Comparative single-dose kinetics and

Transcription

1 Comparative single-dose kinetics and dynamics of lorazepam, alprazolam, prazepam, and placebo Thirty-nine healthy volunteers received single oral doses of either alprazolam (1 mg), lorazepam (2 mg), prazepam (20 mg), or placebo in a randomized, double-blind, parallel group study. Plasma drug concentrations, subjective self-ratings, and the digit symbol substitution test (DSST) were evaluated during 24 hours after dosage. Alprazolam was absorbed rapidly and produced correspondingly rapid sedation and impaired DSST performance. These effects also resolved rapidly, being similar to placebo by 4 to 6 hours after dosage. Sedative and DSST-impairing effects of loraz,epam were of slower onset but longer duration than those of alprazolam. After oral prazepam, appearance of desmethyldiazepam in plasma was slow, with minimal sedative and DSST-impairing effects. Twenty-four hours after dosage, both alprazolam and lorazepam significantly impaired recall of a list of 16 words learned previously 3 hours after dosage. Thus benzodiaz,epines with approximately equivalent clinical anxiolytic properties may have different sedative, performance-impairing, and amnesic profiles after single doses in healthy volunteers; these differences are explained at least in part by pharmacokinetic variations. (CLIN PHARMACOL TREK 1988;44: ) David J. Greenblatt, Jerold S. Harmatz, Cynthia Dorsey, and Richard I. Shader Boston, Mass. In controlled clinical trials of short or intermediate duration (2 to 4 weeks), appropriately chosen doses of various benzodiazepine antianxiety agents are more effective than placebo and comparable in efficacy to each other in the treatment of anxious outpatients with neuroses.1-3 However, clinical experience and some controlled studies suggest that, despite similar anxiolytic efficacy, benzodiazepines may have differing secondary pharmacologic properties, including the capacity to produce nonspecific sedation, impairment of intellectual performance and motor function, and alterations in information acquisition and recall.4-'5 Whether such van- From the Division of Clinical Pharmacology, Departments of Psychiatry and Medicine, Tufts University School of Medicine and New England Medical Center. Presented in part at the Eighty-seventh Annual Meeting of the American Society for Clinical Pharmacology and Therapeutics, Washington, D.C., March 21, Supported in part by grants MH and AG from the U.S. Public Health Service and a grant-in-aid from Parke- Davis/Warner-Lambert, Morris Plains, N.J., and Ann Arbor, Mich. Received for publication Dec. 12, 1987; accepted March 29, Reprint requests: David J. Greenblatt, MD, Division of Clinical Pharmacology, TuftsNew England Medical Center, Box 1007, 171 Harrison Ave., Boston, MA ations among benzodiazepines, should they exist, reflect actual qualitative differences in intrinsic neuropharmacologic properties, as opposed to pharmacokinetic differences, is not clearly established."'" The present study evaluated the character, time course, and intensity of secondary pharmacologic effects of three benzodiazepines, selected because of their differing pharmacokinetic properties, in a series of healthy volunteers after single therapeutic doses. METHODS Thirty-nine healthy male and female volunteers aged 19 to 44 years participated after giving informed consent. All were healthy, active, ambulatory adults with no history of medical disease and receiving no other medications. Women were not taking oral contraceptive preparations. Before participation, each subject underwent plasma sample analysis by electron capture GC" to exclude the presence of detectable amounts of other benzodiazepines. Subjects were randomly assigned to one of four treatment conditions, with dosage of the three benzodiazepines chosen as those considered to be of approximately equivalent anxiolytic efficacy during chronic treatment. The conditions were inactive placebo; lorazepam, 2 mg; 326

2 VOLUME 44 NUMBER 3 Kinetics and effects of benzodiazepines 327 Table I. Subject characteristics and pharmacokinetic variables Values are mean ± SE. *Smoking data not available for one subject taking alprazolam and one subject taking lorazepam. Placebo Alprazolam Lorazepam Prazepam Subject characteristics No Age (yr) 26(± 2) 27 (± 1) 30 (± 3) 26 (± 1) Weight (kg) 78 (± 8) 69 (± 2) 76 (± 4) 69 (± 5) Male/Female 5/3 9/3 9/0 5/5 Smoking (no/yes)* 6/2 6/5 4/4 7/3 Kinetic variables Cma (ng/ml) 16.5 (± 1.5) 18.0 (± 1.1) 105 (± 12) Ta. (hr after dose) 1.25 (± 0.3) 1.78 (± 0.2) 9.2(± 3) Elimination N (hr) 9.5 (± 0.7) 11.4 (± 1.1) Oral clearance (ml/min) 94 ( ± 7) 124 (± 15) alprazolam, 1 mg; and prazepam, 20 mg. All medications consisted of the commercially available oral dosage form, which was repackaged in an opaque capsule with a rapid disintegration rate. The four medications were identical, and treatment was double-blind. Subjects fasted overnight before each trial and ingested a light liquid breakfast (6 to 8 ounces of orange juice) 2 to 3 hours before drug administration. The study medication was given at approximately 9 AM together with 100 to 200 ml tap water. Subjects remained fasting until 3 hours after administration, at which time they resumed a normal diet. Venous blood samples were drawn into heparinized tubes before administration and at the following postdosage times: 1/2, 1, 2, 3, 4, 6, 8, and 24 hours. Venous blood samples were centrifuged, and plasma was separated and frozen until the time of assay. Subjective perceptions of sedative effects and mood state were obtained by self-rated 100 mm visual analog scales described previously Ratings were obtained twice before dosing and at the times indicated above corresponding to blood sampling. Assessment of behavioral manifestations of sedation, using an observer form of the same rating instrument, was also done by an investigator (C. D.) who was blind to the treatment condition. The digit symbol substitution test (DSST), adapted from the Wechsler Adult Intelligence Scale, was administered twice before dosage and 1, 2, 4, 8, and 24 hours after dosage. Each time, subjects completed one of 100 randomly selected test variants and were asked to make as many correct substitutions as possible in a period of 2 minutes. To reduce practice effects, the second of the two predosing scores on both the rating scales and the DSST was taken as the "base- line," with all subsequent scores expressed as an increment or decrement over the baseline score. Information acquisition and recall were evaluated with a standard word-list free-recall procedure.25 Three hours after administration a series of 16 words, taken from four different categories, were read in random order in "shopping list" fashion. Recall was tested immediately after presentation of the list; subjects wrote list items in any order. The list was then presented in a different random order, with subjects again writing down the items immediately. This was repeated six times. Twenty-four hours after dosing, subjects were asked to remember as many words as possible from the list; thereafter the learning and recall procedure was repeated again with the same list in six different random sequences. For subjects receiving alprazolam or lorazepam, plasma concentrations were determined by GC with electron capture detection.26'27 The peak plasma concentration (C,,,) and time of peak concentration (tmax) were used as indexes of the rate of drug absorption.' The slope of the terminal log-linear phase of the plasma concentration curve was determined by linear regression analysis and used to calculate the apparent elimination half-life (t112). The AUC until the final detectable concentration was determined by the linear trapezoidal method and extrapolated to infinity by addition of the final concentration divided by the terminal rate constant. Apparent oral clearance was calculated as the administered dose divided by total AUC. For individuals receiving prazepam, only desmethyldiazepam was measured in plasma samples,' consistent with the known pharmacokinetic properties of prazepam.3031 and t,,,. taken as indexes of the rate of desmethyldi-

3 CLIN PHARMACOL THER 328 Greenblatt et al. SEPTEMBER 1988 o 18- Cr 16-0_ '6 14- a_ N CC 0 10 co 8 cc Z 4 02 J 6 o, 11 / LORAZEPAM v_ f' ALPRAZOLAM _v DESMETHYLDIAZEPAM (after PRAZEPAM) HOURS AFTER DOSE o_ Lii -100 N -90 >- I -80 Lu (t) ) 0 CC 30 I J a- Fig. 1. Plasma concentrations of the three drugs after single oral doses. Each point is the mean for all subjects receiving that particular compound at the corresponding time. azepam appearance in plasma. However, elimination t1/2 and clearance could not be calculated, because the duration of sampling was insufficient relative to the long elimination ti,2 of desmethyldiazepam.'"' Statistical procedures included ANOVA, Student's t test, and Dunnett's t test. RESULTS The four treatment groups were statistically comparable in age (F = 0.95; df = 3,35), weight (F = 1.08; df -=- 3,35), gender composition (x2 = 6.28; df = 3), and smoking history (x2 = 1.60; df = 3) (Table I). Alprazolam appeared in the systemic circulation more rapidly than lorazepam (Table I; Fig. 1), with mean Lax values of 1.25 and 1.8 hours after administration, respectively. Alprazolam also was eliminated more rapidly than lorazepam, with a mean ti,2 of 9.5 hours as opposed to 11.4 hours for lorazepam. At 24 hours the plasma concentration of alprazolam averaged only 15% of the peak value, whereas the lorazepam concentration averaged 22% of the peak value. After oral prazepam, desmethyldiazepam appeared in the systemic circulation slowly, with a mean tmax value of 9.2 hours after dosage. By 24 hours the plasma concentra- tion had fallen to an average of only 79% of the maximum. Self-rated sedation scores after placebo administration indicated no significant change from baseline; if anything there was a trend toward reduced sedation scores in the placebo condition (Fig. 2). All three active medications produced significant increases over baseline in sedation. The greatest sedative effects were seen with alprazolam, with peak values occurring 1 and 2 hours after dosage. Thereafter sedative effects declined rapidly, being not significantly different from placebo 4 hours after dosage and thereafter. Effects of lorazepam were of slightly slower onset, being maximal 3 hours after dosage and remaining significantly elevated above baseline at 6 hours. Sedative effects of prazepam were of slowest onset, with significant differences from baseline observed only 2 and 3 hours after dosage. Findings on other self-ratings including fatigue, feeling "spacey," and the feeling of "thinking slowed down" followed similar patterns (Fig. 3), as did the observer's ratings of the degree of sedation. Self-ratings and observer ratings of sedation were significantly correlated over all treatments and sampling times (r = 0.62; p <0.001). Scores on the DSST after placebo administration im-

4 VOLUME 44 NUMBER 3 Kinetics and effects of benzodiazepines 329 C.o ca ca -o (/) CC \ Placebo o----o Lorazepam Alprazolam Praiepam I,-:;4.\ *. *17 /* \ //,>Sr%." \\\ * N7 76 (/) 10 Pr I e Hours After Dose 6 Fig. 2. Self-rated sedation scores based on the visual analog scale. Values are expressed as the increment or decrement over the predose baseline. Each point is the mean for all subjects in the indicated treatment group at corresponding times. Asterisks along the x axis indicate significant overall differences among the four treatment conditions based on ANOVA. (* p < 0.05; ** p < 0.01). Asterisks next to individual points indicate changes scores significantly different from zero change proved over time, consistent with learning effects (Fig. 4). Overall differences among the four groups reached statistical significance at 1, 2, and 3 hours after dosage. Impairment of DSST performance after alprazolam administration was of rapid onset, being maximal 1/2 hour after dosage; thereafter the learning curve in the alprazolam group moved toward the placebo curve, and by 3 to 4 hours after dosage, placebo and alprazolam values were similar. On the other hand, lorazepam produced more sustained decrements, lasting at least 3 to 4 hours after dosage (Fig. 4). Differences between placebo and prazepam treatment conditions at all times were minimal When tested 3 hours after administration, the number of words learned during the six learning trials differed among the four treatments (Table II), although the overall difference was not significant (F = 1.59; df = 3,35). Learning performance was best with placebo, with an average of 15.3 out of 16 words learned during the six trials. The number of words learned in all three active treatment groups was lower, with the individual lorazepam vs placebo difference being significant (p < 0.05, Dunnett's one-tailed t test). When recall was tested 24 hours after dosage, the four treatment conditions were significantly different (F = 4.3; df = 3,35; p < 0.02); an average of 14.3 words were recalled by placebo recipients, with significantly fewer words recalled in the alprazolam and lorazepam groups (9.8 and 11.1, respectively). The prazepam treatment group did not differ significantly from the placebo group. After the six relearning trials performed 24 hours after dosage, the four treatment conditions did not significantly differ (F = 1.27; df = 3,35), although the number of words relearned was lowest in the lorazepam treatment condition. DISCUSSION The present study evaluated the secondary pharmacologic properties of three benzodiazepine derivatives after single oral doses in healthy volunteers. Appro-

5 330 Greenblatt et al. CLIN PHARMACOL THER SEP1 EMBER / \ 0--- Placebo 0 Lorazepam Alprazolam Prazepam PRE A HOURS AFTER DOSE A Fig. 3. Self-rated scores on the visual analog scale for the item "thinking slowed downthinking speeded up." Values are expressed as the increment or decrement over the predose baseline. Each point is the mean for all subjects in the indicated treatment group at corresponding times. Asterisks along the x axis indicate significant overall differences among the four treatment conditions based on ANOVA (* p <0.05; ** p <0.01). Asterisks next to individual points indicate changes scores significantly different from zero change. priate choice of dosage is of course critical in such studies. Our choice of dosage levels for the three active medications corresponded, as best we could determine, to dosages that would have approximately equivalent anxiolytic efficacy during intermediate-duration treatment of anxious outpatients. Our results must be interpreted in light of this assumption. In any case, our findings indicate that all three active medications differed significantly from placebo, and the three active drugs also differed importantly from each other. Both alprazolam and lorazepam are reported to be moderately rapidly absorbed benzodiazepines, with peak plasma concentrations occurring between 1 and 2 hours after oral dosage.25'32-" In the present study the rate of alprazolam absorption was more rapid than that of lorazepam. On the other hand, prazepam acts almost as a "slow-release" benzodiazepine, with appearance in the systemic circulation of desmethyldiazepam occurring at a slow rate.' Differences among the three drugs in absorption rate were evident in the time course of self-rated sedative effects, as well as impairment of performance on the DSST. Sedative and performanceimpairing effects were of most rapid onset for alprazolam, followed by lorazepam. On the other hand, sedative effects of prazepam were of slower onset, and differences between prazepam and placebo in DSST performance were minimal. The relative durations of action of alprazolam and lorazepam also were consistent with their rate of disappearance from the systemic circulation as a result of distribution or elimination. Alprazolam was eliminated more rapidly from the blood than was lorazepam; sedative and DSST-impairing effects of lorazepam likewise were more sustained than were those associated with alprazolam. However, it is evident that the plasma concentration and pharmacokinetic properties of alprazolam and lorazepam do not completely explain their phannacodynamic profiles. For example, mean plasma lorazepam concentrations did not decline markedly between 1 and 6 hours after dosage (Fig. 1), whereas there were large decrements

6 VOLUME 44 NUMBER 3 Kinetics and effects of benzodiazepines 331 ca 0 co c 0- I CI) -4-,/- I - /s \ -o-- - Placebo o o Lorazepam v--- Alprazolam Prazepam Pre 4 8 Hours After Dose Fig. 4. Scores on the DSST in each of the four treatment conditions. Values are expressed as the increment or decrement over the pretreatment baseline score. Each point is the mean for all subjects in the indicated treatment assay at the corresponding time. Asterisks along the x axis indicate overall significant differences among the four treatment conditions (* p < 0.05; ** p < 0.01). Table H. Effect of benzodiazepines on information acquisition and recall Treatment group 3 Hr after dosage (six learning trials) Data are mean ± SE. *Significant (p <0.05) individual difference vs placebo with Dunnett's one-tailed t-test. t ANOVA; df = 3,35. Number of words learned or recalled Free recall 24 Hr after dosage Six relearning trials Placebo 15.3 ( ± 0.3) 14.3 (± 0.7) 15.4 ( 0.5) Alprazolam 13.2 (± 0.8) 9.8 (± 1.2)* 15.0 (± 0.6) Lorazepam 12.8 ( -± 1.1)* 11.1 (± 1.1)* 13.4 ( ± 1.1) Prazepam 13.6 (± 0.8) 13.2 ( -± 0.5) 14.8 ( ± 0.6) Value of Ft 1.59 (NS) 4.30 (p <0.02) 1.27 (NS) in the extent of impairment on the DSST ( Fig. 4). Likewise for alprazolam, effects on the DSST and selfratings of "thinking speed" were similar to those of placebo by 6 to 8 hours after dosage, whereas significant amounts of alprazolam remained in plasma at these time points. It is unlikely that these discrepancies are explained by a failure of plasma drug concentrations to reflect concentrations in the brain. On the contrary, both lorazepam and alprazolam rapidly equilibrate between systemic plasma and the central nervous system,3' with a constant brain-plasma concentration ratio thereafter maintained over time.38'39 Furthermore, the amount of drug available to interact with the presumed pharmacologic receptor is determined mainly by the amount present in whole brain.38'39 Rather the reduction in pharmacodynamic effects despite the persistence of active

7 332 Greenblatt et al. CLIN PHARMACOL THER SEPTEMBER 1988 drug in plasma is probably explained by the functional phenomenon of acute tolerance. Effects of benzodiazepines on information acquisition and recall are controversial.' Some reports suggest that certain benzodiazepines, such as lorazepam and triazolam, have a greater propensity to impair memory than other benzodiazepines.'" However, this is not a consistent finding, and it may be that the qualitative nature of memory impairment induced by various benzodiazepines is similar and closely linked with the degree of nonspecific sedation.485' Furthermore, apparent differences among drugs may be explained principally by variations in dosage, plasma concentration, and duration of action:7 In the present study, benzodiazepineinduced impairment of information acquisition and recall was clearly evident. When tested 3 hours after dosage, differences were observed in number of words learned among the three active compounds and placebo, although only the lorazepam-placebo difference attained statistical significance. The apparently greater amnesic effect of lorazepam in this setting may be related to the relatively sustained plasma level profile, yielding high plasma concentrations at 3 hours, together with the more sustained effect on self-rated sedation and DSST impairment. When recall of those words learned 3 hours after dosage was tested 24 hours after administration, placebo and the three active treatment conditions differed significantly, with the individual alprazolam-placebo and lorazepam-placebo differences being statistically significant. Although recall among prazepam recipients was less than in subjects receiving placebo, this individual difference did not reach significance. Finally, when relearning of the same word list was tested thereafter, the four treatment conditions did not differ significantly. These amnesic and relearning effects are consistent with previous studies,' indicating that benzodiazepines have the capacity to impair acquisition of information when testing is done at or near the plasma concentration peak after oral dosage. Furthermore, there was evidence of failure to recall even what was learned when retesting was conducted 24 hours after dosage. Differences among drugs appear to be largely quantitative, with no evidence of a qualitative variation in the character of the memory impairment. After single oral doses in healthy volunteers, all three of these benzodiazepines produced pharmacologic effects that were significantly different from those of placebo. These included significant sedative effects, impairment of performance on a simple psychomotor test, and alterations in information acquisition and recall. Differences among the three drugs in the time course and intensity of such effects were at least partly explained by differing pharmacokinetic properties. We are grateful for the assistance of Ann Locniskar and Rita Maths. References Riekels K, Schweizer EE. Current pharmacotherapy of anxiety and panic. In: Meltzer HY, ed. Psychopharmacology: the third generation of progress. New York: Raven Press, 1987: Greenblatt DJ, Shader RI. Benzodiazepines in clinical practice. New York: Raven Press, 1974: Greenblatt DJ, Shader RI. Pharmacotherapy of anxiety with benzodiazepines and p-adrenergic blockers. In: Lipton MA, DiMascio A, Killam KF, eds. Psychopharmacology: a generation of progress. New York: Raven Press, 1978: Shader RI, Pary RJ, Harmatz JS, Allison S, Locniskar A, Greenblatt DJ. Plasma concentrations and clinical effects after single oral doses of prazepam, clorazepate, and diazepam. J Clin Psychiatry 1984;45: Jaffe JH, Ciraulo DA, Nies A, Dixon RB, Monroe LL. Abuse potential of halazepam and of diazepam in patients recently treated for acute alcohol withdrawal. CLIN PHAR- MACOL THER 1983;34: Roache JD, Griffiths RR. Repeated administration of diazepam and triazolam to subjects with histories of drug abuse. Drug Alcohol Depend 1986;17: Subham Z, Harrison C, Hindmarch I. Alprazolam and lorazepam: single- and multiple-dose effects on psychomotor skills and sleep. Eur J Chin Pharmacol 1986; 29: Lucki I, Rickels K, Geller AM. Chronic use of benzodiazepines and psychomotor and cognitive test performance. Psychopharmacology 1986;88: Aranko K. Task-dependent development of crosstolerance to psychomotor effects of lorazepam in man. Acta Pharmacol Toxicol 1985;56: Aranko K, Mattila MJ, Bordignon D. Psychomotor effects of alprazolam and diazepam during acute and subacute treatment, and during the follow-up phase. Acta Pharmacol Toxicol 1985;56: Ellinwood EH, Heatherly DG, Nikaido AM, Bjomsson TD, Kilts C. Comparative pharmacokinetics and pharmacodynamics of lorazepam, alprazolam and diazepam. Psychopharmacology 1985;86: Griffiths RR, McLeod DR, Bigelow GE, Liebson IA, Roache JD, Nowowieski P. Comparison of diazepam and oxazepam: preference, liking and extent of abuse. J Pharmacol Exp Ther 1984;229: Griffiths RR, McLeod DR, Bigelow GE, Liebson IA, Roache JD. Relative abuse liability of diazepam and oxazepam: behavioral and subjective dose effects. Psychopharmacology 1984;84:

8 VOLUME 44 NUMBER 3 Kinetics and effects of benzodiazepines 333 McNicholas LF, Martin WR, Cherian S. Physical dependence on diazepam and lorazepam in the dog. J Pharmacol Exp Ther 1983;226: Griffiths RR, Lamb RJ, Ator NA, Roache JD, Brady JV. Relative abuse liability of triazolam: experimental assessment in animals and humans. Neurosci Biobehav Rev 1985;9: Mewaldt SP, Ghoneim MM, Hinrichs JV. The behavioral actions of diazepam and oxazepam are similar. Psychopharmacology 1986;88: Greenblatt DJ, Shader RI. Pharmacokinetics of antianxiety agents. In: Meltzer HY, ed. Psychophannacology: the third generation of progress. New York: Raven Press, 1987: Greenblatt DJ. Analysis of benzodiazepines by electroncapture gas-liquid chromatography. In: Sunshine I, Jatlow PI, eds. Methodology for analytical toxicology, vol 2. Boca Raton, FL: CRC Press, 1982: Greenblatt DJ, Hannatz JS, Zinny MA, Shader RI. Effect of gradual withdrawal on the rebound sleep disorder after discontinuation of triazolam. N Engl J Med 1987; 317: Greenblatt DJ, Shader RI, Harmatz JS, Franke K, Koch- Weser J. Absorption rate, blood concentrations, and early response to oral chlordiazepoxide. Am J Psychiatry 1977;134: Shader RI, Georgotas A, Greenblatt DJ, Hannatz JS, Allen MD. Impaired absorption of desmethyldiazepam from clorazepate by magnesium aluminum hydroxide. CLIN PHARMACOL THER 1978;24: Ochs HR, Greenblatt DJ, Eckardt B, Harmatz JS, Shader RI. Repeated diazepam dosing in cirrhotic patients: cumulation and sedation. CLIN PHARMACOL THEE 1983; 33: Greenblatt DJ, Abernethy DR, Morse DS, Harmatz JS, Shader RI. Clinical importance of the interaction of diazepam and cimetidine. N Engl J Med 1984;310: Scavone JM, Greenblatt DJ, Harmatz JS, Shader RI. Kinetic and dynamic interaction of brotizolam and ethanol. Br J Clin Pharmacol 1986;21: Shader RI, Dreyfuss D, Gerrein JR, Harmatz JS, Allison Si, Greenblatt DJ. Sedative effects and impaired learning and recall after single oral doses of lorazepam. CLIN PHARMACOL THEE 1986;39: Greenblatt DJ, Franke K, Shader RI. Analysis of lorazepam and its glucuronide metabolite by electroncapture gas-liquid chromatography: use in phannacokinetic studies of lorazepam. J Chromatogr 1978;146:311-20, Greenblatt DJ, Divoll M, Shader RI. Automated gas chromatographic determination of plasma alprazolam concentrations. J Clin Psychopharmacol 1983;3: Greenblatt DJ, Koch-Weser J. Clinical pharmacokinetics. N Engl J Med 1975;293:702-5; Greenblatt DJ, Ochs HR, Lloyd BL. Entry of diazepam and its major metabolite into cerebrospinal fluid. Psychopharmacology 1980;70: Allen MD, Greenblatt DJ, Harmatz JS, Shader RI. Desmethyldiazepam kinetics in the elderly after oral prazepam. CLIN PHARMACOL THER 1980;28: Brodie RR, Chasseaud LF, Taylor T. Concentrations of N-descyclopropylmethylprazepam in whole blood, plasma, and milk after administration of prazepam to humans. Biopharm Drug Dispos 1981;2: Smith RB, Kroboth PD, Vanderlugt JT, Phillips JP, Juhl RP. Pharmacokinetics and phannacodynamics of alprazolam after oral and IV administration. Psychopharmacology 1984;84: Greenblatt DJ, Divoll M, Abernethy DR, Ochs HR, Shader RI. Benzodiazepine kinetics: implications for therapeutics and pharmacogeriatrics. Drug Metab Rev 1983;14: Greenblatt DJ, Divoll M, Abernethy DR, Ochs HR, Shader RI. Clinical pharmacokinetics of the newer benzodiazepines. Clin Pharmacokinet 1983;8: Greenblatt DJ, Divoll M, Hannatz JS, Shader RI. Pharmacokinetic comparison of sublingual lorazepam with intravenous, intramuscular, and oral lorazepam. J Pharm Sci 1982;71: Greenblatt DJ. Clinical pharmacokinetics of oxazepam and lorazepam. Clin Pharmacokinet 1981;6: Arendt RM, Greenblatt DJ, dejong RH, et al. In vitro correlates of benzodiazepine cerebrospinal fluid uptake, phannacodynamic action, and peripheral distribution. JPharmacol Exp Ther 1983;227: Miller LG, Greenblatt DJ, Paul SM, Shader RI. Benzodiazepine receptor occupancy in vivo: correlation with brain concentrations and pharmacodynamic actions. J Pharmacol Exp Ther 1987;240: Miller LG, Greenblatt DJ, Barnhill JG, Deutsch SI, Shader RI, Paul SM. Benzodiazepine receptor binding of triazolobenzodiazepines in vivo: increased receptor number with low-dose alprazolam. J Neurochem 1987; 49: Greenblatt DJ, Shader RI, Harmatz JS, Georgotas A. Self-rated sedation and plasma concentrations of desmethyldiazepam following single doses of clorazepate. Psychopharmacology 1979;66: Greenblatt DJ, Shader RI, Abernethy DR. Current status of benzodiazepines. N Engl J Med 1983;309:354-8; Kales A, Kales JD, Bixler EO, Scharf MB, Russek E. Hypnotic efficacy of triazolam: sleep laboratory evaluation of intermediate-term effectiveness. J Clin Pharmacol 1976;16: Scharf MB, Khosla N, Brocker N, Goff P. Differential amnestic properties of short- and long-acting benzodiazepines. J Clin Psychiatry 1984;45:51-3. Shader RI, Greenblatt DJ. Triazolam and anterograde

9 334 Greenblatt et al. CLIN PHARMACOL THER SEPTEMBER 1988 amnesia: all is not well in the z-zone. J Clin Psychophannacol 1983;3:273. Healy M, Pickens R, Meisch R, McKenna T. Effects of clorazepate, diazepam, lorazepam, and placebo on human memory. J Clin Psychiatry 1983;44: Kothary SP, Brown ACD, Pandit UA, Samra SK, Pandit SK. Time course of antirecall effect of diazepam and lorazepam following oral administration. Anesthesiology 1981;35: Bixler EO, Kales A, Brubaker BH, Kales JD. Adverse reactions to benzodiazepine hypnotics: spontaneous reporting system. Pharmacology 1987;35: Kumar R, Mac DS, Gabrielli WF, Goodwin DW. Anxiolytics and memory: a comparison of lorazepam and alprazolam. J Clin Psychiatry 1987;48: Block RI, Berchou R. Alprazolam and lorazepam effects on memory acquisition and retrieval processes. Pharmacol Biochem Behav 1984;20: Lister RG. The amnesic action of benzodiazepines in man. Neurosci Biobehav Rev 1985;9: Patat A, Klein MJ, Hucher M. Effects of single oral doses of clobazam, diazepam and lorazepam on performance tasks and memory. Eur J Clin Pharmacol 1987; 32:461-6.