity after cholecystectomy. Forty-two patients undergoing elective cholecystectomy were randomly assigned to two groups: one to receive interpleural administration of bupivacaine-adrenaline mixture (Group 1 = 22 patients) and the other standard administration of intramuscular meperidine (Group 2 = 20 patients) for postoperative pain relief. The intensity of pain was evaluated by a visual analogue scale (VAS) preoperatively as well as at 2, 8, 24 and 48 hr postoperatively. At the same time, FVC and FEV1.0 measurements were obtained for all patients. The group given interpleural bupivacaine had better pain relief with mean VAS of 0.6 +/- 0.9 (mean +/- SD) 1.1 +/- 1.4, 0.6 +/- 0.9 and 0.8 +/- 1.2 compared with 5.2 +/- 2.2, 5.8 +/- 2.7, 5.5 +/- 2.2 and 4.5 +/- 1.8 for patients receiving meperidine (P less than 0.001). The patients in Group 1 also had larger FVC and FEV than those in Group 2: FVC 22 +/- 14.5 per cent vs 32 +/- 15.2 per cent (P less than 0.005), FEV1.0 25 +/- 15.5 vs 38 +/- 14.8 per cent (P less than 0.001) (mean +/- SD). We conclude that the interpleural analgesia can achieve better pain relief with greater ventilatory capacity than a standard analgesic regimen in the first two days after cholecystectomy.     

Haemodynamic changes during the apnoea test for diagnosis of brain death

Haemodynamic responses to the apnoea test jor the diagnosis of brain death were investigated in nine patients with severe head injury or cerebrovascular disease. To prove apnoea, the ventilator was disconnected for ten minutes and oxygen was insufflated to avoid hypoxaemia. No respiratory movement was séen in any patient. Ten minutes after disconnecting the ventilator, PaCO₂ was increased to 78 ± 3 mmHg and pH was reduced to 7.17 ± 0.02. Adequate oxygenation was maintained in all patients. Cardiac output increased from 4.8 ± 0.7 to 5.7 ± 0.8 L · min^{? 1} (P < 0.05), and mean pulmonary artery pressure increased from 11 ± I to 17 ± 2 mmHg (P < 0.01). However, mean arterial pressure, heart rate, pulmonary artery wedge pressure and right a trial pressure did not change. Plasma catecholamines were measured in three patients. Plasma norepinephrine concentrations increased in all three patients but the changes in plasma epinephrine were minimal. These circulatory responses to acute hypercapnia were less than those reported in awake volunteers and in patients during general anaesthesia. However, since plasma norepinephrine concentration increased during the test, some sympathoadrenal response, probably of spinal origin, was present, and may have prevented the direct depressant circulatory effects of acute hypercapnia. In conclusion, the apnoea test did not produce haemodynamic disturbances when respiratory acidosis was limited toapH 7.17±0.02 and PaCO₂ 60–80 mmHg.