In vitro lipophilicity of a series of benzodiazepines was evaluated by octanol: buffer partition ratio at physiological pH, and by retention time on a reverse-phase high-pressure liquid chromatographic (HPLC) system with a neutral-pH mobile phase. Both approaches ranked diazepam as highly lipophilic, but overall the two indices were poorly correlated (r = 0.23). For seven of the benzodiazepines, the in vivo volume of distribution (Vd) was determined in pharmacokinetic studies. After correlation for individual values of protein binding, Vd for unbound drug was significantly correlated with octanol: buffer partition ratio (r = 0.74), and to a greater extent with HPLC retention (r = 0.81). Thus, lipid solubility at least partly determines the extent of benzodiazepine distribution in vivo, which in turn is a major determinant of the duration of clinical action after single doses.

The mechanism by which ethanol induces an increase in portal vein blood flow was studied in rats using radiolabeled microspheres. Ethanol (2 g/kg) by gavage resulted in an increase of 50-70% in portal vein blood flow. The ethanol-induced increase in portal blood flow was suppressed by the adenosine receptor blocker 8-phenyltheophylline [ethanol, 61.8 +/- 4.1 ml.kg-1.min-1; ethanol + 8-phenyltheophylline (0.2 mg.kg-1.min-1), 44.2 +/- 2.0 ml.kg-1.min-1; P less than 0.05]. By itself, 8-phenyltheophylline (0.2 mg.kg-1.min-1) was without effect on cardiac output or portal blood flow. Adenosine infusion resulted in a dose-dependent increase in portal blood flow with a maximal effect at a dose of 0.17 mg.kg-1.min-1 (control, 41.3 +/- 2.3; adenosine, 81.7 +/- 8.0 ml.kg-1.min-1; P less than 0.05). This adenosine-induced increase in portal blood flow was inhibited by 8-phenyltheophylline in a dose-dependent manner [adenosine, 81.7 +/- 8.0 ml.kg-1.min-1; adenosine + 8-phenyltheophylline (0.2 mg.kg-1.min-1), 49.8 +/- 6.6 ml.kg-1.min; P less than 0.05]. Both alcohol and adenosine significantly reduced preportal vascular resistance by 40% (P less than 0.02) and 60% (P less than 0.01), respectively. These effects were fully suppressed by 8-phenyltheophylline. It is concluded that adenosine is a likely candidate to mediate the ethanol-induced increase in portal vein blood flow. It is suggested that an increase in circulating acetate and liver hypoxia may mediate the effects of alcohol by increasing tissue and interstitial adenosine levels.

In eight normal volunteers, the combination of ethanol (0.5 g/kg) and diazepam (10 mg) administered orally produced a greater decrease in motor performance on a pursuit rotor than diazepam alone. The pharmacologic effect of diazepam was enhanced by 73% and this potentiation was associated with significantly greater diazepam concentrations (p less than 0.01) than after diazepam alone. The failure to observe any increase in the concentrations of the principal metabolite, N-desmethyl diazepam, during the period of enhanced pharmacologic effect precludes any change in the demethylating metabolic process as being responsible. The data suggest (0.10 greater than p greater than 0.05) a trend to a smaller volume of distribution of diazepam when ethanol is administered prior to diazepam ingestion. The subjects showed acute tolerance to the effects of diazepam. Lower plasma concentrations on the ascending side of the plasma diazepam concentration versus time profile were linked with the same pharmacologic responses associated with a greater drug concentration on the descending portion, of the same curve.

When disease impairs clearance of drugs, multiple-dose therapy may result in cumulation. The disposition of chlordiazepoxide (CDX), 50 mg infused intravenously over 10 min, was studied in 14 normal subjects and in 11 patients with biopsy-proven cirrhosis. In the normal subjects, mean (+/- SE) kinetic parameters were: t 1/2 beta, 10.0 (+/- 0.9) hr; Vd, 0.38 (+/- 0.04) l/kg; clearance, 0.54 (+/- 0.13) ml/min/kg. Clearance of total drug correlated inversely with serum albumin concentration in normal subjects (r = -0.63). Values in cirrhotic patients were: t 1/2 beta, 34.9 (+/- 8.7) hr; Vd, 0.34 (+/- 0.024) 1/kg; and clearance, 0.185 (+/- 0.34) ml/min/kg. Desmethylchlordiazepoxide (DMCDX), the major metabolite of CDX, appeared in blood of cirrhotic patients less rapidly than in normal subjects. Severity of liver disease did not indicate the impairment of CDX clearance. In 5 of the same cirrhotic patients, mean t 1/2 beta for oxazepam (7.1 +/- 1.0 hr) was 27% longer than in control subjects (5.6 +/- 0.7 hr); the difference is not significant. On kinetic grounds oxazepam may be preferable to chlordiazepoxide in cirrhotic patients since its elimination kinetics are not greatly altered in cirrhosis.