Buprenorphine kinetics was determined in surgical patients using radioimmunoassay. Buprenorphine was measured in the plasma of 24 patients who had ...
Buprenorphine kinetics kinetics was determined in surgical patients using radioimmunoassay. was measured in the plasma of 24 patients who had received 0.3 mg intraoperatively. After 3 hr 10 of these patients then received a further 0.3 mg intravenously for postoperative pain relief, and 11 patients were given 0.3 mg intramuscularly: again, plasma levels were measured for 3 hr. The data fitted closely to a triexponential decay curve. There was a very fast initial phase, with a half (t1/2) of 2 Min. The terminal t1/2 was slow, approximately 3 hr. Comparison of the kinetics of the same patient, awake and anesthetized, showed that the clearance was significantly lower in the anesthetized state. A notable feature of the drug given intramuscularly is rapid systemic availability, so that peaks are obtained in 2 to 5 min, and in 10 min the resulting levels are the same as for the intravenous and intramuscular routes.
Buprenorphine Buprenorphine buprenorphine buprenorphine
Roy E. S. Bullingham, M.D., Henry J. McQuay, M.D., Andrew Moore, D.Phil., and Martin R. D. Bennett, B.A.* Oxford, England Nuffield Departments of Anaesthetics and Clinical Biochemistry, Radcliffe Infirmary
Buprenorphine is a synthetic opiate analgesic with mixed agonist-antagonist properties and a low abuse potentia1.5 Its long action and unusual receptor kinetics' lead to particular interest in its kinetics. We have used the drug intraoperatively and postoperatively in man, and now report its plasma kinetics as determined by radioimmunoassay. This trial was designed to provide information on three aspects of buprenorphine kinetics: (1) the kinetic parameters of a single intravenous dose of buprenorphine given under carefully controlled conditions, i.e., under general anesthesia; (2) a comparison of intravenous kinetics under general anesthesia
Received for publication Feb. 20, 1980. Accepted for publication March 25, 1980. Reprint requests to: R. E. S. Bullingham, Nuffield Department of Anaesthetics, Radcliffe Infirmary, Oxford OX2 6HE, England. *M. R. D. Bennett was funded by Reckitt and Colman Pharmaceutical.
with intravenous kinetics in the awake postoperative patient; (3) mobilization data for intramuscular buprenorphine in the awake postoperative patient. Materials and methods
Drugs. 3H-Buprenorphine with a specific activity of 25 Ci/mM was produced by a Custom synthesis in the Radiochemical Centre, and was a gift from Reckitt and Colman. The buprenorphine (Temgesic) given to patients was that marketed in Britain. Subjects. The subjects were 24 fit patients undergoing elective total hip replacement at the Nuffield Orthopedic Centre, Oxford. Patients were selected sequentially if they were less than 75 yr old and were excluded if taking medication other than mild oral analgesics or thiazide diuretics. In all cases, preoperative electrocardiogram (ECG), hemoglobin, and plasma biochemistry were normal.
0009-9236/801110667+06$00.60/0 C) 1980 The C. V. Mosby Co.
667
668
Clin. Pharmacol. Ther. November 1980
Bullingham et al.
Table I. Data of patient groups Phase II, intravenous
Phase I Number of patients Age (yr ± SEM) Sex distribution (male: female) Weight (kg ± SEM) Intraoperative Pa((,,, (kPa ± SEM) Surgery start to reversal time (min -± SEM) Intraoperative blood loss (ml ± SEM)
24
10
64.5 ± 1.6
Phase II, intramuscular 11
67.5 ± 6.5
62.1 ± 2.6
14: 10
6:4
7:4
67.7 ± 2.4 5.4 -± 0.15
67.5 ± 2.1 5.35 -± 0.16
66.5 ± 3.6 5.5 ± 0.28
96.3 ± 5.8
95.3 ± 10.1
88.6 ± 6.5
487
-±
All patients received diazepam 10 mg orally preoperatively. Anesthesia was induced with intravenous thiopental 4 mg/kg, followed by pancuronium 0.1 mg/kg. After tracheal intubation, intermittent positive-pressure ventilation was instituted on the Bain circuit6 at a tidal volume of 10 ml/kg. The fresh gas flow of 70 ml/kg was chosen to achieve intraoperative normocapnia, and this was checked by arterial blood gas measurement. Fresh gas flow 2 to 4 hr
consisted of nitrous oxide and oxygen in a ratio of 2: 1, with 0.5% halothane. Halothane was used to avoid awareness during this light anesthesia and because quantitative information is available on its effect on hepatic blood flow.2' Radial artery cannulation was performed and then (phase I) 0.3 mg buprenorphine, diluted to 10 ml with 0.154 M NaC1, was injected intravenously over 1.0 min. All patients received this initial dose. Zero time was taken as the end of injection. Arterial samples were drawn intraoperatively and into the postoperative period, at 2, 5, 7.5, 10, 15, 20, 30, 40, 60, 80, 100, 120, 150, and 180 min. A basal venous blood sample was drawn before induction of anesthesia. All blood samples were taken into lithium heparin tubes, centrifuged, and the plasma separated at room temperature. Plasma was stored at 20° until required for analysis. Intraoperatively, all patients were kept horizontal on a warming blanket. Nasopharyngeal temperature, ECG, and direct arterial pressure were monitored throughout. Hartmann's solution was infused at 5 ml/kg/hr for 2 hr and then blood was transfused according to a standard formula. Intraoperative blood loss was moni-
53.6
539
-±
94.0
394 ± 58.8
tored by weighing of swabs and calibrated suction. No supplements of muscle relaxant were given. Halothane was withdrawn about 10 min before reversal. Reversal was with intravenous neostigmine 2.5 mg and atropine 1.2 mg. Postoperatively, patients breathed spontaneously, using controlled oxygen therapy (28% Ventimask). Phase II started at 180 min. The patients were divided into two groups and a second dose of 0.3 mg buprenorphine was given. One group again received the dose intravenously. The second group was given the dose intramuscularly into the vastus lateralis on the side opposite to the hip replacement. Arterial samples were drawn after the second dose at 2, 5, 10, 15, 20, 40, 60, 80, 120, 150, and 180 min, and treated in the same way as the phase I samples. Analytical procedures. Buprenorphine was measured in whole heparinized plasma by a radioimmunoassay procedure' using specific antibodies raised to N-hemisuccinyl-nor-buprenorphine linked to bovine serum albumin in rabbits. Standard displacement curves were prepared by measuring the displacement of 3H-buprenorphine binding to antiserum by unlabeled buprenorphine in drug-free human plasma. Samples with high values were diluted with drug-free plasma, and those with low values reassayed in a system where the volume of plasma samples was increased from 100 to 300 ill. Nonspecific binding of plasma to 3H-buprenorphine in the absence of antiserum was less than 5%. The precision of the assay was excellent, and the intra-assay and interassay
Buprenorphine kinetics
Volume 28
Number 5
Table II. Buprenorphine plasma levels Sample time (min) 2 5
7.5 10 15
20 30 40 60 80 100 120 150 180 *n
Phase
(X
± SD) (nglml)
1
(n = 24) 18.62 ± 5.60* 9.37 ± 2.99 6.03 ± 1.60 4.60 ± 1.09
3.12 2.30
-± -±
1.59 ± 1.13 ±
0.85 0.67 0.53 0.49 0.46 0.37
-±
± ± ± ± ±
669
0.85 0.56 0.40 0.32 0.18 0.24 0.16 0.15 0.16 0.17
Phase 11, intravenous (n = 10)
Phase II, intromuscular (n = 11)
6.28t 14.62 5.23 ± 1.67
3.61
3.15 ± 0.92 2.15 ± 0.56 1.77 -± 0.43
3.31 2.46 2.19
1.05 ± 0.27 0.77 ±- 0.17 0.70 ± 0.16
1.22 ± 0.76 0.79 -± 0.49 0.64 ± 0.36
0.47 ± 0.17 0.42 ± 0.18 0.15 0.38
0.49 ± 0.27 0.43 ± 0.24 0.36 ± 0.16
3.56 ± 2.80-t ± 2.97
± 2.52 ± 1.60 1.26
= 23.
tn = 9. in = 8.
variation at various plasma levels was less than 5%. In these short-term experiments, metabolite cross-reaction does not cause significant
problems.' Analysis of results. Plasma levels were fitted to a sum of exponentials curve, using a computer program based on a stripping method.9 The quality of the fit was judged by a minimum least-squares criterion. Before analysis, the computed terminal decay portion of the first dose of drug was removed from all plasma values measured after the second dose of drug. Justification for this came from the close exponential fits obtained with the first dose, showing behavior corresponding to that of a linear kinetic system. Kinetic parameters were calculated from the exponential fits." For calculation of percentage availability of the intramuscular dose, the area under the curve (AUC) from time 0 to 180 min in phase II was calculated using a numerical integration program.' The AUC from 180 min to infinity was estimated by calculation using an exponential fitted to the last six points of the curve for the intramuscular dose. The total AUC could then be compared with one similarly calculated for patients who received their second dose of drug intravenously. Since these were different patients, two different methods were used for the comparison. In one method,
the patient receiving the second dose intramuscularly was matched in terms of firstdose plasma levels of buprenorphine to all patients receiving the second dose intravenously. The best match was taken as that with minimum sum of squared differences, and then that patient's second-phase data was used in the comparison. In the second method, the averaged second-phase data for the intravenous group was used for comparison with each of the in-
tramuscular patients. Results
Three patients who completed the first phase subsequently did not complete the second phase because of arterial line failure. One patient who completed both phases subsequently did not complete the succeeding analgesia-measurement phase because of instrument failure. Plasma levels from these patients are included with those of the basic 20 who completed all phases. Patient data is shown in Table I. There was no significant difference between the groups with respect to age, weight, sex distribution, duration of anesthesia, or the total blood loss. Average buprenorphine plasma level data is given in Table II. The intravenous plasma data from individual patients fitted a multiexponential decay curve. In 27 of the total of 34 intrave-
Clin. Pharmacy'. Ther.
Bullingham et al.
670
November 1980
Table III. Average kinetic parameters (t ± SEM) Phase II, intravenous
Phase II, intramuscular
24
10
7
26.5 -± 3.2 2.1 -± 0.14 5.8 ± 0.57 11.2 ± 0.69 1.1 ± 0.08 139.6 ± 14.4 901.3 ± 39.7 97.3 ± 7.66 33.5 ± 2.98
18.1 ± 3.1 2.2 ± 0.29
Phase I (intraoperative) Number of patients A (ng/ml) 0/2,, (min) B (ng/ml) t1/20 (min)
C (ng/ml)
(min) Plasma clearance (ml/min) Vd (1) Initial C (ng/ml) V/25
Parameter values are for the triexponential fit: Plasma concentration (ng/ml) = Aexp pt + Cexp - yt. Other parameters are calculated from these fits.
Bexp
-
nous curves analyzed, the triexponential fit was better than the biexponential. In the remaining 7 patients, although a biexponential fit was better than a triexponential, the difference was small, and a triexponential analysis was used on their data. The quality of the fit obtained for each patient was assessed by the following method. A sum of squared differences was calculated for each data point between a given patient and the corresponding data points of every other patient, then the minimum value obtained from this match was compared with the sum of squares obtained for that patient from his best triexponential fit. By this criterion, for 75% of the patients the differences in plasma levels between patients were substantially greater than the differences between plasma level and computed fit within the patient. In the remaining patients, further analysis showed that the major source of error was caused by the 2-min sample point. The timing of the early samples always poses the greatest practical problems. Average kinetic parameters derived from individual triexponential fits are presented in Table III, as values in plasma concentraC exp at + B exp tion = A exp yt. Intramuscular uptake calculated by either method was very similar. Seven of eleven showed systemic availabilities of above 90%. The other four patients had poor availability of 40% to 60% of the dose. One of these had had previous hip surgery on the side of injection.
-
-
-
4.7 ± 0.79
2.8 ± 0.60 3.16 0.14 37.0 88.9 35.26 21.6 ± 3.54
16.5 ± 2.46 1.4 ± 0.26 138.5 ± 41.8
18.7 ± 0.72 ± 183.6 ± 1,275 ± 187.8 ±
992.7 ± 70.3 148.1 ± 51.3
- at + Bexp - pt + Cexp - yt, or biexponential fit:
Plasma levels were correspondingly low and irregular in this group. A notable feature of those patients with good drug mobilization was the rapidity with which the peak level occurred. The data then showed an excellent fit to a biexponential curve. Average kinetic parameters derived from these individual biexponential fits are also shown in Table III, in the form plasma concentration = B exp ot + C exp yt. Fig. 1 shows a plot of the average data of phase II for the intramuscular and intravenous groups as given in Table II. The closeness of the plasma levels for these two routes is evident.
-
-
Discussion The plasma level of intravenous buprenorphine declines very rapidly, with a t1/2 of 2.0 min, as expected for this very lipophilic drug. This is followed by a slow terminal phase with a t1/2 of 2 to 3 hr. In view of the 3-hr duration of the experiment, this t1/2 can only be considered to be an estimate. Plasma clearance under anesthesia is about 900 ml/min. The plasma/red blood cell ratio of buprenorphine is close to unity,* and so the blood clearance has a similar value. Buprenorphine is almost completely metabolized in vivo.t The clearance obtained is thus very close to the expected hepatic blood flow under these anesthetic conditions.2' '° Consequently, the *Moore RA: Unpublished observations. tRance MI: Personal communication.
Buprenorphine kinetics
Volume 28 Number 5
671
20
18
16
14 _c
12
0 10
0.
7
co co
te,
ro
4
0
20
40
80
60
100
120
140
160
180
Time (min)
Fig. 1. Plasma levels of buprenorphine in phase ll by intravenous (o----o) routes of administration.
extraction ratio is close to 1.0, and a very large first-pass effect is to be expected. Analysis of the plasma levels after a second dose, when a measurable plasma level is still present as a result of the first dose, requires that assumptions be made about the behavior of the drug. The assumption that the behavior is linear in this experiment is justified because of the good fit to a sum of exponential terms that was obtained. Removing the tail of the first dose from the total plasma level after the second dose will increase the total error in the phase II fit parameters. However, the plasma level of buprenorphine from the first dose at the time of administration of the second dose is small (about 0.4 ng/ml). Further, the long terminal t1/2 means that this value shows little absolute change during the time of measurement of the second dose. When the second dose is given intravenously with high initial plasma levels, the relative error is likely to be small. This is supported by the fits with data treated in this way, which were as good as those obtained for the first dose alone. The relative error will be greater with intramuscularly administered drug,
(
e) and intramuscular
especially if poor absorption should lead to little rise in plasma level. Comparison of the results from the same patient, anesthetized and awake postoperatively, shows two interesting differences. Plasma levels awake were all lower (p < 0.025, paired t test) than the anesthetized values, up to 30 min. This arose mainly from higher initial plasma values in the anesthetized state, implying lower initial volumes of distribution when anesthetized. This presumably reflects the lowered cardiac output known to occur in anesthesia,'° even in the normocapnic situation achieved here. The second effect of anesthesia is to lower the clearance by a consistent 30% with respect to the awake patient. As shown, the drug is almost completely cleared by the liver, and this figure agrees well with the measured effect of halothane anesthesia on hepatic blood flow .2. 7 The reported effects of anesthesia on meperidine kinetics8 are in the same direction and of the same magnitude as reported here. We believe sufficient attention has not been paid to the effects of anesthesia on drug kinetics. There will be a predictable difference in the behavior of
672
Clin. Pharmacol. Ther.
Bullingham et al.
drugs used in the anesthetized and awake conditions, and failure to recognize this may lead to important clinical differences. A notable feature of buprenorphine given intramuscularly is its very rapid uptake. Peak plasma level was usually at 5 min after the dose, and in some patients had occurred by 2 min. Mean plasma levels of intramuscular and intravenous buprenorphine differ little beyond 5 min, although the variance is larger at a given level. Systemic availability is generally close to 100%. Poor drug uptake presumably occurs with misplacement of the drug, an inherent risk with this route of administration. The clinical effects of these doses in this trial will be reported elsewhere. As anticipated,4 the effect of the drug far outlasts the plasma level and there is no direct relationship between plasma level and pharmacologic effect. Furthermore, poor systemic availability by the intramuscular route does not imply correspondingly poor analgesia. We thank the surgeons of the Nuffield Orthopedic Centre for allowing us to study their patients, and the Anesthetic and Recovery staff for their help. Reckitt and Colman donated the drug and assay materials; we gratefully acknowledge their support.
References JG, Rance MJ, Flockhart IR, Dockray G, Bennett MRD, Moore RA: The radioimmunoassay of buprenorphine. Eur J Clin Pharmacol. (In press.)
1. Bartlett AJ, Lloyd-Jones
November 1980
Epstein RM, Deutsch S, Cooperman LH, Clement AJ, Price HL: Splanchnic circulation during halothane anaesthesia and hypercapnia in normal man. Anesthesiology 27:654-661, 1966. Greville TNE: Theory and application of spline functions. London, 1969, Academic Press, Inc., pp. 156-167.
Hambrook JM, Rance MJ: The interaction of buprenorphine with the opiate receptor: Lipophilicity as a determining factor in drug-receptor kinetics, in Opiates and endogenous opioid peptides. Amsterdam, 1976, Elsevier/North Holland Biomedical Press, pp. 295-301. Heel RC, Brogden RN, Speight TM, Avery GS: Buprenorphine: A new strong analgesic. Curr Ther 5:29-33, 1979. Henville JD, Adams AP: The Bain anaesthetic system. Anaesthesia 31:247-256, 1976. Juhl B, Einer-Jensen N: Hepatic blood flow and cardiac output during halothane anaesthesia: An animal study. Acta Anaesthiol Scand 18:114122, 1974.
Mather LE, Tucker GT, Pflug AE, Lindop MJ, Wilkerson C: Meperidine kinetics in man. CLIN PHARMACOL THER 17:21-30, 1975. Perl W: A method for curve fitting by exponential functions. Int J Appl Radiat Isot 8:211-222, 1960.
Prys-Roberts C, Kelman GR, Greenbaum R, Robinson RH: Circulatory influences of artificial ventilation during nitrous oxide anaesthesia in man. Br J Anaesth 39:533-548, 1967. Wagner JG: Linear pharmacokinetic equations allowing direct calculation of many needed pharmacokinetic parameters from the coefficients and exponents of polyexponential equations which have been fitted to the data. J Pharmacokinet Biopharm 4:443-467, 1976.
