Br.J. Anaesth. (1978), 50, 1237
PHARMACOKINETICS OF THIOPENTONE: EFFECTS OF ENFLURANE AND NITROUS OXIDE ANAESTHESIA AND SURGERY M. M. GHONEIM AND M. J. VAN HAMME SUMMARY
A complete pharmacokinetic study of the drug in man, using the classical compartmental analysis, has not been reported. Understanding the pharmacokinetics of a drug is important for its practical use (Greenblatt and Koch-Weser, 1975). Previous attempts at studying the pharmacokinetics of thiopentone in man suffered from the lack of a sensitive analytical method, insufficient duration of sampling, and the use of doses greater than those commonly employed. We have developed a sensitive and reproducible assay for thiopentone in plasma (Van Hamme and Ghoneim, 1978); together with an adequate sampling design this allowed us to assess the kinetics of the drug in subjects receiving a single i.v. dose. We studied also the effects of nitrous oxide and enflurane (Ethrane) anaesthesia and surgery on the pharmacokinetic profile of the drug. METHODS
Subjects
Six patients undergoing eye, ear or oral surgery and six volunteers were studied after their informed consent had been obtained. All were healthy, apart from the disease requiring surgery in the patient group, and they had received no regular medication. The sex, age, body weight and duration of anaesthesia are shown in table I. The patients received diazepam
(Valium) 10 mg and hyoscine 0.4 mg injected i.m. 30 min before surgery, but the volunteers received no premedication. A venous cannula was inserted to one arm through which a solution of dextrose 5% in lactated Ringer's solution was administered during the study. Thiopentone 3.5 mg kg" 1 (table I) was then injected i.v. over a 30-s period to the other arm. In the patient group, anaesthesia was maintained with enflurane (Ethrane) in nitrous oxide in oxygen. The system inflow concentration of enflurane administered from a copper kettle vaporizer and through a semiclosed circuit was 1-2% (v/v). In some patients, suxamethonium 40-60 mg was given i.v. to facilitate tracheal intubation. Sampling design
Ten millilitre of venous blood was taken into heparinized glass syringes before and 1,2,4,8,15,30 and 60 min after injection, then hourly for 6 h; thereafter samples were taken every 2 h for 12 h. The blood samples were transferred immediately to 15-ml glass-stoppered centrifuge tubes containing heparin. The plasma was separated by centrifugation at 2000 rev min" 1 for 15 min and was frozen and stored at —15 °C until analysed within 3 weeks of collection. Analytical procedure
M.
M.
GHONEIM, M.B., B.CH., F.F.A.R.C.S. ; M.
J. VAN
HAMME, PH.D. ; Department of Anaesthesia, University of Iowa College of Medicine, Iowa City, Iowa 52242, U.S.A. 0007-0912/78/0050-1237 $01.00
Thiopentone was extracted from plasma and analysed by flame and ionization chromatography, with butabarbitone as internal standard (Van Hamme © Macmillan Journals Ltd 1978
Downloaded from http://bja.oxfordjournals.org/ at University of Birmingham on August 25, 2015
Plasma concentrations of thiopentone following the injection of 3.5 mg kg" 1 i.v. were studied in six patients who received enflurane and nitrous oxide anaesthesia and six volunteers. We identified a three-compartment open model system containing both a "shallow" and a "deep" peripheral compartment in all the patients and half of the controls, and a two-compartment open model for the remaining volunteers. The distribution of thiopentone to the tissues was very rapid, the a and (3 distribution half-lives averaging 2.5 min and 46.4 min respectively for the patient group and 2.8 min and 48.7 min for the control group. The wide distribution of the drug was indicated by the apparent volume of distribution, the means of which varied between 2 and 1.5 times body weight. The mean elimination half-life was 5.1 h for the patients and 5.7 h for the volunteers. The return of the drug to the central compartment from the deep peripheral compartment was the rate-controlling factor in its elimination. Neither enflurane and nitrous oxide anaesthesia nor the stress of surgery affected the distribution or clearance of the drug from plasma.
BRITISH JOURNAL OF ANAESTHESIA
1238 TABLE I.
Age (yr) Group Anaesthetized patients Control volunteers
Subject data Thiopentone (mg)* (mean + SD)
Sex
(mean±SD)
Weight (kg) (mean + SD)
3M 3F 3M 3F
25±9
67.4 ±16.1
235.4 ±56.1
25±5
65.6 ±15.9
231.4 ±54.8
Duration of anaesthesia (h) (mean±SD) 2.89 ±1.81
* Individual plasma concentrations of thiopentone will be supplied by the authors upon request.
Statistical and kinetic analyses
Student's t test was used to evaluate the difference between the volunteer and the patient groups. A probability value less than 0.05 was considered statistically significant. Plasma concentration data were analysed with the aid of an IBM 360/65 computer, and a program for non-linear least squares regression analysis (Metzler, 1969). Data from each individual subject were fitted sequentially to a monoexponential and then to successive polyexponential equations of the general form
CD(0 = Ajt-W+A&-** ... Ap-X>1 consistent with a one, two, three or four-compartment open mamillary model (Rescigno and Segre, 1966). The exactness of fit was determined in each case by calculation of the squared correlation coefficient. The acceptance of a particular equation as the "best" fit was based upon the incremental percentage improvement in the squared correlation coefficient reaching a pre-selected criterion (Sedman and Wagner, 1974). In the equation, the quantity C p represents plasma concentration of thiopentone and t represents the time after injection; i is the number of exponential terms required for description of the plasma concentration decay curve and the number of compartments in the corresponding kinetic model. The coefficients A 1J2 i are ordinate axis intercepts, and the exponents A12 i are rate constants for each exponential phase. These are "hybrid" quantities
influenced by all of the individual processes involved in the disposition of the drug (Gibaldi and Perrier, 1975). Half-lives of the disappearance phases, volume of the central compartment (Fc) total apparent volume of distribution by the area method (Fd area), total plasma clearance (C/p), first-order rate constants for drug transfer between compartments (k12, k2v k13 and k31) and the elimination rate constant k10 were calculated from the coefficients and exponents as described by Gibaldi and Perrier (1975). The fraction of drug in the body which is in the central compartment (A3/&10) was computed also. Composite data were obtained from a computer fit of average plasma concentration data of the two groups of subjects. Calculations were then conducted as described to obtain rate constants and pharmacokinetic parameters for each group. RESULTS
The coefficient of variations obtained by extraction and analysis of standard curve plasma samples was 3.0% with a reproducibility of ±2.0%. Concentrations remained unchanged after storage for 6 weeks. Mean plasma concentrations of thiopentone are plotted in figure 1. The data from the patients fitted best a triexponential equation consistent with a threecompartment open pharmacokinetic model with firstorder drug elimination from the central compartment (fig. 2). The plasma concentration data of half the volunteer group fitted best a tri-exponential equation while the rest was expressed best bi-exponentially. The latter was consistent with a two-compartment open kinetic model (fig. 2). Kinetic parameters for the two groups are shown in table II. Neither the plasma concentrations nor the kinetic parameters were significantly different between the two groups. DISCUSSION
The pharmacokinetic profile of thiopentone is described best in the entire patient group and half of the volunteer group by a tri-exponential equation,
Downloaded from http://bja.oxfordjournals.org/ at University of Birmingham on August 25, 2015
and Ghoneim, 1978). A Bendix 2500 Research Gas Chromatograph containing a 183 mm x 4 mm (i.d.) x 0.635 mm (o.d.) siliconized glass column packed with 5% OV-1 on 100/120 mesh HP Chromosorb W provided desired barbiturate separations when operated at 205 °C, with the injection port and detector at 230 °C. The detector provided an optimum response when all gases were regulated at 207 kPa at the followingflowrates: hydrogen 40 ml min" 1 ; air 425 ml m k r 1 and nitrogen 45 ml min" 1 . All samples were analysed in duplicate.
1239
PHARMACOKINETICS OF THIOPENTONE TABLE II. Kinetics of thiopentone in anaesthetized patients and in control volunteer subjects* Anaesthetized patients f
^(Hgml- 1 ) ^(ngml-1) ^(Hgml- 1 ) Axtfi- 1 )
A^h- 1 ) A3 0 0
*2i(h"1) *3i(h-)
rjA^h) TjA2(h) TjA3(h) *./*iot Kc (litre kg-1) Fd area (litre kg"1) C/p (litre kg- 1 h- 1 )
Mean+SEM
Group composite §
Mean±SEM
Group composite §
21.12 ±6.50 3.81 ±0.61 1.34 + 0.27 28.29 ±6.53 0.99 ±0.15 0.14 ±0.01
7.52 3.92 1.51 16.44 0.90 0.12
9.90 + 5.13 2.53 + 0.43 1.66 + 0.33 9.69 + 4.09 0.55 ±0.18 0.12±0.01
6.93 3.30 1.21 14.81 0.85 0.10
1.57 ±0.28 24.92 + 7.64 10.76 ±3.91 2.91 + 1.36 0.56 ±0.23
0.75 8.26 7.29 0.83 0.33
0.85 + 0.32 3.25 + 1.62 2.98 ±1.07 1.27 ±0.51 0.37 ±0.06
0.71 7.63 6.25 0.89 0.29
0.04 + 0.01 0.78±0.11 5.14 ±0.30 0.11+0.02 0.17 + 0.04 1.88 + 0.38 0.26 + 0.06
0.04 0.77 5.68 0.16 0.25 1.53 0.19
0.22 + 0.10 2.69 ±0.89 5.74 + 0.78 0.27 ±0.08 0.44 ±0.13 1.61 ±0.11 0.22 ±0.02
0.05 0.81 6.73 0.14 0.28 1.96 0.20
* P>0.05 for all the parameters; f Six subjects in each group; § Data were obtained from a computerized best fit of the average group plasma concentrations; % n = 2 for a two-compartment model, n = 3 for a three-compartment model.
k
A
: 12
1
2 o Volunteer average o Patient average
k.
i
21 10
B
k 13
*12
3
1
2 21
' 1' 10
FIG. 1. Average plasma concentration of thiopentone in 12 subjects after a dose of 3.5 mg kg"1 i.v. The three decay lines were fitted by computer using non-linear least squares regression analysis.
FIG. 2. (A) TWO- and (B) three-compartment open pharmacokinetic models. Compartment "1" is the central compartment, and "2" and "3" are peripheral compartments. In (B), "2" is the shallow peripheral compartment and "3" is the deep peripheral compartment.
while the other members of the volunteer group exhibit two-compartment kinetics. Two-compartmental and three-compartmental kinetic patterns have been observed in studies of different subjects receiving the same drug (Greenblatt et al., 1975). According to a three-compartment open model, the
drug is administered into a central compartment, followed by its distribution into two peripheral compartments, one of which is rapidly accessible, while the other is only slowly accessible. Elimination occurs only from the central compartment which
Downloaded from http://bja.oxfordjournals.org/ at University of Birmingham on August 25, 2015
^loCh"1)
Control volunteers j"
1240
drug within a central compartment and a shallow peripheral compartment which includes the richly perfused organs, and then to a deeper peripheral compartment. The drug is slowly released from the latter compartment and eliminated from the body. The elimination half-life ranges from 5.1 h for the patient group to 5.7 h for the control group. All potent anaesthetic drugs reduce the hepatic blood flow (Cooperman, 1972; Juhl and EinerJensen, 1976). They would be expected to reduce the plasma clearance of drugs with high hepatic extraction by reducing the delivery of the drug to the liver. As an example, the rate of elimination of lignocaine, which has a high hepatic extraction, was substantially prolonged in patients undergoing laparoscopy under general anaesthetic (Ti= 2.8 h) as compared with normal conscious volunteers (Tj = 1.3 h) (AdjeponYamoach, Scott and Prescott, 1973). Thiopentone has a low hepatic extraction. Mark and others (1965), measured the drug extraction directly in man, although their data were obtained primarily from patients with cirrhosis. In six subjects the fraction of thiopentone removed from hepatic blood flow varied from 10% to 50%, with a mean of 28%. In another five subjects, the extraction was zero. One subject with a normal liver had a trans-hepatic arterio-venous difference in thiopentone concentration of 32-35%. These values were obtained minutes to hours following injection of the drug and it is possible that they may be even less if the samples are taken earlier. It has been shown that a decrease in hepatic blood flow will have a significant effect on the half-life of a drug only if the hepatic extraction ratio of the drug is high (Swartz, Sidell and Cucinell, 1974). Inhalation anaesthetic agents compete for active sites in or about cytochrome P-450 with barbiturates (Brown, 1971). Acute stress, such as that of surgery, inhibits the metabolism of the ultra-short acting barbiturate hexobarbitone because of increased release of corticosterone (Chung and Brown, 1976). However, halothane (Rahn, Dayton and Frederickson, 1969) enflurane anaesthesia and the stress of surgery seem not to affect the clearance of the drug. Novelli, Marsili and Lovenzi (1975) found that injection of thiopentone to the portal vein or pretreatment with enzyme inducers or inhibitors did not modify the effects of the drug. It is possible that, with a low metabolic clearance of the drug, estimated to be 10-15% per hour (Brodie et al., 1950), the effect of further slowing of metabolism may not be readily apparent. It should be remembered that the effect of
Downloaded from http://bja.oxfordjournals.org/ at University of Birmingham on August 25, 2015
contains the organs of metabolism and excretion. The two-compartment open model is simpler and assumes a small central compartment and a larger peripheral one. These compartments do not necessarily correspond to specific anatomical entities, although for many drugs the central compartment consists of the vascular space, with or without the richly perfused tissues such as the brain and heart. The peripheral compartment or compartments are usually formed by less perfused tissues, such as muscle, skin and fat. The distribution of thiopentone into tissues is very rapid, the a and fl distribution half-lives averaging 2.8 min and 48.7 min, respectively, for the volunteer group and 2.5 min and 46.4 min for the patient group. The high lipid solubility of the drug is mainly responsible for this effect. The wide distribution of the drug is indicated by the apparent volume of distribution of the drug, the means of which varied between 2 and 1.5 times body weight. Even these figures underestimate the extent of distribution of the drug, because they were calculated from the total thiopentone plasma concentration. The plasma protein binding of the drug may be as great as 86% in the therapeutic range (Becker, 1976) and only the unbound fraction can leave the blood stream and reach the different tissues. The rate of transfer of a drug from one compartment to another is associated with the rate constant, K. There are one or more rate constants associated with the transfer process. In general these rate processes are first order. The transfer of thiopentone between the central and peripheral compartments is described by k12> k2V k13 and k31 (fig. 2). The mean ratio ofk3jk13 of 0.19 for the patient group and 0.29 for the control group reflecting drug movement between the central and deep peripheral compartment, suggests slow equilibration between these compartments. On the other hand, the shallow compartment equilibrates more quickly as reflected by two to three times greater k21/12 ratios. The elimination rate constant k10 is, on average, twice k^, the rate of return of drug from the deep peripheral compartment. The mean ratio V^io °f °- 0 9 f° r th e patient group and 0.15 for the control group indicates that only 9-15% of thiopentone in the body is in the central compartment, available for elimination at any time. These findings suggest that the return of the drug to the central compartment from the deep peripheral compartment is rate-controlling in the elimination of thiopentone. In summary, the kinetic profile of thiopentone is charaaerized by an initial rapid distribution of the
BRITISH JOURNAL OF ANAESTHESIA
1241
PHARMACOKINETICS OF THIOPENTONE general anaesthesia and surgery on the-kinetics of concomitantly administered drugs will vary, at least in part, because of the drugs themselves.
PHARMACOCINETIQUE DU THIOPENTONE: EFFETS DE L'ANESTHESIE PAR L'ENFLURANE ET LE PROTOXYDE D'AZOTE ET DE L'INTERVENTION CHIRURGICALE RESUME
On a etudie les concentrations de thiopentone dans le plasma sur six malades anesthesias a l'aide d'enflurane et de protoxyde d'azote ainsique sur six volontaires, apres une injection intraveineuse a raison de 3,5 mg kg"1. Nous avons identifie un systeme a modele ouvert a trois compartiments contenant a la fois un compartiment peripWrique "peu profond" et un autre "profond" sur tous les malades et sur la moitie des ttaioins, et un modele ouvert a deux compartiments sur les autres volontaires. La repartition du thiopentone dans tous les tissus a ete tres rapide, la repartition des demi-vies a et p etant en moyenne de 2,5 min et de 46,4 min respectivement pour le groupe de malades et de 2,8 min et de 48,7 min pour le groupe temoin. La vaste repartition du medicament a ete indiquee par le volume apparent de la repartition, dont les moyennes ont varie entre 2 fois et 1,5 fois suivant le poids du corps. L'eiimination moyenne de la demi-vie a ete de 5,1 h pour les malades et de 5,7 h pour les volontaires. Le retour du medicament au compartiment central, en provenance du compartiment peripherique profond, a ete le facteur controlant le taux pendant son elimination. Ni l'anesthesie par l'enflurane et le protoxyde d'azote, ni le stress chirurgical n'ont affecte la repartition ou l'eiimination du medicament se trouvant dans le plasma. PHARMAKOKINETIK VON THIOPENTON: WIRKUNGEN VON ENFLURAN UND SALPETRIGER OXYDANASTHESIE UND CHIRURGIE ZUSAMMENFASSUNG
Untersucht wurden die Plasmakonzentrationen von Thiopenton in sechs Patienten, die nach einer intravenOsen Injektion von 3,5 mg kg- 1 eine Enfluran und salpetrige Oxydanasthesie bekamen, und in sechs Testpersonen. Wir identifizierten in alien sechs Patienten und in der Halfte der freiwilligen Kontrollgruppe das System als ein dreiteiliges, offenes Modell mit einem "flachen" und einem "tiefen" peripheralen Teil, und in den restlichen Testpersonen als ein zweiteiliges, offenes Modell. Die Verteilung von Thiopenton im Gewebe ging sehr rasch, die Alpha- und Beta- Verteilungshalbwertzeiten waren durchschnittlich je 2,5 und 46,4 Minuten fur die Patientengruppe, und je 2,8 und 48,7 Minuten fiir die Kontrollgruppe. Die grosse Verbreitung der Droge wurde von dem augenscheinlichen Volumen der Verteilung angezeigt, dessen Mittelwerte zwischen 2- und 1,5-mal KSrpergewicht wechselten. Die durchschnittlichen Ausscheidungshalbwertzeiten waren 5,1 Stunden fiir die Patientengruppe und 5,7 Stunden fiir die Kontrollgruppe. Die Ruckkehr der Droge vom tiefen, peripheralen Teil zum zentralen Teil war der geschwindigkeitskontrollierende Umstand der Ausscheidung. Weder Enfluran und salpetrige Oxydanasthesie noch
Downloaded from http://bja.oxfordjournals.org/ at University of Birmingham on August 25, 2015
Adjepon-Yamoach, K. K., Scott, D. B., and Prescott, L. F. (1973). Impaired absorption and metabolism of oral lignocaine in patients undergoing laparoscopy. Br. J. Anaesth., 45,143. Becker, K. E., jr (1976). Gas chromatographic assay for free and total plasma levels of thiopental. Anesthesiology, 45, 656. Brodie, B. B., Mark, L. C , Papper, E. M., Lief, P. A., Bernstein, E., and Rovenstine, E. A. (1950). The fate of thiopental in man and a method for its estimation in biological material. J. Pharmacol. Exp. Ther., 98,85. Brown, B. R., jr (1971). The diphasic action of halothane on the oxidative metabolism of drugs by the liver: an in vitro study in the rat. Anesthesiology, 35, 241. Chung, H., and Brown, D. R. (1976). The mechanism of the effect of acute stress on hexobarbital metabolism. Toxicol. Appl. Pharmacol, 37, 313. Cooperman, L. H. (1972). Effects of anaesthetics on the splanchnic circulation. Br. J. Anaesth., 44,967. Gibaldi, M., and Perrier, D. (1975). Multi Compartments Models in Pharmacokinetics, 1st edn, p. 45. New York: Marcel Dekker, Inc. Greenblatt, D. J., and Koch-Weser, J. (1975). Clinical pharmacokinetics. N. Engl.J. Med., 293, 702. Shader, R. I., Franke, K., and Koch-Weser, J. (1975). Pharmacokinetics of intravenous chlordiazepoxide. Clin. Pharmacol. Ther., 17, 235. Juhl, B., and Einer-Jensen, N. (1976). Hepatic blood flow and cardiac output during Fluoromar anaesthesia. An animal study. Acta Anaesthesiol. Scand., 20, 271. Mark, L. C , Brand, L., Kamvyssi, S., Britton, R. C , Perel, j ; M., Landrau, M. A., and Dayton, P. G. (1965). Thiopental metabolism by human liver in vivo and in vitro. Nature {Lond.), 206,1117. Metzler, C. M. (1969). A User's Manual for NONLIN. Technical report 7292/69/7292/005. Kalamazoo, Michigan, U.S.A.: The Upjohn Co. Novelli, G. P., Marsili, M., and Lovenzi, P. (1975). Influence of liver metabolism on the actions of Althesin and thiopentone. Br.J. Anaesth., 47, 913. Rahn, E., Dayton, P. G., and Frederickson, E. L. (1969). Lack of effect of halothane on the metabolism of thiopentone in man. Br.J. Anaesth., 41, 503. Rescigno, A., and Segre, G. (1966). Drug and Tracer Kinetics, 1st edn, p. 91. London: Blaisdell Publishing Co. Sedman, A. J., and Wagner, J. G. (1974). AUTO AN: A Decision-Making Pharmacokinetic Computer Program. 615 East University Avenue, Ann Arbor, Michigan, U.S.A.: Publication Distribution Services. Swartz, R. D., Sidell, F. R., and Cucinell, S. A. (1974). Effects of physical stress on the disposition of drugs eliminated by the liver in man. J. Pharmacol. Exp. Ther., 188,1. 101
Van Hamme, M. J., and Ghoneim, M. M. (1978). A sensitive gas chromatographic assay for thiopentone in plasma. Br.J. Anaesth., 50,143.
BRITISH JOURNAL OF ANAESTHESIA
1242 der Stress der Chirurgie hatten Einfluss auf Verteilung oder Ausscheidung der Droge aus dem Plasma. FARMACOCINETICA DE TIOPENTONA: EFECTOS EJERCIDOS POR ANESTESIA DE ENFLURANO Y OXIDO NITROSO Y POR CIRUGIA SUMARIO
Downloaded from http://bja.oxfordjournals.org/ at University of Birmingham on August 25, 2015
Se estudiaron las concentraciones de tiopentona en la plasma de seis pacientes y de seis voluntaries que recibieron anestesia de enflurano y oxido nitroso tras la inyeccion intravenosa de 3,5 mg kg"1. Identificamos un sistema de modelo abierto de tres compartimentos que contenia tanto un compartimento periferico "poco profundo" como uno
"profundo" en todos los pacientes y en la mitad de los controles, y un modelo abierto de dos compartimentos en los voluntarios restantes. La distribucion de tiopentona por los tejidos fue rapida, promediendo los periodos de media vida de distribucion a y (3 2,5 y 46,4 min respectivamente en el grupo de los pacientes y 2,8 min y 48,7 min en el grupo de control. La extensa distribucion de la droga fue indicada por el volumen de distribuci6n aparente, cuyo promedio varid entre 2 y 1,5 veces el peso del cuerpo. El promedio de media vida de eliminaci6n fue de 5,1 h para los pacientes y 5,7 h para los voluntarios. El retorno de la droga al compartimento central desde el compartimento periferico profundo fue el factor que controld la rapidez de su eliminacion. Ni la anestesia de enflurano y 6xido nitroso, ni el "stress" causado por la cirugia afectaron la distribucidn o eliminacidn de la droga de la plasma.
PHARMACOKINETICS OF THIOPENTONE: EFFECTS OF ENFLURANE AND NITROUS OXIDE ANAESTHESIA AND SURGERY M. M. GHONEIM AND M. J. VAN HAMME SUMMARY
A complete pharmacokinetic study of the drug in man, using the classical compartmental analysis, has not been reported. Understanding the pharmacokinetics of a drug is important for its practical use (Greenblatt and Koch-Weser, 1975). Previous attempts at studying the pharmacokinetics of thiopentone in man suffered from the lack of a sensitive analytical method, insufficient duration of sampling, and the use of doses greater than those commonly employed. We have developed a sensitive and reproducible assay for thiopentone in plasma (Van Hamme and Ghoneim, 1978); together with an adequate sampling design this allowed us to assess the kinetics of the drug in subjects receiving a single i.v. dose. We studied also the effects of nitrous oxide and enflurane (Ethrane) anaesthesia and surgery on the pharmacokinetic profile of the drug. METHODS
Subjects
Six patients undergoing eye, ear or oral surgery and six volunteers were studied after their informed consent had been obtained. All were healthy, apart from the disease requiring surgery in the patient group, and they had received no regular medication. The sex, age, body weight and duration of anaesthesia are shown in table I. The patients received diazepam
(Valium) 10 mg and hyoscine 0.4 mg injected i.m. 30 min before surgery, but the volunteers received no premedication. A venous cannula was inserted to one arm through which a solution of dextrose 5% in lactated Ringer's solution was administered during the study. Thiopentone 3.5 mg kg" 1 (table I) was then injected i.v. over a 30-s period to the other arm. In the patient group, anaesthesia was maintained with enflurane (Ethrane) in nitrous oxide in oxygen. The system inflow concentration of enflurane administered from a copper kettle vaporizer and through a semiclosed circuit was 1-2% (v/v). In some patients, suxamethonium 40-60 mg was given i.v. to facilitate tracheal intubation. Sampling design
Ten millilitre of venous blood was taken into heparinized glass syringes before and 1,2,4,8,15,30 and 60 min after injection, then hourly for 6 h; thereafter samples were taken every 2 h for 12 h. The blood samples were transferred immediately to 15-ml glass-stoppered centrifuge tubes containing heparin. The plasma was separated by centrifugation at 2000 rev min" 1 for 15 min and was frozen and stored at —15 °C until analysed within 3 weeks of collection. Analytical procedure
M.
M.
GHONEIM, M.B., B.CH., F.F.A.R.C.S. ; M.
J. VAN
HAMME, PH.D. ; Department of Anaesthesia, University of Iowa College of Medicine, Iowa City, Iowa 52242, U.S.A. 0007-0912/78/0050-1237 $01.00
Thiopentone was extracted from plasma and analysed by flame and ionization chromatography, with butabarbitone as internal standard (Van Hamme © Macmillan Journals Ltd 1978
Downloaded from http://bja.oxfordjournals.org/ at University of Birmingham on August 25, 2015
Plasma concentrations of thiopentone following the injection of 3.5 mg kg" 1 i.v. were studied in six patients who received enflurane and nitrous oxide anaesthesia and six volunteers. We identified a three-compartment open model system containing both a "shallow" and a "deep" peripheral compartment in all the patients and half of the controls, and a two-compartment open model for the remaining volunteers. The distribution of thiopentone to the tissues was very rapid, the a and (3 distribution half-lives averaging 2.5 min and 46.4 min respectively for the patient group and 2.8 min and 48.7 min for the control group. The wide distribution of the drug was indicated by the apparent volume of distribution, the means of which varied between 2 and 1.5 times body weight. The mean elimination half-life was 5.1 h for the patients and 5.7 h for the volunteers. The return of the drug to the central compartment from the deep peripheral compartment was the rate-controlling factor in its elimination. Neither enflurane and nitrous oxide anaesthesia nor the stress of surgery affected the distribution or clearance of the drug from plasma.
BRITISH JOURNAL OF ANAESTHESIA
1238 TABLE I.
Age (yr) Group Anaesthetized patients Control volunteers
Subject data Thiopentone (mg)* (mean + SD)
Sex
(mean±SD)
Weight (kg) (mean + SD)
3M 3F 3M 3F
25±9
67.4 ±16.1
235.4 ±56.1
25±5
65.6 ±15.9
231.4 ±54.8
Duration of anaesthesia (h) (mean±SD) 2.89 ±1.81
* Individual plasma concentrations of thiopentone will be supplied by the authors upon request.
Statistical and kinetic analyses
Student's t test was used to evaluate the difference between the volunteer and the patient groups. A probability value less than 0.05 was considered statistically significant. Plasma concentration data were analysed with the aid of an IBM 360/65 computer, and a program for non-linear least squares regression analysis (Metzler, 1969). Data from each individual subject were fitted sequentially to a monoexponential and then to successive polyexponential equations of the general form
CD(0 = Ajt-W+A&-** ... Ap-X>1 consistent with a one, two, three or four-compartment open mamillary model (Rescigno and Segre, 1966). The exactness of fit was determined in each case by calculation of the squared correlation coefficient. The acceptance of a particular equation as the "best" fit was based upon the incremental percentage improvement in the squared correlation coefficient reaching a pre-selected criterion (Sedman and Wagner, 1974). In the equation, the quantity C p represents plasma concentration of thiopentone and t represents the time after injection; i is the number of exponential terms required for description of the plasma concentration decay curve and the number of compartments in the corresponding kinetic model. The coefficients A 1J2 i are ordinate axis intercepts, and the exponents A12 i are rate constants for each exponential phase. These are "hybrid" quantities
influenced by all of the individual processes involved in the disposition of the drug (Gibaldi and Perrier, 1975). Half-lives of the disappearance phases, volume of the central compartment (Fc) total apparent volume of distribution by the area method (Fd area), total plasma clearance (C/p), first-order rate constants for drug transfer between compartments (k12, k2v k13 and k31) and the elimination rate constant k10 were calculated from the coefficients and exponents as described by Gibaldi and Perrier (1975). The fraction of drug in the body which is in the central compartment (A3/&10) was computed also. Composite data were obtained from a computer fit of average plasma concentration data of the two groups of subjects. Calculations were then conducted as described to obtain rate constants and pharmacokinetic parameters for each group. RESULTS
The coefficient of variations obtained by extraction and analysis of standard curve plasma samples was 3.0% with a reproducibility of ±2.0%. Concentrations remained unchanged after storage for 6 weeks. Mean plasma concentrations of thiopentone are plotted in figure 1. The data from the patients fitted best a triexponential equation consistent with a threecompartment open pharmacokinetic model with firstorder drug elimination from the central compartment (fig. 2). The plasma concentration data of half the volunteer group fitted best a tri-exponential equation while the rest was expressed best bi-exponentially. The latter was consistent with a two-compartment open kinetic model (fig. 2). Kinetic parameters for the two groups are shown in table II. Neither the plasma concentrations nor the kinetic parameters were significantly different between the two groups. DISCUSSION
The pharmacokinetic profile of thiopentone is described best in the entire patient group and half of the volunteer group by a tri-exponential equation,
Downloaded from http://bja.oxfordjournals.org/ at University of Birmingham on August 25, 2015
and Ghoneim, 1978). A Bendix 2500 Research Gas Chromatograph containing a 183 mm x 4 mm (i.d.) x 0.635 mm (o.d.) siliconized glass column packed with 5% OV-1 on 100/120 mesh HP Chromosorb W provided desired barbiturate separations when operated at 205 °C, with the injection port and detector at 230 °C. The detector provided an optimum response when all gases were regulated at 207 kPa at the followingflowrates: hydrogen 40 ml min" 1 ; air 425 ml m k r 1 and nitrogen 45 ml min" 1 . All samples were analysed in duplicate.
1239
PHARMACOKINETICS OF THIOPENTONE TABLE II. Kinetics of thiopentone in anaesthetized patients and in control volunteer subjects* Anaesthetized patients f
^(Hgml- 1 ) ^(ngml-1) ^(Hgml- 1 ) Axtfi- 1 )
A^h- 1 ) A3 0 0
*2i(h"1) *3i(h-)
rjA^h) TjA2(h) TjA3(h) *./*iot Kc (litre kg-1) Fd area (litre kg"1) C/p (litre kg- 1 h- 1 )
Mean+SEM
Group composite §
Mean±SEM
Group composite §
21.12 ±6.50 3.81 ±0.61 1.34 + 0.27 28.29 ±6.53 0.99 ±0.15 0.14 ±0.01
7.52 3.92 1.51 16.44 0.90 0.12
9.90 + 5.13 2.53 + 0.43 1.66 + 0.33 9.69 + 4.09 0.55 ±0.18 0.12±0.01
6.93 3.30 1.21 14.81 0.85 0.10
1.57 ±0.28 24.92 + 7.64 10.76 ±3.91 2.91 + 1.36 0.56 ±0.23
0.75 8.26 7.29 0.83 0.33
0.85 + 0.32 3.25 + 1.62 2.98 ±1.07 1.27 ±0.51 0.37 ±0.06
0.71 7.63 6.25 0.89 0.29
0.04 + 0.01 0.78±0.11 5.14 ±0.30 0.11+0.02 0.17 + 0.04 1.88 + 0.38 0.26 + 0.06
0.04 0.77 5.68 0.16 0.25 1.53 0.19
0.22 + 0.10 2.69 ±0.89 5.74 + 0.78 0.27 ±0.08 0.44 ±0.13 1.61 ±0.11 0.22 ±0.02
0.05 0.81 6.73 0.14 0.28 1.96 0.20
* P>0.05 for all the parameters; f Six subjects in each group; § Data were obtained from a computerized best fit of the average group plasma concentrations; % n = 2 for a two-compartment model, n = 3 for a three-compartment model.
k
A
: 12
1
2 o Volunteer average o Patient average
k.
i
21 10
B
k 13
*12
3
1
2 21
' 1' 10
FIG. 1. Average plasma concentration of thiopentone in 12 subjects after a dose of 3.5 mg kg"1 i.v. The three decay lines were fitted by computer using non-linear least squares regression analysis.
FIG. 2. (A) TWO- and (B) three-compartment open pharmacokinetic models. Compartment "1" is the central compartment, and "2" and "3" are peripheral compartments. In (B), "2" is the shallow peripheral compartment and "3" is the deep peripheral compartment.
while the other members of the volunteer group exhibit two-compartment kinetics. Two-compartmental and three-compartmental kinetic patterns have been observed in studies of different subjects receiving the same drug (Greenblatt et al., 1975). According to a three-compartment open model, the
drug is administered into a central compartment, followed by its distribution into two peripheral compartments, one of which is rapidly accessible, while the other is only slowly accessible. Elimination occurs only from the central compartment which
Downloaded from http://bja.oxfordjournals.org/ at University of Birmingham on August 25, 2015
^loCh"1)
Control volunteers j"
1240
drug within a central compartment and a shallow peripheral compartment which includes the richly perfused organs, and then to a deeper peripheral compartment. The drug is slowly released from the latter compartment and eliminated from the body. The elimination half-life ranges from 5.1 h for the patient group to 5.7 h for the control group. All potent anaesthetic drugs reduce the hepatic blood flow (Cooperman, 1972; Juhl and EinerJensen, 1976). They would be expected to reduce the plasma clearance of drugs with high hepatic extraction by reducing the delivery of the drug to the liver. As an example, the rate of elimination of lignocaine, which has a high hepatic extraction, was substantially prolonged in patients undergoing laparoscopy under general anaesthetic (Ti= 2.8 h) as compared with normal conscious volunteers (Tj = 1.3 h) (AdjeponYamoach, Scott and Prescott, 1973). Thiopentone has a low hepatic extraction. Mark and others (1965), measured the drug extraction directly in man, although their data were obtained primarily from patients with cirrhosis. In six subjects the fraction of thiopentone removed from hepatic blood flow varied from 10% to 50%, with a mean of 28%. In another five subjects, the extraction was zero. One subject with a normal liver had a trans-hepatic arterio-venous difference in thiopentone concentration of 32-35%. These values were obtained minutes to hours following injection of the drug and it is possible that they may be even less if the samples are taken earlier. It has been shown that a decrease in hepatic blood flow will have a significant effect on the half-life of a drug only if the hepatic extraction ratio of the drug is high (Swartz, Sidell and Cucinell, 1974). Inhalation anaesthetic agents compete for active sites in or about cytochrome P-450 with barbiturates (Brown, 1971). Acute stress, such as that of surgery, inhibits the metabolism of the ultra-short acting barbiturate hexobarbitone because of increased release of corticosterone (Chung and Brown, 1976). However, halothane (Rahn, Dayton and Frederickson, 1969) enflurane anaesthesia and the stress of surgery seem not to affect the clearance of the drug. Novelli, Marsili and Lovenzi (1975) found that injection of thiopentone to the portal vein or pretreatment with enzyme inducers or inhibitors did not modify the effects of the drug. It is possible that, with a low metabolic clearance of the drug, estimated to be 10-15% per hour (Brodie et al., 1950), the effect of further slowing of metabolism may not be readily apparent. It should be remembered that the effect of
Downloaded from http://bja.oxfordjournals.org/ at University of Birmingham on August 25, 2015
contains the organs of metabolism and excretion. The two-compartment open model is simpler and assumes a small central compartment and a larger peripheral one. These compartments do not necessarily correspond to specific anatomical entities, although for many drugs the central compartment consists of the vascular space, with or without the richly perfused tissues such as the brain and heart. The peripheral compartment or compartments are usually formed by less perfused tissues, such as muscle, skin and fat. The distribution of thiopentone into tissues is very rapid, the a and fl distribution half-lives averaging 2.8 min and 48.7 min, respectively, for the volunteer group and 2.5 min and 46.4 min for the patient group. The high lipid solubility of the drug is mainly responsible for this effect. The wide distribution of the drug is indicated by the apparent volume of distribution of the drug, the means of which varied between 2 and 1.5 times body weight. Even these figures underestimate the extent of distribution of the drug, because they were calculated from the total thiopentone plasma concentration. The plasma protein binding of the drug may be as great as 86% in the therapeutic range (Becker, 1976) and only the unbound fraction can leave the blood stream and reach the different tissues. The rate of transfer of a drug from one compartment to another is associated with the rate constant, K. There are one or more rate constants associated with the transfer process. In general these rate processes are first order. The transfer of thiopentone between the central and peripheral compartments is described by k12> k2V k13 and k31 (fig. 2). The mean ratio ofk3jk13 of 0.19 for the patient group and 0.29 for the control group reflecting drug movement between the central and deep peripheral compartment, suggests slow equilibration between these compartments. On the other hand, the shallow compartment equilibrates more quickly as reflected by two to three times greater k21/12 ratios. The elimination rate constant k10 is, on average, twice k^, the rate of return of drug from the deep peripheral compartment. The mean ratio V^io °f °- 0 9 f° r th e patient group and 0.15 for the control group indicates that only 9-15% of thiopentone in the body is in the central compartment, available for elimination at any time. These findings suggest that the return of the drug to the central compartment from the deep peripheral compartment is rate-controlling in the elimination of thiopentone. In summary, the kinetic profile of thiopentone is charaaerized by an initial rapid distribution of the
BRITISH JOURNAL OF ANAESTHESIA
1241
PHARMACOKINETICS OF THIOPENTONE general anaesthesia and surgery on the-kinetics of concomitantly administered drugs will vary, at least in part, because of the drugs themselves.
PHARMACOCINETIQUE DU THIOPENTONE: EFFETS DE L'ANESTHESIE PAR L'ENFLURANE ET LE PROTOXYDE D'AZOTE ET DE L'INTERVENTION CHIRURGICALE RESUME
On a etudie les concentrations de thiopentone dans le plasma sur six malades anesthesias a l'aide d'enflurane et de protoxyde d'azote ainsique sur six volontaires, apres une injection intraveineuse a raison de 3,5 mg kg"1. Nous avons identifie un systeme a modele ouvert a trois compartiments contenant a la fois un compartiment peripWrique "peu profond" et un autre "profond" sur tous les malades et sur la moitie des ttaioins, et un modele ouvert a deux compartiments sur les autres volontaires. La repartition du thiopentone dans tous les tissus a ete tres rapide, la repartition des demi-vies a et p etant en moyenne de 2,5 min et de 46,4 min respectivement pour le groupe de malades et de 2,8 min et de 48,7 min pour le groupe temoin. La vaste repartition du medicament a ete indiquee par le volume apparent de la repartition, dont les moyennes ont varie entre 2 fois et 1,5 fois suivant le poids du corps. L'eiimination moyenne de la demi-vie a ete de 5,1 h pour les malades et de 5,7 h pour les volontaires. Le retour du medicament au compartiment central, en provenance du compartiment peripherique profond, a ete le facteur controlant le taux pendant son elimination. Ni l'anesthesie par l'enflurane et le protoxyde d'azote, ni le stress chirurgical n'ont affecte la repartition ou l'eiimination du medicament se trouvant dans le plasma. PHARMAKOKINETIK VON THIOPENTON: WIRKUNGEN VON ENFLURAN UND SALPETRIGER OXYDANASTHESIE UND CHIRURGIE ZUSAMMENFASSUNG
Untersucht wurden die Plasmakonzentrationen von Thiopenton in sechs Patienten, die nach einer intravenOsen Injektion von 3,5 mg kg- 1 eine Enfluran und salpetrige Oxydanasthesie bekamen, und in sechs Testpersonen. Wir identifizierten in alien sechs Patienten und in der Halfte der freiwilligen Kontrollgruppe das System als ein dreiteiliges, offenes Modell mit einem "flachen" und einem "tiefen" peripheralen Teil, und in den restlichen Testpersonen als ein zweiteiliges, offenes Modell. Die Verteilung von Thiopenton im Gewebe ging sehr rasch, die Alpha- und Beta- Verteilungshalbwertzeiten waren durchschnittlich je 2,5 und 46,4 Minuten fur die Patientengruppe, und je 2,8 und 48,7 Minuten fiir die Kontrollgruppe. Die grosse Verbreitung der Droge wurde von dem augenscheinlichen Volumen der Verteilung angezeigt, dessen Mittelwerte zwischen 2- und 1,5-mal KSrpergewicht wechselten. Die durchschnittlichen Ausscheidungshalbwertzeiten waren 5,1 Stunden fiir die Patientengruppe und 5,7 Stunden fiir die Kontrollgruppe. Die Ruckkehr der Droge vom tiefen, peripheralen Teil zum zentralen Teil war der geschwindigkeitskontrollierende Umstand der Ausscheidung. Weder Enfluran und salpetrige Oxydanasthesie noch
Downloaded from http://bja.oxfordjournals.org/ at University of Birmingham on August 25, 2015
Adjepon-Yamoach, K. K., Scott, D. B., and Prescott, L. F. (1973). Impaired absorption and metabolism of oral lignocaine in patients undergoing laparoscopy. Br. J. Anaesth., 45,143. Becker, K. E., jr (1976). Gas chromatographic assay for free and total plasma levels of thiopental. Anesthesiology, 45, 656. Brodie, B. B., Mark, L. C , Papper, E. M., Lief, P. A., Bernstein, E., and Rovenstine, E. A. (1950). The fate of thiopental in man and a method for its estimation in biological material. J. Pharmacol. Exp. Ther., 98,85. Brown, B. R., jr (1971). The diphasic action of halothane on the oxidative metabolism of drugs by the liver: an in vitro study in the rat. Anesthesiology, 35, 241. Chung, H., and Brown, D. R. (1976). The mechanism of the effect of acute stress on hexobarbital metabolism. Toxicol. Appl. Pharmacol, 37, 313. Cooperman, L. H. (1972). Effects of anaesthetics on the splanchnic circulation. Br. J. Anaesth., 44,967. Gibaldi, M., and Perrier, D. (1975). Multi Compartments Models in Pharmacokinetics, 1st edn, p. 45. New York: Marcel Dekker, Inc. Greenblatt, D. J., and Koch-Weser, J. (1975). Clinical pharmacokinetics. N. Engl.J. Med., 293, 702. Shader, R. I., Franke, K., and Koch-Weser, J. (1975). Pharmacokinetics of intravenous chlordiazepoxide. Clin. Pharmacol. Ther., 17, 235. Juhl, B., and Einer-Jensen, N. (1976). Hepatic blood flow and cardiac output during Fluoromar anaesthesia. An animal study. Acta Anaesthesiol. Scand., 20, 271. Mark, L. C , Brand, L., Kamvyssi, S., Britton, R. C , Perel, j ; M., Landrau, M. A., and Dayton, P. G. (1965). Thiopental metabolism by human liver in vivo and in vitro. Nature {Lond.), 206,1117. Metzler, C. M. (1969). A User's Manual for NONLIN. Technical report 7292/69/7292/005. Kalamazoo, Michigan, U.S.A.: The Upjohn Co. Novelli, G. P., Marsili, M., and Lovenzi, P. (1975). Influence of liver metabolism on the actions of Althesin and thiopentone. Br.J. Anaesth., 47, 913. Rahn, E., Dayton, P. G., and Frederickson, E. L. (1969). Lack of effect of halothane on the metabolism of thiopentone in man. Br.J. Anaesth., 41, 503. Rescigno, A., and Segre, G. (1966). Drug and Tracer Kinetics, 1st edn, p. 91. London: Blaisdell Publishing Co. Sedman, A. J., and Wagner, J. G. (1974). AUTO AN: A Decision-Making Pharmacokinetic Computer Program. 615 East University Avenue, Ann Arbor, Michigan, U.S.A.: Publication Distribution Services. Swartz, R. D., Sidell, F. R., and Cucinell, S. A. (1974). Effects of physical stress on the disposition of drugs eliminated by the liver in man. J. Pharmacol. Exp. Ther., 188,1. 101
Van Hamme, M. J., and Ghoneim, M. M. (1978). A sensitive gas chromatographic assay for thiopentone in plasma. Br.J. Anaesth., 50,143.
BRITISH JOURNAL OF ANAESTHESIA
1242 der Stress der Chirurgie hatten Einfluss auf Verteilung oder Ausscheidung der Droge aus dem Plasma. FARMACOCINETICA DE TIOPENTONA: EFECTOS EJERCIDOS POR ANESTESIA DE ENFLURANO Y OXIDO NITROSO Y POR CIRUGIA SUMARIO
Downloaded from http://bja.oxfordjournals.org/ at University of Birmingham on August 25, 2015
Se estudiaron las concentraciones de tiopentona en la plasma de seis pacientes y de seis voluntaries que recibieron anestesia de enflurano y oxido nitroso tras la inyeccion intravenosa de 3,5 mg kg"1. Identificamos un sistema de modelo abierto de tres compartimentos que contenia tanto un compartimento periferico "poco profundo" como uno
"profundo" en todos los pacientes y en la mitad de los controles, y un modelo abierto de dos compartimentos en los voluntarios restantes. La distribucion de tiopentona por los tejidos fue rapida, promediendo los periodos de media vida de distribucion a y (3 2,5 y 46,4 min respectivamente en el grupo de los pacientes y 2,8 min y 48,7 min en el grupo de control. La extensa distribucion de la droga fue indicada por el volumen de distribuci6n aparente, cuyo promedio varid entre 2 y 1,5 veces el peso del cuerpo. El promedio de media vida de eliminaci6n fue de 5,1 h para los pacientes y 5,7 h para los voluntarios. El retorno de la droga al compartimento central desde el compartimento periferico profundo fue el factor que controld la rapidez de su eliminacion. Ni la anestesia de enflurano y 6xido nitroso, ni el "stress" causado por la cirugia afectaron la distribucidn o eliminacidn de la droga de la plasma.
