COMPARISON OF INFUSIONS OF ALFENTANIL OR PETHIDINE FOR SEDATION OF VENTILATED PATIENTS ON THE ITU

Transcription

1 Br.J. Anaesth. (1986), 58, COPARISON OF INFUSIONS OF ALFENTANIL OR PETHIDINE FOR SEDATION OF VENTILATED PATIENTS ON THE ITU P.. YATE, D. THOAS, S.. SHORT, P. S. SEBEL AND J. ORTON The ideal agent with which to sedate patients in the intensive care unit should provide cardiovascular stability, respiratory depression, reduction of intracranial pressure and analgesia. Ideally, it should have a pharmacokinetic profile which promotes its use by infusion. In addition, there should be minimal immunological or metabolic effects, a high therapeutic ratio and absence of active metabolites. The withdrawal of Althesin, reports of cortisol suppression with etomidate (Lambert et al., 1983) and disordered pharmacokinetics with midazolam (Byatt et al., 1984), and a general dissatisfaction with some of the more traditional agents (Gast, Fisher and Sear, 1984) have necessitated a search for new agents in this field. Alfentanil, a new short acting opioid, would appear to fulfil many of the above criteria. These include good cardiovascular stability even in high dose (Sebel, Bovill and van der Haven, 1982) and a short elimination half-life of approximately 90 min (Bovill et al., 1982) which has led to its successful use by infusion both during surgery (Ausems, Hug and de Lange, 1983) and to provide postoperative analgesia (Andrews et al., 1983; O'Connor, Escarpa and Prys-Roberts, 1983). A pilot study (Yate, Thomas and Sebel, 1984) suggested that alfentanil could be useful when given by infusion in the ITU. The purpose of the present study was to investigate the suitability of alfentanil by infusion in a comparison with pethidine. Dose requirements, adequacy of sedation, quality of recovery, P.. YATB,.B., B.S., F.F.A.R.C.S., S.. SHORT,.B., BJ., F.F.A.R.CJ., P. S. SEBEL,.B., BJ., PHJ>., F.F.A.R.CS.I. (Anaesthetics Unit); D. THOAS,.B., B.S., F.F.A.R.CS. (Department of Anaesthetics); J. ORTON, JC (Department of Chemical Pathology); London Hospital edical College, Whitechapel, London El IBB. SUARY Sedation was studied in 30 patients requiring overnight ventilation in the intensive therapy unit (ITU). Patients received an infusion of either alfentanil or pethidine, supplemented with midazolam. The infusion rates were adjusted to provide optimal sedation as judged by a nurse, and measurements were made of quality of sedation, recovery and serum cortisol concentration. In addition, blood concentrations of alfentanil were measured to permit pharmacokinetic and pharmacodynamic analysis. Satisfactory sedation was achieved in both groups. The required infusion rate for alfentanil was between 0.4 and 0.5(jg kg* 1 min* 1. Recovery was good in both groups, apart from one patient in the alfentanil group, in whom recovery was greatly prolonged and alfentanil pharmacokinetics were abnormal. A difference was found in the metabolic response to surgery between the two groups, the response in the alfentanil group being significantly less marked. metabolic effects and pharmacokinetics have been assessed. PATIENTS AND ETHODS Thirty patients requiring postoperative ventilation after open heart surgery were studied. All patients gave informed consent and the study had the approval of the local Ethics Committee. The patients were randomly allocated, using a computer-generated randomization code, to one of two groups: alfentanil (A) or pethidine (P). A standard anaesthetic technique was used: premedication was with lorazepam 2-4 mg by mouth 2 h before surgery and promethazine mg plus

2 1092 BRITISH JOURNAL OF ANAESTHESIA papaveretum mg i.m. 1 h later. Anaesthesia was provided with thiopentone, nitrous oxide in oxygen, halothane, midazolam (up to 10 mg) and fentanyl (up to 1000 ng). After surgery the patients were transferred to the intensive therapy unit and artificial ventilation (to normocapnia) continued until 8.00 a.m. the next day. Once in the intensive care unit, patients were not sedated until they had awoken from the anaesthetic defined as the time when they could open theneyes and move all four limbs to command. Patients in group A received midazolam 2.5 mg and alfentanil 15 ng kg" 1 i.v. followed by alfentanil 0.4 ug kg" 1 min" 1 by infusion. Patients in group P received midazolam 2.5 mg followed by an infusion of pethidine 10 mg h" 1. The injection of midazolam was given 3 min after the administration of the bolus of alfentanil, as earlier experience had shown that the simultaneous administration of alfentanil and midazolam could result in severe hypotension. The infusion rates in both groups were then adjusted by the ITU staff to provide optimal sedation. In addition, staff were given the option of incremental doses (alfentanil 15 ug kg" 1 or pethidine 10 mg) of opioid if the patient became unsettled, or of midazolam 2.5 mg i.v. if the patient was considered to be pain free and accepting IPPV, but too aware. After preliminary experience it was decided to give electively midazolam 2.5 mg to all the patients in the evening, to induce sleep. The quality of sedation was assessed hourly by the nursing staff using a four point sedation score: 1 = Patient asleep or awake but needing no more analgesia. 2 = ildly restless, but answers "no" when asked if more analgesia required. 3 = ildly restless, answering "yes" to above. 4 = Restless, difficult to ventilate or in obvious pain. As sedation was scored by a variety of nurses, an observer reliability study was conducted on similar patients using the method of Gelfand and Hartmann (1975). The infusions were stopped at 8.00 a.m. the next day. Regular attempts were made to initiate satisfactory spontaneous ventilation (as judged by a ventilatory rate >12 b.p.m.). The trachea was extubated after approximately a further 15 min if the Paco, was less than 6.0 kpa. The hourly ventilatory rate and arterial Pco, were recorded for 6 h after the cessation of the infusion in the spontaneously breathing patients, as were the times to spontaneous ventilation and extubation, and the first demand for further analgesic. As it is the practice in this unit to give prophylactic analgesics before removal of the chest drain, all patients were given pre-mixed 50 % nitrous oxide in oxygen to breathe during this procedure. The degree of drowsiness following the end of the infusion was assessed hourly for 6 h using an assessment scale: 1 = Asleep and unrousable. 2 = Asleep but reusable. 3 = Awake but drowsy. 4 = Wide awake and alert. This assessment was performed by one person, a research nurse, who was unaware which sedative had been prescribed. Blood was taken for the measurement of serum cortisol concentration at h on the day before surgery, h and h in the ITU on the day of surgery (during the infusion) and at h on the first day after operation. Samples to permit measurement of the concentration of alfentanil (group A) were taken before commencing the infusion of alfentanil, at 1 h,6 h, 12 h, immediately before the cessation of the infusion, and then at 2, 5, 10, 15, 30, 60, 120, 360, 400, 600 and 960 min after the discontinuation of the infusion. Samples were frozen and assayed at a later date, cortisol by a standard radioimmunoassay (Amerlex) and alfentanil by gas chromatography (Woestenborghs, ichielsen and Heykants, 1981). The distribution and elimination half-lives for alfentanil were calculated using the plasma concentrations obtained after stopping the infusion of alfentanil. These were fitted to single and bi-exponential curves by non-linear regression (BDP3R A Statistical Package, copyright Regents of the University of California). The clearance of alfentanil was calculated from the area under the curve as measured by the trapezoid rule. The volume of distribution was calculated from clearance x elimination half-life/ Statistical analyses were performed using analysis of variance, Student's t test, ann-whitney and Chi Square tests, where appropriate. Results were considered significant at P < 0.05 and are expressed as mean± SE. RESULTS Thirty patients were studied; one patient was excluded as a clinical decision was made not to ventilate the patient overnight. The demographic

3 ?Plas ALFENTANIL AND PETHIDINE COPARED ON THE ITU 1093 TABLB I. Details of patients studied (mean ± SE) Age(yr) Weight (kg) Bypass time (min) Fentanyl dose (ug) Group A (alfentanil) (n = 14) 58.3 ± ± ±56 Group P (pethidine) (n = 15) 57.4 ± ± ± ±35 data, surgical and anaesthetic details were similar between the two groups (table I). All patients had normal preoperative hepatic and renal function as determined by routine biochemical screening. The mean durations of the infusion in both groups were not significantly different: 17.2± 0.56 (range ) h in group A and 17.8 ± 0.45 (range ) h in group P. The dose of midazolam 6.16 ±2.6 mg in group P was larger, but not significantly so, than in group A, 4.28 ± 0.56 mg. The mean infusion rates of alfentanil and pethidine were 0.46 ± ug kg" 1 min" 1 and 0.3±0.02mgkg- 1 h- 1, respectively. The total dose of drug administered was 0.48 ± mg kg- 1 of alfentanil and 5.3±0.43 mg kg" 1 of pethidine. In group A, 22 extra incremental doses of alfentanil were given and the infusion rate had to be adjusted 14 times; in group P 10 incremental doses of pethidine were given and the rate had to be adjusted 56 times significantly (P < 0.05) more often than in group A. The mean infusion rates for the two groups are shown in figure 1. There was no significant increase in the infusion rate with time in either group; the initial increase in rate in the pethidine group, although not significant, possibly reflects an inadequate starting rate in some patients. Conversely, the mean plasma concentrations of alfentanil showed a significant increase (P < 0.05) during the infusion of that drug. There was no statistically significant difference in the sedation scores between the two groups (fig. 2). On only 2 % of the hourly assessments was the sedation considered totally unsatisfactory, in both groups. No clear relationship between the plasma concentrations of alfentanil and the sedation score could be found. The reliability of the nurse assessment score was 90 %. The mean times to extubation, spontaneous ventilation and first demand for analgesia are shown in table II. In both groups 80% of patients were breathing spontaneously in 30 min and in 70% the trachea was extubated within 1 h of stopping the infusion. The prolonged recovery Pethidine o Alfentanil Plasma alfentanil Q6n Q [-0.4 a g usi inf c CD ent 0.3 Q2 0.1 Time (h) l CD 2? c o 'in I -Q3 1 -a nil <o c a> "5 p FIG. 1. ean infusion rate of alfentanil and pethidine over the first 15 h of the infusion and plasma concentrations of alfentanil taken at 1, 6 and 12 h (mean ± SE).

4 1094 BRITISH JOURNAL OF ANAESTHESIA 100 -i Alfentanil n n Pethidine % 50- o n n Sedation score FIG. 2. Distribution of sedation scores expressed as a percentage of total number of hourly assessments at each score. TABLE II. Recovery times {median and range) Time end of infusion to spontaneous ventilation (min) Time to extubation (min) Time to first demand for analgesic (min) Group A (alfentanil) 15.5 (5-230) 40 (25-231) 180 Group P (pethidine) 10 (1-60) 36 (1-110) 210 times in group A were attributable to one patient (No. 15) and, if she is excluded, the mean times to spontaneous ventilation and extubation were, respectively, 16.7 ± 11.4 min and 51 ±24 min in group A and 16.8 ±5.2 min and 49.5 ±7.8 min in group P. This patient behaved quite differently to the others in the group; extubation was not possible until 230 min after the cessation of the infusion and then only after naloxone 0.3 mg i.v. Respiratory rate and arterial Pco, were satisfactory for a further 420 min when she suddenly had a kpa 55-i mm Hg o Alfentanil Pethidine Time Ch) FIG. 3. ean arterial (± SE) Pco, for 6 h after cessation of the infusion in those patients breathing spontaneously. i 6

5 ALFENTANIL AND PETHIDINE COPARED ON THE ITU 1095 oaifentanil Pethidine s S I Time (h) FIG. 4. ean ventilatory rate (± SE) for 6 h after cessation of the infusion in those patients breathing spontaneously. respiratory arrest which was treated with a further naloxone 0.4 mg i.v. and i.m. Subsequently, she made a good recovery. Examination of the patient's records could not identify any special features apart from her general physical state, which was poor, and that she was one of only two patients in the trial to receive erythromycin. There was a wide range in post-infusion demands for analgesia: two patients in group A and three in group P did not require any in the first 10 h after infusion. The mean arterial Pco, and ventilatory rate in the first 6 h in the spontaneously breathing patients are shown in figures 3 and 4. There were no significant differences between the two groups. In addition, no difference was detected in the degree of postoperative drowsiness between the two groups (fig. 5); no patient was classified as unrousable at any time. Cortisol concentration These results are shown in figure 6. There were no significant differences between the groups in the preoperative cortisol values or in those 50-i Alfentanil Pethidine 28 D O Score FIG. 5. Results of assessment scale of drowsiness in the post-infusion period. This is expressed as percentage of the total number of hourly assessments at each score.

6 1096 BRITISH JOURNAL OF ANAESTHESIA o o Alfentanil Pethidine ~ J Sample time FIG. 6. Changes in plasma cortisol concentration (mean ± SE) before, during and after the sedative infusions. *P < TABLE III. Pharmacokinetic data of alfentanil infusions. VD = Volume of distribution No. Age (yr) Weight (kg) Sex Distribution half-life (min) Elimination half-life (min) Clearance (ml kg" 1 min" 1 ) KD (litre kg" 1 ) ean SE ± ±2.3 F F ± ± ± ±0.09 obtained on the first day after operation. A marked response to the stress of surgery was seen in both groups. However, the mean cortisol concentrations at h during the infusion were significantly lower in the patients receiving alfentanil (P < 0.01). The mean duration of the infusion at this time was similar in both groups. Pharmacokinetics Thirteen complete sets of plasma concentrations were collected, of which 12 could be fined to single and bi-exponential curves. However, in all 12 a better fit, as judged by smaller residuals, was found with a bi-exponential decay. The derived phannacokinetic variables are shown in table III. No curve could be fitted to the results from patient No. 15. The mean values of the 12 patients are plotted in figure 1; the results from patient 15 are shown separately. This patient can be seen to have kinetics markedly different from those of the other 12 patients, with a greatly prolonged half-life which was estimated at 720 min and a clearance,

7 ALFENTANIL AND PETHIDINE COPARED ON THE ITU 1097 c <p as XXX) Time (h) FIG. 7. ean (± SE) plasma decay curve in the 12 patients (B) and a separate plot of the post-infusion plasma concentrations in patient No. 15 (A). estimated by extrapolating the plot, of 0.66 ml kg" 1 mkr 1. DISCUSSION Alfentanil has many theoretical advantages over other opioids in view of its high therapeutic ratio and previously published pharmacokinetic profile (Bovill et al., 1982). Concern has been expressed about pethidine on account of its potential cardiotoxicity (de Castro et al., 1979) and neurotoxicity related to nor-pethidine (iller and Jickj 1978). orphine has been reported to have unsatisfactory kinetics in renal failure (cquay and oore, 1984), a common problem in the ITU. There is little experience of the use of other opioids by infusion, most having a pharmacokinetic profile unsuited to prolonged administration by infusion. We have compared an infusion of alfentanil with one of pethidine, both supplemented with a very small dose of midazolam. Pethidine was chosen for comparison on account of its regular use in this unit. oreover, infusions of pethidine for postoperative analgesia have been extensively studied (Church, 1979; Stapleton, Austin and ather, 1979). The group of patients chosen for this study, although not typical of the patients in a general ITU (40% of the patients in our ITU) provided a homogeneous group for study. The design of this study was to use variable infusion rates to provide good sedation and rapid recovery. We were able to provide comparable sedation with both groups. However, it is suggested by the significant increase in the number of changes in rate with the pethidine group that the sedation was easier to achieve with alfentanil. The infusion rates of alfentanil required were in the range ug kg" + 1 min" 1 similar to those reported previously (Yate, Thomas and Sebel, 1984) with unsupplemented alfentanil, although supplemented with midazolam in the present study. Although the previous report suggested that alfentanil could be used as the sole agent, midazolam had to be added in the pethidine group for satisfactory sedation, and in an effort to maintain comparability between the groups midazolam was given to the alfentanil

8 1098 BRITISH JOURNAL OF ANAESTHESIA patients as well. The mean infusion rate of pethidine was comparable with that previously reported for postoperative pain relief in nonventilated patients (Stapleton, Austin and ather, 1979). The plasma concentrations of alfentanil required during the infusion were in the region ng ml" 1, a concentration the same as that required for maintenance during nitrous oxidesupplemented alfentanil for lower abdominal surgery (Ausems and Hug, 1983). The results for recovery (apart from patient No. 15) were acceptable in terms of lack of sedation and adequacy of respiratory rate and P&co, and, in terms of Pco 2 and ventilatory rate, better than previously reported (Andrews'et al., 1983) after a short infusion of alfentanil for postoperative analgesia. Although the post-infusion plasma concentrations obtained in this study have been associated with respiratory depression as measured by carbon dioxide response curves (O'Connor, Escarpa and Prys-Roberts, 1983), the mean plasma concentrations (excluding patient No. 15) of alfentanil after infusion had decreased to less than 150 ng ml" 1 by 30 min, a figure previously quoted (Ausems and Hug, 1983) for satisfactory spontaneous ventilation, a finding supported by the clinical results in this study. Although this was not primarily a pharmacokinetic study, any planned administration of a drug by infusion for a long period should take into account the known pharmacokinetics of the drug in that particular situation. ost of the previously published pharmacokinetic data of alfentanil have been obtained after a single i.v. dose in healthy volunteers (Bower and Hull, 1982) or young healthy patients undergoing surgery (cdonnell et al., 1982; Fragen et al., 1983). In one small study with patients undergoing open heart surgery (Hug, de Lange and Burm, 1983) in which the kinetics were studied before and after cardiopulmonary bypass, a prolonged elimination half-life was found after the postbypass dose, as a result of an increase in the volume of distribution. There are no reports of the pharmacokinetics of alfentanil after prolonged infusions in the intensive care unit. The results obtained in the 12 patients with an acceptable bi-exponential fit show plasma half-lives similar to those obtained by Hug, de Lange and Burm (1983) in cardiac surgical patients, when a large dose was given after surgery. In that study the prolonged elimination half-life was attributable to an increase in the volume of distribution. In our study the total volume of distribution was smaller (0.59 litre kg" 1 ) more like that seen in general surgical patients (Schuttler and Stoeckel, 1982), the prolonged elimination half-life of 162 min in our study being the result of the reduced rate of clearance, 2.66 ml kg" 1 min" 1. The possible advantages of alfentanil are clouded by the potentially serious problem encountered in patient No. 15. Assuming the respiratory arrest was the result of the termination of the effect of the first dose of naloxone, the long duration of action of naloxone is surprising, although a wide variation in the half-life of naloxone has been reported (Aitkenhead et al., 1984). The plasma concentrations measured in this patient showed a greatly reduced clearance, leading to a very prolonged half-life. The reasons for this are unclear; there was no biochemical evidence of hepatic dysfunction, which might be expected to reduce clearance. This patient was receiving erythromycin, a drug that has been reported to interfere with drug excretion (Green and Clement, 1983). However, another patient in the trial (No. 7) was also receiving erythromycin and she had a normal clearance of 3.32 ml kg" 1 min" 1, although a rather prolonged elimination half-life (253 min). There have been other reports of prolonged elimination half-life of alfentanil as a result of a reduced clearance, although not of this order. Notably, cdonnell, Bartokowski and Kahn (1984) suggested that it may be caused by an abnormality of hepatic hydroxylation similar to that seen with debrisoquine and phenacetin an abnormality found in approximately 10% of the population. The introduction of any new drug to the ITU requires careful investigation of any metabolic effect. Previous studies of the metabolic effects of alfentanil have been confined to the effect of large doses administered before open-heart surgery. In one study (de Lange et al., 1983) alfentanil 42 ug kg" 1 given before surgery caused marked suppression of the metabolic responses to surgery, similar to that seen with sufentanil and fentanyl. In another study which investigated the metabolic effects of the administration of fentanyl 50 ng kg" 1 after surgical stimulus (Bent et al., 1984), it was suggested that, once established, the normal stress response could not be suppressed by opioids. In our study, a surprising finding was the marked difference between the two groups. The reason for this is unclear; it may be an effect, on a reduced

9 ALFENTANIL AND PETHIDINE COPARED ON THE ITU 1099 scale, similar to that seen with alfentanil during cardiac surgery, or a reflection of better sedation produced by alfentanil. If alfentanil is to be used for prolonged infusions in the ITU, further investigations of the metabolic effect are required. In conclusion, we have shown that alfentanil at an infusion rate of ng kg" 1 min" 1 or pethidine at a mean infusion rate of 0.3 mg kg" 1 h" 1 can be used as the main agent for sedation of ventilated patients in the ITU, with good recovery in the majority. ACKNOWLEDGEENTS The authors thank Janssen Pharmaceuticals for financial support and supplies of alfentanil, Dr H. akin for performing cortisol assays, iss C. Rutherfoord, S.R.N. for performing the assessments and the cardiac surgeons, anaesthetists and ITU nursing staff for their help and co-operation. Also, thanks to rs Amanda Plumer for her secretarial assistance. REFERENCES Aitkenhead, A. R., Derbyshire, D. R., Pinnock, C. A., Achola, K., and Smith, G. (1984). Phannacokinetics of intravenous naloxone in healthy volunteers. Anesthesiology, 61, A381. Andrews, C. J. H., Sinclair,., Prys-Roberts, C, and Dye, A. (1983). Ventilatory effects during and after continuous infusion of fentanyl or alfentanil. Br.J. Anaesth., 55, 21 IS. Ausems,. E., and Hug, C. C. (1983). Plasma concentrations of alfentanil required to supplement nitrous oxide anaesthesia for lower abdominal surgery. Br.J. Aruusth., 55, 191S. de Lange, S. (1983). Variable rate infusion of alfentanil as a supplement to nitrous oxide anesthesia for general surgery. Anesth. Analg., 62, 982. Bent, J.., Paterson, J. L., ashiter, K., and Hall, G.. (1984). Effects of high-dose fentanyl anesthesia on the established metabolic and endocrine response to surgery. Anesthesiology, 39, 19. Bovill, J. G., Sebel, P. S., Blackburn, C. L., and Heykants, J. (1982). The pharmacokinetics of alfentanil (R39209): A new optoid analgesic. Anesthesiology, SI, 439. Bower, S., and Hull, C. J. (1982). The comparative pharmacokinetics of fentanyl and alfentanil. Br. J. Anaesth., 54, 871. Byatt, C.., Lewis, L. D., Dawling, S., and Cochrane, G.. (1984). Accumulation of midazolam after repeated dosage in patients receiving mechanical ventilation in an intensive care unit. Br. ed. J., 289, 799. de Castro, J., van de Water, A., Wuters, L., Xhonneux, R., Reneman, R., and Kay, B. (1979). Comparative study of cardiovascular, neurological and metabolic side-effects of eight narcotics in dogs. Ada Anaesthesiol. Belg., 30, 5. Church, J. J. (1979). Continuous narcotic infusions for relief of postoperative pain. Br. ed. J., 1, 977. Fragen, R. J., Booij, L. H. D. J., Braak, G. J. J., Vree, T. B., Heykants, J., and Crul, J. F. (1983). Pharmacokinetics of the infusion of alfentanil in man. Br. J. Anaesth., 55, Gast, P. H., Fisher, A., and Sear, J. W. (1984). Intensive care sedation now. Lancet, 2, 863. Gelfand, D.., and Hartmann, D. P. (1975). Child Behaviour Analysis and Therapy, p Oxford: Pergamon Press. Green, J. A., and Clementi, W. A. (1983). Decrease in theophylline clearance after the administration of erythromycin to a patient with obstructive lung disease. Drug Intell. Clin. Pharm., 17, 370. Hug, C. C, de Lange, S., and Burm, A. G. L. (1983). Alfentanil pharmacokinetics in patients before and after cardiopulmonary bypass (CPB). Anesth. Analg., 62, 245. Lambert, A., itchell, R., Frost, J., Ratcliffe, J. G., and Robertson, W. R. (1983). Direct in vitro inhibition of adrenosteroidogenesis by etomidate. Lancet, 2, de Lange, S., Stanley, T. H., Boscoe,. J., de Bruijn, N., Berman, L., Green, O., and Robertson, D. (1983). Catecholamine and cortisol responses to sufentanil-o, and alfentanil-o, anaesthesia during coronary artery surgery. Can. Anaesth. Soc. J., 30, 248. cdonnell, T. E., Bartkowski, R. R., Bonilla, F. A., Henthorn, T. K., and Williams, J. J. (1982). Nonuniformity of alfentanil pharmacokinetics in healthy adults. Anesthesiology, 57, A236. Kahn, C. (1984). Evidence for polymorphic oxidation of alfentanil in man. Anesthesiology, 61, A284. cquay,., and oore, A. (1984). Be aware of renal function when prescribing morphine. Lancet, 2, 284. iller, R. R., and Jick, H. (1978). Clinical effects of meperidine in hospitalized medical patients. J. Pharmacol. Exp. Ther., 112, 191. O'Connor,., Escarpa, A., and Prys-Roberts, C. (1983). Ventilatory depression during and after infusion of alfentanil in man. Br. J. Anaesth., 55, 217S. Schuttler, J., and Stoeckel, H. (1982). Clinical pharmacokinetics of alfentanil. Anaesthesist, 31, 10. Sebel, P. S., Bovill, J. G., and van der Haven, A. (1982). Cardiovascular effects of alfentanil anaesthesia. Br. J. Anaesth., 54, Stapleton, J. V., Austin, K. L., and ather, L. E. (1979). A pharmacokinetic approach to postoperative pain: Continuous infusion of pethidine. Anaesth. Intens. Care, 7, 25. Wocstenborghs, R., ichielsen, L., and Heykants, J. (1981). Rapid and sensitive gas chromatographic method for the determination of alfentanil and sufentanil in biological samples. J. Chromatogr., 224, 122. Yate, P.., Thomas, D., and Sebel, P. S. (1984). Alfentanil infusion for sedation and analgesia in intensive care. Lancet, 1, 396.