General Tranquilizers


Current methodology for determining plasma levels of diazepam and its active metabolites employs electron-capture gas chromatography and can measure 1- to 10-ng quantities ().

When a group of people are administered diazepam at a particular dose, there is a wide range of plasma concentrations of diazepam (). This range, in 13 subjects administered 15 mg of oral diazepam daily for a week or more, is 16 to 400 ng/ml (). Garattini et al. () found a range of 10 to 250 ng/ml in 27 subjects given a single 15-mg oral dose. A half-life of 38 to 92 minutes has been measured by this group in five female subjects. Kleijn et al. () calculated a half-life of 20 to 42 hours after 10 mg three times a day, about 0.5 mg/kg/day orally in five subjects.

This great variability in plasma concentration is a major problem in interpreting drug effects on human performance. Perhaps the factors that influence plasma concentration curves can be identified and controlled. With seven subjects, Linnoila et al. () showed that ingestion of food increases plasma levels of diazepam 6, 7, and 8 hours after intravenous administration of 0.3 mg of diazepam per kg. They suggest that enterohepatic recycling of diazepam is occurring. Kleijn et al. () had also suggested this mechanism.

The influence of ethanol on diazepam plasma levels was also studied. In both the Linnoila et al. study () and that of Haffner et al. () diazepam levels were higher when alcohol was also given. By contrast, Linnoila et al. () did not see significant alterations in chlordiazepoxide or thioridazine plasma concentrations after ethanol ingestion.

When diazepam dosage was administered on a body weight basis, great plasma concentration variations were still usually encountered (). However, Linnoila et al. () found much closer agreement in the levels of their seven subjects when both body weight and food intake were controlled. Unfortunately, the individual values are not presented in this article, but their standard error appears from the table to be about ± 20 for values between 300 and 400 ng/ml.

The sampling time is important. The peak after oral absorption occurs 0.5 to 3 hours after administration ().

Kanto et al. () suggest several reasons for believing that diazepam may induce its own metabolism to desmethyldiazepam in man. They point to the appearance, of high levels of the N-desmethyl metabolite in the plasma of chronic diazepam users after a single diazepam dose and to the very low levels seen in control subjects given the same dose. They also demonstrated a marked drop in the N-desmethyl metabolite plasma response to a diazepam dose after abstinence from diazepam in a chronic user. They present evidence that patients who have been taking diazepam for periods of months or years have lower plasma levels than those who have been on diazepam a few weeks, and show in a series of eight patients that after 1 to 6 weeks on 15 mg/day orally, there is a fall in plasma diazepam levels. Enzyme induction cannot explain the great variations seen in plasma diazepam levels because the N-desmethyl metabolite concentrations vary as much or more than the diazepam levels and generally are higher when diazepam levels are higher and lower when diazepam levels are lower ().

Diazepam binds to protein (). Relatively little diazepam can be recovered from hemolyzed red blood cells (). The proportion of a diazepam dose bound to tissue may be a critical factor in the variability of plasma levels. Interestingly, Zingales () finds low levels of the N-desmethyl metabolite in erythrocytes after cessation of diazepam administration and at a time when there is no longer a measurable plasma level of diazepam or its metabolites.

As with diazepam, a given dose of chlorpromazine produces a wide range of plasma levels (). Ingestion of food may alter the chlorpromazine plasma concentration. Fasting may result in rapid attainment of high plasma levels when chlorpromazine is subsequently administered orally. Since this effect is not always obtained, the authors suggest that residual food from a previous meal may be interfering with rapid absorption of the drug. Curry et al. () also present evidence for phenobarbital induction of increased metabolism of chlorpromazine. They found, in addition, that only about 1 percent of a l-g oral chlorpromazine dose was excreted in urine and feces. This study covers many ideas and presents a number of kinds of measurement. It suffers from a lack of numbers of subjects. In most cases the plasma levels represent measurements on only a single individual.

Cordon () reports that diazepam can produce the classic effects of addiction: tolerance, phychic dependence, and physical dependence. He presents only one case, but that is sufficient to demonstrate the matter.

Kleinknecht and Donaldson () review 23 articles which present data on the effects of diazepam on cognitive and psychomotor performance. They organize the many tests employed into groups according to the attributes they believe the tests to measure. They suggest that there is need to control and/or analyze for age, sex, personality (e.g., introversion vs. extroversion), and subject population (e.g., student vs. mental patient). Three of the articles reviewed by Kleinknecht and Donaldson are represented in the’ present summary (). Two additional articles dealing with the effects of tranquilizers and ethanol on simulated driving performance are added to the present survey ().

Haffner et al., Mørland et al. (), Linnoila (), and Loomis () have shown that diazepam (), chlordiazepoxide (), thioridazine (), haloperidole (), flupenthixole (), chlorpromazine (), secobarbital (), and meprobamate () in therapeutic doses can impair human performance in complex tests like simulated driving. Similarly, alcohol can impair this performance. When tranquilizers and ethanol are administered to the same subject, complex performance is impaired to a greater degree (). When separate elements of a complex task are looked at individually or tests for less complex behavior are administered, there is less consistency in the data (). With ethanol and diazepam, performance on some elements of simulated driving was improved over that with placebo and lower doses of the same agent ().

Lawton and Cahn () found only slight impairment with diazepam on three of five less complex psychomotor tests. Their oral dosage regimen was 5 mg three times a day for 3 days prior to testing and 5 mg just before testing on the morning of the fourth day. The 20 subjects also took 3 ounces of 100-proof vodka in grape juice or pure grape juice just before testing. Alcohol did not impair performance on these tests nor did it further increase the effect of diazepam. No drug concentrations were reported. The blood ethanol concentrations varied widely, from 68 to a surprising 140 mg/100 ml, and the change in concentration in 3 hours exceeded 60 mg/100 ml in every case. The average change was 82 mg/100 ml or over 27 mg/100 ml/hour, far too high to be consistent with values of 15 to 20 mg/100 ml/hour usually reported. There seems to have been a problem with the method of analysis, although other possibilities include improper storage before analysis and wide variations in sampling times.

General Tranquilizers: Comment

It is difficult to interpret the relation of diazepam plasma levels to behavioral responses because of the great variation in concentrations achieved with any particular dose. Food and alcohol intake patterns contribute to the variation.

There are some problems with the entire group of five behavioral studies (). All the subjects were healthy young males. No attempt was made to identify individuals suffering from anxiety. There seems to be great difficulty in designing or selecting tests for the mental and psychomotor elements of complex tasks. For instance, critical flicker-fusion frequency tests are very sensitive to diazepam effects, but the relationship of the test performance of complex human tasks is not obvious, In addition, either no plasma levels of the tranquilizers were measured () or large variations were reported among the subjects ().

However, these are careful studies. Haffner et al. () and Mørland et al. () presented a large range of mental and psychomotor tests, and their use of two diazepam dose levels permits delineation of the threshold for drug effects. All of the papers reviewed here report the use of placebos, and in all but one (), ethanol blood levels were measured. Different subjects were used for each test () or a randomized rotation procedure was used with at least a week between testing procedures ().


Selections from the book: “Drugs and Driving”. Robert Willette, Ph.D., editor. State-of-the art review of current research on the effects of different drugs on performance impairment, particularly on driving. National Institute on Drug Abuse Research Monograph 11. March 1977.