PLASMA CONCENTRATIONS AND CLINICAL EFFECTS OF LORAZEPAM AFTER ORAL ADMINISTRATION

Br.J. Anaesth. (1981), 53, 517

PLASMA CONCENTRATIONS AND CLINICAL EFFECTS OF LORAZEPAM AFTER ORAL ADMINISTRATION E. G. BRADSHAW, A. A. A L I , B. A. MULLEY AND R. M. RYE

SUMMARY

Lorazepam (Ativan, Wyeth) has become widely used for premedication because of good sedative and amnesic properties, and relative freedom from undesirable side-effects (Wilson and Ellis, 1973; Knapp and Fierro, 1974; Dundee et al., 1977). Blood concentrations of the drug and its kinetics have been studied in volunteers (Knowles, Comer and Ruelius, 1971; Greenblatt et al., 1979), but few reports have related the findings to clinical effects in individuals and there are inconsistencies in published data. This study describes the amnesic, sedative and anxiolytic effects of lorazepam in patients and volunteers and relates these effects to drug concentrations. The study provided further information about variability in blood concentrations and response, and the time of onset and duration of drug action following oral administration. This is important for the correct timing of administration before operation and for deciding when patients may be safely discharged from hospital. The study was confined to oral administration of the drug, a route with clinical advantages but which has received less attention in other studies. The results also provided further information on the half-life and other kinetic information.

METHODS

Blood concentrations Drug administration and sample collection. Male

and female patients undergoing minor surgery and two volunteers were studied under the same ward conditions after informed consent was obtained (table I). TABLE I. Subject data and drug dose. * Volunteers

Age

Subject

(yr)

Sex

1 2 3 4* 5* 6 7 8 9 10 11 12 13 14 15

34 21 44 34 30 27 36 55 57 31 39 26 43 35 26

F M M M F F F M F F M F F

F M

Weight (kg)

Dose (rag)

Dose (mgkg-')

57.4

3.0 3.5 3.5 3.5 3.0 3.0 2.5 3.5 2.5 4.0 3.0 3.0 3.5 2.5 3.5

0.052 0.050 0.049 0.050 0.054 0.058 0.056 0.056 0.056 0.053 0.057 0.050 0.049 0.052 0.055

70 72 70 56 52

44.5 63

44.5 76 53 60 72 48 64

Measurements of blood concentrations were made in subjects 1-9 only. No subject had received E. G. BRADSHAW, M.B., B.S., M.R.C.S., L.R.C.P., F.F.A.R.C.S., D.A. any drugs over the previous 24 h and all had fasted M.D., Department of Anaesthetics, University of Manchester Hope Hospital, Salford M6 8HD. A. A. ALI,* B.PHARM., D.C.C. from midnight. Lorazepam 50 ug kg ~ ' was adminPH.D.; B. A. MULLEY, B.PHARM., PH.D., C.CHEM., M.R.I.C. istered orally approximately 6h before surgery. M.P.S.; R. M. RYE, B.SC., PH.D., C.CHEM., M.R.I.C, M.P.S. Blood samples were obtained at 0, 0.5, 1, 1.5, 2 h Postgraduate School of Pharmacy, University of Bradford and then hourly for 6h. Subjects were either Bradford, W. Yorkshire BD7 1DP. * Present address: Faculty of Pharmacy, University of sitting or lying down during this period. Further Khartoum, P.O. Box 1996, Khartoum, Sudan. samples were taken at 9,24,33, and 48 h after drug 0007-0912/81/050517-06 801.00

© Macmillan Publishers Ltd 1981

Downloaded from http://bja.oxfordjournals.org/ at New York University on July 3, 2015

The relation between plasma concentrations and properties of lorazepam administered orally as premedication was studied in patients undergoing minor surgery. The drug was found to be reliably absorbed in most subjects producing plasma concentrations similar to those reported after injection. Effective concentrations were obtained within 30-60 min, maintained for 4-6 h and were closely related to the action of the drug on memory and its sedative and anxiolytic properties. Its kinetics were described in about half the subjects by a one-compartment model, but in others two compartments were required. Variable half-lives were found (range 9.3-32.1 h, mean about 20 h). Sufficient drug remains after 24 h to suggest that residua] c.n.s. effects may still be present.

518

tree selected in the same order for each subject, but avoiding any collusion between them. This avoided any possibility that some objects may be more readily remembered than others. At 24 h after drug administration, subjects were asked to recall these pictures. Those which had been forgotten were mixed with another group of cards, similar in colouring and dimension, and the subjects were then asked to recognize any they had seen before. The number of subjects recalling at each time and numbers either recalling or recognizing the cards were recorded. Short term recall was measured by asking the subjects to name a playing card seen and then to recall it 5 min later. Sedation was measured on a five-point scale: 1 being no sedation, 5 being so sedated that the patient could not be aroused by gentle shaking. The grading criteria were the same as those used by Pandit, Heisterkamp and Cohen (1976). Anxiety was measured by asking the subjects to estimate and mark a 10-cm visual analog scale in the range "calm" to "nervous". This test was omitted by the two volunteers since they were not undergoing operation. Judgement of time. All subjects were asked to estimate a 10-s time interval and the estimate recorded. The results of the clinical tests were analysed by the Mann-Whitney U Test (amnesia and sedation), Chi-square test (short-term recall) and paired t test (anxiety). RESULTS

Table II shows the plasma lorazepam concentrations in all nine subjects. Absorption of the drug in most cases was rapid so that plasma concentrations generally exceeded 30ngml~' after 30 min and peak concentrations were attained within 90 min. Subject 1, however, had much slower absorption characteristics. The drug concentration did not peak until about 5 h and a significant time-lag was observed before absorption commenced. A timelag was found in subject 7, although absorption was subsequently rapid. Table III shows the halfClinical study lives from the computer-fitting process for all nine A number of tests were carried out concurrent subjects. Data from five of the subjects fitted onewith the sampling of blood in the same subjects compartment kinetics and data from four, two (1-9) and in six others (subjects 10-15), over a 6-h compartments. The results of the clinical tests are shown in period after drug administration or until the patients underwent operation. Retrograde and table IV. The ability of the patients and volunteers anterograde amnesia were measured by showing to recall memory cards after 24 h was greatly the subject a different postcard-size coloured pic- reduced between 1-5 h after drug administration, ture of everyday objects such as a house, keys or a the effect being greatest at 2h. Even when the

Downloaded from http://bja.oxfordjournals.org/ at New York University on July 3, 2015

administration or as near these times as convenient. The samples were centrifuged and the plasma deep frozen until the concentration of lorazepam was determined by gas-liquid chromatography (GLC). GLC assay for lorazepam. A method similar to that described by Greenblatt, Franke and Shader (1978) was used, but the chromatograph was a Pye-Unicam Model 104 equipped with a nickel-63 electron capture detector and the carrier gas was nitrogen 40mlmin~'. Column temperature was 270 °C. Stock solutions of oxazepam (internal standard) and lorazepam were prepared in toluene. Peaks corresponding to the re-arranged forms of lorazepam and the internal standard oxazepam had retention times of 3.8 and 3.0 min respectively. An additional peak was sometimes observed in plasma samples left for a few hours at room temperature before extraction. Samples were therefore immediately deep-frozen after collection. Plasma samples (lml) to which internal standard (50 ng) had been added, were extracted once with benzene 5ml containing 1.5% isoamyl alcohol. A 4-ml aliquot of the organic phase was evaporated to dryness and the residue redissolved in toluene 50ulitre containing 15% isoamyl alcohol before injection (0.5 ulitre). Five calibration standards (10-50 ng ml ~ ') were analysed with each set of unknown samples. The slope of the calibration curve varied by approximately 25% over a period of about 3 months, but was much less for assays carried out within a weekly period and the precision was similar to that reported by Greenblatt and his co-workers. In plasma samples containing less than lOngml" ', the precision and accuracy of the assay was less. Results reported are the mean of duplicated determinations. Pharmacokinetic analysis. The drug concentrations in each individual were fitted to either a one-compartment or a two-compartment pharmacokinetic model by computer, using least squares regression analysis.

BRITISH JOURNAL OF ANAESTHESIA

PLASMA CONCENTRATIONS AND ACTIONS OF LORAZEPAM

519

TABLE II. Plasma lorazepam concentrations (ng ml '). * Time for some samples varied from l)\e column headlines. In these cases the time is gtven in parentheses after the blood concentration

Time (h)* Subject

0.5

1 2

42.1

1

1.5

2

3

4

5

6

9

24

23.6 30.3

26.4

33.2

28.8 24.7

33.6 24.6

48.6 20.2

46.0 19.4

38.2 14.6

28.4 12.0

0

3

37.4

46.6

44.8

—

43.5

44.8

43.4

37.0

—

34.6 (8)

4

53.0

5

41.4

42.3

6

42.4

55.1 34.4

36.2

35.8

42.1

34.5

37.8

35.8

33.5

27.1 33.1 25.4

27.1 30.4 —

17.6 (10) 22.2 (10) 16.6

17.0 (23) 10.9 9.4

11.4

48

8.0

6.1

(31) 11.4

(37)

8.5

5.3

(28)

(44)

2.9

1.8

(28)

(44)

9.6

6.9

(48.5)

(8) 7

0

37.4

36.0

—

32.4

8

45.7

25.6

29.9

26.3

23.4

9

33.8

?8.5

—

40.3

31.7 (4.5) 21.3

31.3

35.3

Number of compartments

Half-life

1

28.4 25.6 17.8 19.3

1 2 3 4 5 6 7 8 9

i 1 2 1 2 1 2 1

31.0 (5.75)

13.7

—

33.5 (5.5)

33.8 (6.5)

24.5 (8.75) 15.5 (8.5) 27.7 (8.75)

11.2 (24.75)

—

—

5.1

—

—

(23) 19.8

—

9.4

patients were asked to recognize the picture cards, there was still significant reduction of memory at these times. Short-term recall of the playing cards was also reduced between 1 and 5h. Sedation scores increased over the same period and anxiety was significantly reduced, even at 6h following administration. Estimates of the lO^s time interval were variable, eight subjects showing an increase and six a decrease. The average results are shown in table IV, but the differences are not significant. From the data obtained, a relationship between plasma lorazepam concentrations and clinical ef-

TABLE III. Half-lives of lorazepam in subjects 1-9

Subject no.

—

8.3

(h)

9.3

32.1 13.6 10.3 23.3

TABLE IV. Clinical effects of lorazepam. Mean for 15 subjects except for anxiety score, front which the two volunteers were excluded. *P<0.05

Time (h) 0

0.5

1

1.5

2

3

4

5

(n = 15) (n= 15) (n = 15) (n=15) (n •= 15) {n = 15) (n = 12) (n = H) Per cent recall of memory cards after 24 h Per cent recall and recognition of memory cards after 24 h Short-term recall (%) Sedation score Anxiety score Estimate of 10-s time interval

6

(« = 6)

100

93

20*

20*

7*

20*

42*

18*

67

100

93

33*

47*

20*

33*

75*

45*

83

100 1

87 1.6

33*

47*

27*

13*

67*

55*

3.3* 3.01* 10.23

3.1* 2.82* 10.55

83 2.0

5.91 8.53

5.36 8.87

2.4* 4.83* 9.73

2.8* 3.43* 9.64

3.06* 3.14* 9.79

2.8* 2.67* 8.70

3.20* 8.83

Downloaded from http://bja.oxfordjournals.org/ at New York University on July 3, 2015

32.1

45.0

39.0

33

BRITISH JOURNAL OF ANAESTHESIA

520

fects was demonstrated. Plasma concentrations in each individual patient were calculated from the computer results for each time at which clinical data had been obtained. Using these best estimates of plasma concentration, the results were analysed in groups at 10-ngml" 1 intervals in the range 0-50ngml" '. The relation between plasma lorazepam concentration and its amnesic, sedative and anxiolytic effects is shown in figure 1. The effects of lorazepam were maximum at plasma concentrations of 30-50 ng ml" 1 while, at less than lOngml" ', there was little action of the drug. Short- term (recall l) 100

10 20 30 40 50 concn (ng ml1) Sedation score

10 20 30 40 50 concn (ng ml"1)

Anxiety

3

10 20 30 40 50 concn (ng ml"1)

10 20 30 40 50 concn (ng ml"1)

FIG. 1. Relation between plasma concentrations and clinical effects. Top left: recall after 24 h (hatched) and recognition after 24 h (clear). The unit of measurement on the anxiety scale is cm. DISCUSSION

The mean peak concentration observed after adjusting the dose for body weight was 42.2 ng ml" '. This is in close agreement with the values of 45.8 and 39.0 ng ml" ' calculated by a similar procedure from the results obtained by Greenblatt and others (1979) using 2-mg and 4-mg doses. However, although some individual variation in blood concentrations is evident in our results (table II), it is considerably less marked than in the study reported by Dundee and others (1978). They also observed that plasma concentrations after oral

Downloaded from http://bja.oxfordjournals.org/ at New York University on July 3, 2015

Recall or recall and recognition after 24h CO

administration were less than after equal doses administered i.m. and that there was no significant difference in mean plasma concentrations after 2-mg and 4-mg doses. Our study showed that some patients had slower absorption characteristics than others. It is uncertain whether the slower absorption shown by subject 1 was caused by physiological factors, differences in release of the drug from individual tablets, a combination of these factors, or other causes. In general, our results obtained with patients are consistent with previously published findings observed in volunteers (Greenblatt et al., 1979). Compared with lorazepam, diazepam and some other benzodiazepines show a more marked multicompartment character. Elimination halflives for lorazepam (table III) range from 9.3 to 32.1 h (mean about 20 h), which is greater than the 16 h reported by Greenblatt and others (1979) in volunteers. Both figures are greater than earlier reports (Elliott, 1976; Greenblatt etal., 1976). The concentration of lorazepam remaining in the plasma after 24 h (mean 14 ng ml ~ ') is comparable to the value of 21 ng ml ~ ' reported by Dundee and others (1979). The effect of oral lorazepam on long-term memory was striking, with maximum failure to recall the pictures between 1-5 h after administration. Retrograde amnesia did not occur. Anterograde amnesia, although slightly slower in onset, was similar to that caused by the injection of 4-mg doses i.v. (Dundee and George, 1976; Pandit, Heisterkamp and Cohen, 1976). The short-term recall of playing cards at zero time was complete, but by 1 h was reduced to about 30%. The maximum effect occurred after 3h. Sedation scores paralleled the effect on memory. No patient was unrousable, but at the times of maximum effect, the subjects had to be shaken vigorously to answer questions and participate in the tests. Since the assessment procedure adopted in this study was identical to that described and used by Pandit, Heisterkamp and Cohen (1976), it is possible to compare directly the sedative effects of 4-mg doses i.v. used in their study with the smaller doses administered orally in this investigation. The onset of sedation following administration i.v. is more rapid, but the depth of sedation and duration of action are similar. These observations provide evidence that the oral administration of lorazepam is as effective as i.v. administration in similar doses.

PLASMA CONCENTRATIONS AND ACTIONS OF LORAZEPAM

521

chirurgicales mineures la relation qui existe entre les concentrations que Ton trouve dans le plasma et les proprietcs du We thank Dr T. V. A. Harry for useful discussions and other lorazepam administre par voie buccale, en tant que medicament assistance, Miss G. Crook and Ms M. I. Marples for help with pre-operatoire. II a etc trouve que ce medicament ctait fiablethe clinical part of the study, and Mrs S. Fox for carrying out ment absorbc par la plupart des sujets et qu'il produisait des some of the drug analyses. We are also grateful to Mr P. Luker concentrations dans le plasma similaires a celles que Ton avait signalces apres l'administration par injections. Les concenfor assistance with the computer calculations. trations ont effectivement etc obtenues dans un delai de 30 a 60 min et elles se sont maintcnues pendant 4 a 6 h de temps. Elles REFERENCES ont etc etroitement reliees a l'action de ce medicament sur la Dundee, J. W., and George, K. A. (1976). The amnesic action memoire et a ses proprictes sedatives et anxiolytiques. Sa of diazepam, flunitrazepam and lorazepam in man. Acta cinetique a etc decrite par un modele a un scul compartiment sur pres de la moitie des sujets, mais sur les autres il a fallu deux Anaesthesiol. Bclg., 27, (Suppl.), 3. Lilburn, J. K. (1978). Ketamine-lorazepam. Attenuation compartments. On a trouve des demi-vics variables (plage de of psychic sequelae of ketamine by lorazepam. Anaesthesia, 9,3 a 32,1 h, la moyenne ctant d'environ 20 h). II reste suffisamment de medicament apres 24 h pour permettre de penser qu'il 34, 213. Nair, S. G., and George, K. A. (1977). Studies of y a toujours des effets rcsiduels sur le systeme nervcux central. drugs given before anaesthesia. XXVI: Lorazepam. Br. J. Anatsth., 49, 1047. Toner, W., and Howard, P. J. (1978). Plasma PLASMAKONZENTRATIONEN UND KLINISCHE lorazepam levels. Anaesthesia, 33, 15. WIRKUNGEN VON LORAZEPAM NACH ORALER McGowan, W. A. W., Lilburn, J. K., McKay, A. C , and VERABREICHUNG Hegarty, J. E. (1979). Comparison of the actions of diazepam and lorazepam. Br. J. Anaesth., 51, 439. ZUSAMMENFASSUNG Elliott, H. W. (1976). Metabolism of lorazepam. Br. J. Anaesth., 48, 1017. Die Beziehung zwischen den Plasmakonzentrationen und den Greenblatt, D. J., Franke, K., and Shader, R. I. (1978). Eigenschaften von oral als Vorbehandlung verabreichtem Analysis of lorazepam and its glucuronide metabolite by Lorazepam wurde an Patienten mit kleineren Operationcn electron-capture gas-liquid chromatography. Use in untersucht. Die Droge wurde in den Korpern der meisten pharmacokinetic studies of lorazepam. J. Chromatogr., 146, Patienten gut absorbiert, mit Plasmakonzentrationen ahnlich jenen, wie sic nach der Injektion gegeben waren. Wirksame 311. Schillings, R. T., Kyriakopoulos, A. A., Shader, R. I., Konzentrationen wurden innerhalb von 30-60 Minuten erzielt, Sisenwine, S. F., Knowles, J. A., and Ruelius, H. W. (1976). wurden fur 4-6 Stunden aufrecht erhalten und standen im Clinical pharmocokinetics of lorazepam. 1. Absorption and engen Verhalmis zur Drogenwirkung auf das Gedachtnis und disposition of oral 14C-lorazepam. Clin. Pharmacol. Ther., zu den sedativen und anxiolytischen Eigenschaften der Droge. Bei etwa der Halfte der Patienten wurden die kinetischen 20, 329. ACKNOWLEDGEMENTS

Downloaded from http://bja.oxfordjournals.org/ at New York University on July 3, 2015

The results of this study confirm previous Greenblatt, D. J., Shader, R. I., Franke, K., MacLaughlin, D. S., Harmatz, J. S., Allen, M. D., Werner, A., and Woo, E. clinical experience with the drug. It is reliably (1979). Pharmacokinetics and bioavailability of intravenous, absorbed after oral administration and is clinically intramuscular, and oral lorazepam in humans. J. Pharm. Set., effective for 1-5 h, which makes it particularly 68, 57. useful for premedication of all patients on an Knapp, R. B., and Fierro, L. (1974). Evaluation of the cardiopulmonary safety and effects of lorazepam as a preoperating list at one time. The pharmacokinetic medicant. Anesth. Analg. (Cleve.), 53, 122. calculations show that lorazepam has a half-life of J. A., Comer, W. H., and Ruelius, H. W. (1971). 9-32 h and that the average plasma concentration Knowles, Disposition of 7-chloro-5-(o-chlorophenyl)-l, 3-dihydro-3after 24 h is 14ngml~ '. Therefore, if patients are hydroxy-2H-l, 4-benzodiazepin-2-one (lorazepam) in discharged from hospital less than 24 h after humans. Arzneim. Forsch., 21, 1055. administration, the blood concentrations of lora- Pandit, S. K., Hcisterkamp, D. V., and Cohen, P. J. (1976). Further studies of the anti-recall effect of lorazepam. A zepam may in some cases still produce sedative and dose-timc-effect relationship. Anesthesiology, 45, 495. amnesic effects. They must return home ac- Wilson, J., and Ellis, F. R. (1973). Oral premedication with companied and avoid additional c.n.s. depressant lorazepam (Ativan): a comparison with heptabarbitone (Medomin) and diazepam (Valium). Br. J. Anaesth., AS, 738. drugs such as alcohol. The study also confirms that lorazepam has not only prolonged sedative effects which are greater than those of diazepam, but also produces intense EFFETS CLINIQUES ET CONCENTRATIONS DE LORAZEPAM DANS LE PLASMA APRES amnesia. The amnesic effects are useful for unADMINISTRATION PAR VOIE BUCCALE pleasant procedures such as gastroscopy, sigmoidoscopy and the emergent delirium of keRESUME tamine (Dundee and Lilburn, 1978). Nous avons etudic sur des patients subissant des interventions

BRITISH JOURNAL OF ANAESTHESIA

522 Vcrhaltnisse dcr Droge durch ein einteiliges Modell beschrieben, wahrend bei den anderen Patienten zweiteilige Modclle crforderlich warcn. Die variablen Halbwertzeiten warcn: (Bereich 9,3-32,1 Stunden, Mittel etwa 20 Stunden). Genugende Drogenmengen verblieben nach 24 Stunden, um erkennen zu lassen, dass restliche c.n.s.-Wirkungen immer noch vorhanden sein konnen.

CONCENTRACIONES EN EL PLASMA Y EFECTOS CLINICOS DEL LORAZEPAM DESPUES DE LA ADMINISTRACION ORAL SUMAWO

Se estudio la relacion cntre las concentraciones en el plasma y

las propiedades del lorazepam, administrado por via oral cuaj premedicacion, en pacientes bajo tratamiento quirurjico menor. Se encontro que en la mayoria de los sujetos la droga era absorbida perfectamente, produciendo concentraciones en el plasma similares a las descritas despues de la inyeccion. Se obtu vieron concentraciones efectivas en el plazo de 30 a 60 min, sc mantuvicron de 4 a 6h y estuvieron extrechamente relacionadas con la acti vidad de la droga en la memoria y con sus propiedades sedanvas y anxioliticas. Sus aspectos cineticos se describieron en, aproximadamente, la mitad de los sujetos mediante el modelo de un compartimento, pero en otros fue necesario el modelo de dos compartimentos. Se enconrxaron vidas medias variables (gama de 93-32,1 h; media de 20 h aproximadamente). Despues de 24 h aim queda suficiente cantidad de droga como para sugerir que los efectos residuales en el sistema nervioso central estan aim presentes. Downloaded from http://bja.oxfordjournals.org/ at New York University on July 3, 2015