Caffeine elimination: a test of liver function

Pharmacokinetics of Caffeine: A Systematic Analysis of Reported Data for Application in Metabolic Phenotyping and Liver Function Testing

Caffeine is by far the most ubiquitous psychostimulant worldwide found in tea, coffee, cocoa, energy drinks, and many other beverages and food. Caffeine is almost exclusively metabolized in the liver by the cytochrome P-450 enzyme system to the main product paraxanthine and the additional products theobromine and theophylline. Besides its stimulating properties, two important applications of caffeine are metabolic phenotyping of cytochrome P450 1A2 (CYP1A2) and liver function testing. An open challenge in this context is to identify underlying causes of the large inter-individual variability in caffeine pharmacokinetics. Data is urgently needed to understand and quantify confounding factors such as lifestyle (e.g., smoking), the effects of drug-caffeine interactions (e.g., medication metabolized via CYP1A2), and the effect of disease. Here we report the first integrative and systematic analysis of data on caffeine pharmacokinetics from 141 publications and provide a comprehensive high-quality data set on the pharmacokinetics of caffeine, caffeine metabolites, and their metabolic ratios in human adults. The data set is enriched by meta-data on the characteristics of studied patient cohorts and subjects (e.g., age, body weight, smoking status, health status), the applied interventions (e.g., dosing, substance, route of application), measured pharmacokinetic time-courses, and pharmacokinetic parameters (e.g., clearance, half-life, area under the curve). We demonstrate via multiple applications how the data set can be used to solidify existing knowledge and gain new insights relevant for metabolic phenotyping and liver function testing based on caffeine. Specifically, we analyzed 1) the alteration of caffeine pharmacokinetics with smoking and use of oral contraceptives; 2) drug-drug interactions with caffeine as possible confounding factors of caffeine pharmacokinetics or source of adverse effects; 3) alteration of caffeine pharmacokinetics in disease; and 4) the applicability of caffeine as a salivary test substance by comparison of plasma and saliva data. In conclusion, our data set and analyses provide important resources which could enable more accurate caffeine-based metabolic phenotyping and liver function testing.

PK-DB: pharmacokinetics database for individualized and stratified computational modeling

A multitude of pharmacokinetics studies have been published. However, due to the lack of an open database, pharmacokinetics data, as well as the corresponding meta-information, have been difficult to access. We present PK-DB (https://pk-db.com), an open database for pharmacokinetics information from clinical trials. PK-DB provides curated information on (i) characteristics of studied patient cohorts and subjects (e.g. age, bodyweight, smoking status, genetic variants); (ii) applied interventions (e.g. dosing, substance, route of application); (iii) pharmacokinetic parameters (e.g. clearance, half-life, area under the curve) and (iv) measured pharmacokinetic time-courses. Key features are the representation of experimental errors, the normalization of measurement units, annotation of information to biological ontologies, calculation of pharmacokinetic parameters from concentration-time profiles, a workflow for collaborative data curation, strong validation rules on the data, computational access via a REST API as well as human access via a web interface. PK-DB enables meta-analysis based on data from multiple studies and data integration with computational models. A special focus lies on meta-data relevant for individualized and stratified computational modeling with methods like physiologically based pharmacokinetic (PBPK), pharmacokinetic/pharmacodynamic (PK/PD), or population pharmacokinetic (pop PK) modeling.

Plasma hydroxy-metronidazole/metronidazole ratio in hepatitis C virus-induced liver disease

It has been suggested that the measurement of metronidazole clearance is a sensitive method for evaluating liver function. The aim of this study was to evaluate the usefulness of plasma hydroxy-metronidazole/metronidazole ratios as indicators of dynamic liver function to detect changes resulting from the various forms of chronic hepatitis C virus (HCV) infection. A total of 139 individuals were studied: 14 healthy volunteers, 22 healthy, asymptomatic, consecutive anti-HCV-positive HCV-RNA negative subjects, 81 patients with chronic hepatitis C (49 with moderate/severe chronic hepatitis and 34 with mild hepatitis), and 20 patients with cirrhosis of the liver. HCV status was determined by the polymerase chain reaction. Plasma concentrations of metronidazole and its hydroxy-metabolite were measured by reverse-phase high-performance liquid chromatography with ultraviolet detection in a blood sample collected 10 min after the end of a metronidazole infusion. Anti-HCV-positive HCV-RNA-negative individuals demonstrated a significantly reduced capacity to metabolize intravenously infused metronidazole compared to healthy individuals (0.0478 +/- 0.0044 vs 0.0742 +/- 0.0232). Liver cirrhosis patients also had a reduced plasma hydroxy-metronidazole/metronidazole ratio when compared to the other groups of anti-HCV-positive individuals (0.0300 +/- 0.0032 vs 0.0438 +/- 0.0027 (moderate/severe chronic hepatitis) vs 0.0455 +/- 0.0026 (mild chronic hepatitis) and vs 0.0478 +/- 0.0044 (anti-HCV-positive, HCV-RNA-negative individuals)). These results suggest an impairment of the metronidazole metabolizing system induced by HCV infection that lasts after viral clearance. In those patients with chronic hepatitis C, this impairment is paralleled by progression of the disease to liver cirrhosis.

Plasma hydroxy-metronidazole/metronidazole ratio can detect early changes in hepatic function in ethanol-induced liver injury

AIMS

To evaluate the usefulness of plasma hydroxy-metronidazole/metronidazole (OH-MET/MET) ratios as a dynamic liver function test in ethanol abusers with or without liver cirrhosis.

METHODS

Metronidazole was administered intravenously for 20 min to healthy volunteers, and to patients with alcohol-induced, non-cirrhotic hepatopathy and liver cirrhosis. Plasma concentrations of metronidazole and hydroxy-metronidazole were measured by high performance liquid chromatography in samples collected 5, 10, 20 and 30 min after the metronidazole infusion.

RESULTS

Patients with non-cirrhotic alcoholic hepatopathy had significantly elevated aminotransferase levels compared to healthy volunteers and Child A patients. Child-Pugh C patients had significantly prolonged prothrombin times when compared to healthy volunteers and patients with non-cirrhotic hepatopathy. Metronidazole metabolism, as measured by the OH-MET/MET ratio following the intravenous administration of 500 mg of the drug, was significantly impaired in all ethanol-abusing individuals, including patients with non-cirrhotic alcoholic hepatopathy.

CONCLUSIONS

Metronidazole metabolism was impaired in ethanol abusers, even in the absence of liver cirrhosis, indicating that ethanol was capable of affecting liver function in the early stages of alcohol-induced liver disease.

Plasma hydroxy metronidazole/metronidazole ratio in anti-HCV carriers with and without apparent liver disease

AIMS

To evaluate plasma hydroxy-metronidazole/metronidazole ratio as a dynamic liver function test in HCV-infected individuals with/without liver disease, in the absence of liver cirrhosis.

METHODS

Metronidazole was administered intravenously in healthy volunteers, asymptomatic anti-HCV-positive blood donors, and in chronic hepatitis C patients. Serology to HCV was determined by a second generation assay and confirmed by gelatin particle agglutination test using recombinant antigens C22-3 and C200. Plasma concentration of metronidazole and hydroxy-metronidazole was measured by high performance liquid chromatography in samples collected 5, 10, 20 and 30 min following the end of metronidazole infusion.

RESULTS

Chronic hepatitis C patients had abnormal liver enzymes, while healthy volunteers and anti-HCV-positive blood donors had normal liver biochemistry tests. Plasma metronidazole concentration was similar in all groups studied. Plasma hydroxy-metronidazole/metronidazole ratio was significantly reduced in HCV-infected subjects, an effect observed 10 min after the end of drug infusion.

CONCLUSIONS

Metronidazole clearance is impaired in anti-HCV-positive blood donors and chronic hepatitis C patients, indicating that HCV is capable of affecting liver function at early stages of the disease. The metronidazole clearance test can detect impaired liver function in HCV-infected individuals even in the absence of liver cirrhosis.

Xenobiotic-induced hepatotoxicity: mechanisms of liver injury and methods of monitoring hepatic function

Xenobiotic-induced liver injury is a clinically important etiology of hepatic disease that, if not recognized, can lead to hepatic failure. In this article we discuss the mechanisms of xenobiotic-induced liver injury, various factors that can alter the risk and severity of injury, the clinical and laboratory manifestations of injury, and the methods used to detect the presence of injury and (or) functioning liver mass.

Determination of caffeine and its metabolites by micellar electrokinetic capillary electrophoresis

The determination of caffeine and its analogues is important for a wide variety of analyses and is performed in an assortment of matrices ranging from food to clinical samples. While reversed-phase HPLC has become the standard analysis protocol in most laboratories, capillary electrophoresis has the advantages of higher separation efficiency and shorter separation time. The micellar capillary electrophoresis (MECC) separation of caffeine and its metabolites, theobromine, paraxanthine, theophylline and 1,3,7-trimethyluric acid was investigated using sodium dodecyl sulphate (SDS) as the micellar phase. The effects of pH, micelle concentration, buffer concentration, ionic strength, buffer salts, applied voltage and injection time were studied to select the optimum conditions for the determination of caffeine and its four analogues in drugs, foods and body fluids. Caffeine and its three analogues were resolved within 120 s with detection limits less than 1 microgram/ml. Samples could be analyzed utilizing direct injection with satisfactory resolution and reproducibility.

The effect of liver disease on urine caffeine metabolite ratios

OBJECTIVES

A number of caffeine metabolite ratios (CMRs) have been proposed to measure CYP1A2 activity in vivo. The effect of liver disease on these ratios is not clear. The objective of this study was to determine the influence of liver disease on caffeine metabolite ratios.

STUDY DESIGN

Two studies were performed. First, in healthy control subjects and in subjects with cirrhosis, caffeine clearance was measured by intravenous infusion of stable isotope-labeled caffeine while subjects consumed oral caffeine. Second, spot urine samples were collected from control subjects and from subjects with either chronic hepatitis or cirrhosis while they consumed caffeine.

RESULTS

In study 1, caffeine clearance was decreased in subjects with cirrhosis (control mean, 0.14 L/hr/kg; cirrhosis mean, 0.04 L/hr/kg; p = 0.003). CMRs were affected by liver disease (e.g., ratio characterizing paraxanthine demethylation [AAMU + 1X + 1U/17U], control median, 8.3; cirrhosis median, 6.2; p = 0.005). AAMU + 1X + 1U/17U correlated significantly with caffeine clearance in the control group (r2 = 0.68; p = 0.0001) and in subjects with cirrhosis (r2 = 0.41; p = 0.05). In study 2, there was a significant difference between control subjects and subjects with cirrhosis for AAMU + 1X + 1U/17U (median for control subjects, 6.2; median for subjects with cirrhosis, 4.3; p = 0.001) but not between control subjects and patients with chronic hepatitis.

CONCLUSIONS

We conclude that AAMU + 1X + 1U/17U is affected by liver disease and reflects the decrease in CYP1A2 activity in subjects with cirrhosis. AAMU + 1X + 1U/17U measured from a spot urine sample reflects caffeine clearance in subjects with cirrhosis and may be useful as a hepatic function test. Despite the large interindividual variation observed, the test can be repeated easily in the same patient and an individual patient's decline in CYP1A2 activity, such as in patients with progressively deteriorating liver function, can be monitored.

Plasma hydroxy-metronidazole/metronidazole ratio in patients with liver disease and in healthy volunteers

Metronidazole pharmacokinetics were studied in patients with different degrees of liver cirrhosis, classified according to the Child-Pugh algorithm (A, B or C, as liver disease severity increases) and in schistosomic patients. Metronidazole (500 mg) was administered i.v. as a slow infusion over 20 min, and blood samples were collected at set intervals after the end of the infusion. The plasma concentrations of metronidazole and its main metabolite hydroxy-metronidazole were quantified by reversed-phase h.p.l.c. with u.v. detection. The metronidazole and hydroxy-metronidazole areas under the curve from 0 to 24 h (AUC0,24h), the metronidazole terminal elimination half-life (t1/2), the total clearance (CL), the metronidazole volume of distribution (V) values and the hydroxy-metronidazole/metronidazole concentration ratios as a function of time were calculated for each group. Comparison of the metronidazole AUC0,24h, t1/2 and CL values revealed that metronidazole metabolism is progressively impaired as the severity of liver disease increases. There were no variations in these parameters between the schistosomic and Child-Pugh A groups. In addition, there were no differences in the V and hydroxy-metronidazole AUC0,24h among the various groups studied. However, metronidazole metabolism was delayed in patients with hepatic disease, as illustrated by the hydroxy-metronidazole/metronidazole ratio 10 min after the end of metronidazole infusion. These results indicate that the clinical assessment of liver disease is paralleled by an impairment of metronidazole metabolism. Of the studied variables, we propose the hydroxy-metronidazole/metronidazole ratio 10 min after metronidazole infusion as a suitable and practical index for liver function evaluation.

Simplified approach for evaluation of hepatic drug-oxidizing capacity with a simultaneous measurement of caffeine and its primary demethylated metabolites in carbon tetrachloride-intoxicated rats

1. Prediction of hepatic injury from changes in blood concentrations of caffeine (CA) and its three primary metabolites (theobromine: TB; paraxanthine: PX; theophylline; TP) after CA administration has been studied in carbon tetrachloride (CCl4)-intoxicated rats. 2. The plasma half-life (t1/2) of CA (10 mg/kg, oral) was increased, and total body clearance (CL) decreased, in CCl4-treated rats; the apparent volume of distribution (Vd) was relatively unchanged. 3. The ratios of CA to the three metabolites (TB/CA, PX/CA, TP/CA) were significantly decreased compared to those of controls. 4. There were good correlations between the ratios of TB/CA, PX/CA, and TP/CA and CL of CA (TB/CA; r = 0.944 at 1 h, r = 0.942 at 2 h: PX/CA; r = 0.974 at 1 h, r = 0.923 at 2 h: TP/CA; r = 0.866 at 1 h, r = 0.962 at 2 h). 5. Results indicate that it is possible to evaluate hepatic drug-oxidizing capacity by measuring the ratio of CA to its metabolites, using single blood sampling 1 or 2 h after CA administration in CCl4-intoxicated rats.

Simultaneous determination of caffeine and its primary demethylated metabolites in human plasma by high-performance liquid chromatography

An improved high-performance liquid chromatographic method for the simultaneous determination of caffeine and its three primary metabolites (theophylline, theobromine and paraxanthine) in human plasma is described. The four substances were separated on a reversed-phase column (5 microns TSK gel ODS-80TM, 150 mm x 4.6 mm I.D.) by use of the mobile phase methanol-0.1 M NaH2PO4 (30:70, v/v) with a flow-rate of 0.8 ml/min. Absorbance was monitored at 274 nm. The detection limit was 5 ng/ml for theobromine and caffeine and 10 ng/ml for paraxanthine and theophylline. The linearity and reproducibility were sufficient for drug monitoring of caffeine and its primary methylxanthines.

Simplified test for determination of drug-oxidizing capacity in rats with chemical-induced liver injury using caffeine and trimethadione as model drugs

We examined the possibility of predicting the extent of hepatic drug-oxidizing capacity by determination of caffeine, trimethadione and their metabolites in three groups of rats with chemically induced liver injuries. Trimethadione (4 mg/kg) and caffeine (10 mg/kg) were simultaneously administered as two probe drugs. In rats with chemically induced liver injuries pretreated with carbon tetrachloride (CCl4: 0.25 ml/kg), alpha-naphthylisothiocyanate (ANIT: 40 mg/kg), or D-galactosamine (GalN: 400 mg/kg), the half-life (t1/2) of caffeine and trimethadione was significantly (P less than 0.01) prolonged compared to those of control groups total body clearance as dramatically reduced (P less than 0.01), whereas apparent volumes of distribution (Vds) were increased in ANIT and GalN groups. In rats with liver damage the production of three metabolites (theobromine, paraxanthine and theophylline) of caffeine as well as the only metabolite (dimethadione) of trimethadione after oral administration of both drugs were significantly decreased compared to those of controls. In rats with liver injuries, total body clearance of caffeine and trimethadione showed a strong correlations (r = 0.99, P less than 0.01); also total body clearance of caffeine correlated well with the ratio of dimethadione/trimethadione after 1, 2, and 4 hr of trimethadione administration (r = 0.93, P less than 0.01; r = 0.97, P less than 0.01 and r = 0.97, P less than 0.01 respectively). Besides, total body clearance of caffeine also correlated well (coefficients ranging from 0.74 to 0.96; P less than 0.01) with the ratio of theobromine/caffeine, paraxanthine/caffeine and theophylline/caffeine after 1, 2 and 4 hr) of caffeine administration.(ABSTRACT TRUNCATED AT 250 WORDS)

Caffeine clearance in cirrhosis. The value of simplified determinations of liver metabolic capacity

Less complex methods of measuring hepatic metabolic capacity are needed. A simplified caffeine clearance test was evaluated in 23 patients with stable alcoholic liver disease. First, saliva caffeine concentrations were measured over a 24-h caffeine-free interval. Clearance was calculated from the rate of elimination of caffeine and an assumed volume of distribution and compared with the results of a formal clearance test using sequential plasma and saliva samples following a 300 mg oral dose. The simplified method was then assessed in 11 hospitalized patients with cirrhosis. Saliva caffeine concentrations remained measurable over the interval of study in 82% of patients. Caffeine clearance as determined by the simplified method did not differ from plasma caffeine clearance after an oral dose. Application of this method was achieved in 11 of 12 patients hospitalized for complications of severe liver disease, and revealed markedly diminished clearance. Thus, caffeine clearance can be accurately estimated in patients with severe liver disease using two or more samples of either saliva or plasma. This simplified determination of caffeine elimination rate provides a more practical assessment of hepatic metabolic capacity than a formal clearance test.

Fasting plasma caffeine level in cirrhotic patients: relation to plasma levels of catecholamines and renin activity

Fasting plasma caffeine concentrations, plasma levels of catecholamines and plasma renin activity were measured in patients with cirrhosis and control patients without hepatic dysfunction. A careful dietary history showed no significant difference in caffeine consumption (mean +/- S.E.) among 46 cirrhotics (86 +/- 7 mg per day) vs. 34 control patients (91 +/- 8 mg per day). Fasting plasma caffeine concentrations, however, were significantly higher (7.68 +/- 1.42 micrograms per ml) in cirrhotics than in controls (1.01 +/- 0.20 micrograms per ml) (p less than 0.01). Fasting plasma caffeine concentrations in cirrhotics varied significantly with Child's criteria, namely Child's A patients (2.06 +/- 0.38 micrograms per ml); Child's B patients (6.92 +/- 1.86 micrograms per ml), and Child's C patients (17.70 +/- 3.65 micrograms per ml) (p less than 0.001). In 44 cirrhotics, fasting plasma caffeine concentrations were compared with plasma levels of catecholamines and plasma renin activity. Plasma epinephrine concentrations were normal; however, plasma norepinephrine concentrations were increased in six cirrhotics, and plasma renin activities were increased in 28 cirrhotics. After a 3-day caffeine abstinence, plasma caffeine concentration and renin activity were significantly decreased (p less than 0.01), and high plasma norepinephrine levels were also decreased in 12 cirrhotics. Plasma caffeine concentration, renin activity and norepinephrine level did not change in a control group of cirrhotics who continued to receive caffeine for 3 days (n = 6). After abstinence from caffeine, the decrease of fasting plasma caffeine concentration correlated well with the decrease of plasma renin activity (r = +0.746, p less than 0.01).(ABSTRACT TRUNCATED AT 250 WORDS)

The pharmacokinetics of caffeine and its dimethylxanthine metabolites in patients with chronic liver disease

1. Serum and salivary concentrations of caffeine (1,3,7-trimethylxanthine) and its dimethylxanthine metabolites were measured in 10 healthy control subjects and in 19 patients with cirrhosis, for up to 96 h following a 400 mg oral caffeine load. 2. Serum and salivary caffeine concentrations correlated significantly (r = 0.954; P less than 0.001) and no significant differences were observed in the pharmacokinetic data derived from the respective concentration-time curves. 3. In the control subjects, basal salivary caffeine concentrations did not exceed 0.4 mg l-1. The median (range) basal salivary caffeine concentrations in patients with compensated cirrhosis (n = 10), 0.2 (0-0.7) mg l-1 and decompensated cirrhosis (n = 9), 0.7 (0-5.8) mg l-1, were not significantly different from control values, although three patients with decompensated cirrhosis had basal salivary caffeine values above 2.0 mg l-1. 4. In the patients with compensated cirrhosis, the median peak salivary caffeine concentration, 10.9 (8.2-16.5) mg l-1 was significantly greater than in controls, 7.1 (4.7-11.8) mg l-1 (P less than 0.01) and the median apparent volume of distribution was significantly reduced, 0.38 (0.19-0.49) vs 0.41 (0.23-0.63) l kg-1 (P less than 0.05).(ABSTRACT TRUNCATED AT 250 WORDS)

Correlation of caffeine elimination and Child's classification in liver cirrhosis

Apparent pharmacokinetic parameters of caffeine elimination from the circulation were determined in 27 patients with histologically confirmed liver cirrhosis, 8 patients with miscellaneous liver disease, and 8 patients with other than liver disease. The usefullness of this quantitative test to assess the severity of liver cirrhosis was compared to the Child-Turcotte or Child-Pugh classification score as well as to the galactose elimination capacity of these patients. Using reversed-phase high pressure liquid chromatography caffeine, paraxanthine, theophylline, and theobromine were analysed in blood plasma collected before and after an oral dose of caffeine. Compared to apparent caffeine pharmacokinetics in patients with normal livers or miscellaneous liver disease, cirrhosis was characterized by a statistically significant reduction in apparent caffeine clearance and prolongation in half-life. The reduced apparent plasma disappearance rate of caffeine in cirrhotics was related to the retarded formation of paraxanthine which was the main metabolite of caffeine in blood plasma both in the absence or presence of liver disease. The apparent caffeine clearance in cirrhosis decreased with increasing Child-Turcotte classification score: Child's class A patients differed significantly from Child's class B or Child's class C patients, whereas the difference between Child's class B and C patients did not reach statistical significance (Wilcoxon's rank test). In addition there was a strong correlation between the Child-Pugh classification score and apparent caffeine clearance (P less than 0.001). However, no correlation existed between Child's classification and galactose elimination capacity. Our data emphasize the value of the Child-Turcotte or Child-Pugh classification in assessing the severity of liver cirrhosis in a simpler and less time-consuming way than using quantitative liver function tests.

Influence of smoking on caffeine elimination in healthy volunteers and in patients with alcoholic liver cirrhosis

The effect of smoking on caffeine elimination was measured in 7 healthy volunteers and in 18 smoking and in 30 nonsmoking patients with alcoholic liver cirrhosis following oral application of 366 mg caffeine. In an intraindividual experiment in smoking health probands, caffeine clearance decreased from 118 +/- 33 to 77 +/- 22 ml per min (p less than 0.05) after abstaining cigarette smoking for 3 weeks. In a control group without liver disease (8 smokers, 15 nonsmokers), we found a caffeine clearance of 114 +/- 40 ml per min in smokers and 64 +/- 20 in nonsmokers (p less than 0.05). Smoking and nonsmoking patients with alcoholic liver cirrhosis did not differ with respect to clinical and laboratory data and hexobarbitone elimination. However, caffeine clearance was 63 +/- 63 ml per min in smoking patients compared to 34 +/- 49 ml per min in nonsmokers (p less than 0.05). Fasting plasma concentrations of caffeine were higher in nonsmokers (5.1 +/- 6.2 micrograms per ml) than in smokers (2.1 +/- 4.5 micrograms per ml, p less than 0.05). We conclude that smoking habits have to be taken into account if caffeine is used as a model compound for measuring quantitative liver function.

The effect of smoking on caffeine elimination: implications for its use as a semiquantitative test of liver function

1. The effects of caffeine ingestion and cigarette smoking on caffeine and antipyrine pharmacokinetics were studied using normal subjects as their own controls before and after cessation of smoking in an attempt to minimize genetic and other environmental influences. 2. Moderate caffeine ingestion had no inducing effect on caffeine or antipyrine clearance. 3. Cessation of cigarette smoking significantly reduced clearance of caffeine and antipyrine. 4. These results demonstrate that cigarette smoking significantly affects caffeine pharmacokinetics and this may contribute to the variable results for caffeine kinetics found in patients with liver disease.

Rapid high-performance liquid chromatography assay for salivary and serum caffeine following an oral load. An indicator of liver function

A rapid isocratic reversed-phase high-performance liquid chromatography (HPLC) system for the quantitative measurement of serum and salivary caffeine is described. The best separation of caffeine from other methylxanthines was achieved by chromatography on an ODS-Hypersil column using a solvent system of 0.1 M ammonium acetate pH 4.6-acetonitrile (85:15, v/v). The effluent was monitored at 280 nm. Caffeine was extracted from diluted serum and saliva samples (10-500 microliter) by adsorption on a small Bond-Elut C18 cartridge and recovered by elution with methanol. Thermospray HPLC-mass spectrometry conditions were optimized to afford a means of directly identifying caffeine in samples. The positive-ion mass spectrum was characterized by an intense protonated molecular ion, MH+, at m/z 195 and negligible fragmentation. When the mass spectrometer was operated in selected ion monitoring mode, caffeine could be detected in less than 1 microliter of serum and saliva at a concentration of 1 microgram/ml. Caffeine (3.5 mg/kg body wt.) was administered orally to healthy adults, children, and newborn infants, and to patients with liver disease. The clearance rate and half-life were determined as a test of liver function. A prolongation in the elimination of caffeine was observed in patients with liver disease and, although there was some overlap in the values obtained for patients with noncirrhotic liver disease and healthy persons, the oral caffeine load test may usefully serve as a dynamic assessment of liver function in the serial follow-up of patients with liver disease.