Clinical Pharmacokinetics

, Volume 43, Issue 11, pp 725–732

Effect of Rifampicin on the Pharmacokinetics of Fluconazole in Patients with AIDS


    • Department of Clinical Pharmacy, Faculty of Pharmaceutical SciencesChulalongkorn University
  • Nawarat Thanompuangseree
    • Department of PharmacySuratthani Hospital
  • Somsit Tansuphaswadikul
    • Department of MedicineBamrajnaradura Hospital
Original Research Article

DOI: 10.2165/00003088-200443110-00003

Cite this article as:
Ayudhya, D.P.N., Thanompuangseree, N. & Tansuphaswadikul, S. Clin Pharmacokinet (2004) 43: 725. doi:10.2165/00003088-200443110-00003



To study the effect of rifampicin on the pharmacokinetics of fluconazole and on clinical outcomes of fluconazole treatment in patients with AIDS-related cryptococcal meningitis.


Forty Thai patients with AIDS and cryptococcal meningitis, of whom 20 had been receiving oral rifampicin for at least 2 weeks to treat tuberculosis.


Patients were treated for cryptococcal meningitis with amphotericin 0.7 mg/kg/day for 14 days followed by fluconazole 400 mg/day, which was reduced to 200 mg/day once culture of cerebrospinal fluid (CSF) became negative. Patients with tuberculosis received oral rifampicin 600 mg/day at night. Blood samples were collected from the first 12 patients in each group and pharmacokinetic parameters for fluconazole were calculated. CSF samples were collected by lumbar puncture and monitored for eradication of Cryptococcus neoformans.


Concomitant administration of rifampicin with fluconazole resulted in significant changes in the pharmacokinetic parameters of fluconazole, including a 39% increase in elimination rate constant, 28% shorter elimination half-life, 22% decrease in area under the concentration-time curve, 17% decrease in maximum concentration and 30% increase in clearance (p < 0.05). Different fluconazole regimens did not affect the extent of change in the elimination rate constant. Although serum concentrations of fluconazole during the time that patients received rifampicin concomitantly with fluconazole 200 mg/day were generally lower than the minimum inhibitory concentration for C. neoformans, there were no significant differences in clinical outcomes between the two groups to date (p = 0.792).


Coadministration of rifampicin with fluconazole caused significant changes in the pharmacokinetic parameters of fluconazole. Long-term monitoring for recurrence rates of cryptococcal meningitis is required to assess the clinical significance of this interaction.

Fluconazole is a water-soluble fluorine-substituted bis-triazole that has been shown to be effective against a variety of fungal infections, including cryptococcal meningitis. The presence of the thiazole rings may contribute to the resistance of fluconazole to first-pass metabolism and to the low lipophilicity and protein binding of the drug.[1] Greater than 90% of an oral dose of fluconazole can be detected in the systemic circulation.[1,2] The mean (± SD) total clearance of fluconazole is 1.17 ± 0.28 L/h, and renal clearance accounts for 78% of plasma clearance. Plasma half-life is approximately 31–34 hours. Extensive tubular reabsorption explains the long half-life. The unchanged fraction of the drug in urine is in the range of 64–90%. Only 11.4% of an administered dose is recovered in urine as metabolites.[1,2]

However, drugs that induce hepatic microsomal enzymes have been shown to accelerate the metabolism of fluconazole. Rifampicin, a potent inducer of cytochrome P450 enzymes, is still one of the most valuable drugs for the treatment of tuberculosis.[3] Occasionally, tuberculosis occurs simultaneously with cryptococcal meningitis in AIDS patients and concomitant administration of rifampicin and fluconazole may be required. The magnitude of the interaction between these two drugs may have a significant impact on the therapeutic outcomes of these patients. Coker et al.[4] reported a clinical relapse of cryptococcal meningitis in three patients associated with the concurrent administration of fluconazole and rifampicin. Apseloff et al.[5] and Lazar and Wilner[6] reported that concomitant administration of fluconazole with rifampicin in healthy volunteers resulted in a 25% faster elimination (i.e. higher elimination rate constant, ke), 22% shorter elimination half-life (t½β) and 23% decrease in the total area under the concentration-time curve (AUC) of fluconazole.[5,6] Nicolau et al.[7] studied the influence of rifampicin on the pharmacokinetics of fluconazole in a small number of critically ill patients and reported an interaction of greater significance than was found in volunteers, with the potential for poor antifungal treatment outcomes. Administration of multiple doses of fluconazole leads to an approximately 2.5-fold increase in peak plasma concentration compared with that achieved after a single dose.[8]

However, no complete study has reported the effect of rifampicin on fluconazole pharmacokinetics after multiple doses of fluconazole in patients and on the clinical outcomes with fluconazole in patients receiving these medications concurrently. Therefore, this clinical study was designed to compare the pharmacokinetic parameters of fluconazole in AIDS patients with cryptococcal meningitis treated with fluconazole alone or receiving concomitant treatment with rifampicin, along with clinical outcome expressed as time to negative cerebrospinal fluid (CSF) culture for Cryptococcus neoformans (‘conversion time’). In addition to the initial determination of pharmacokinetic parameters after 8 days of treatment with fluconazole 400 mg/day, ke was also determined after 15 days of treatment and during maintenance therapy with fluconazole 200 mg/day so as to examine the effect of time and dosage of fluconazole on the extent of interaction.

Patients and Methods

Study Design

This study was conducted from October 2000 to July 2001 at Bamrajnaradura Hospital, Nonthaburi, Thailand. The study protocol was reviewed and approved by the institutional ethical review board.

In order to obtain 80% power to detect a difference in conversion time of 2 weeks or longer, at least 20 patients per group were required.[9] Forty AIDS patients with cryptococcal meningitis who met the inclusion criteria participated in this study. Twenty of them had been receiving oral rifampicin 600 mg/day at night for more than 2 weeks to treat tuberculosis; the other group of 20 were not receiving rifampicin. The subjects were 18 years or older, consented to enrol in this study and received no concomitant therapy that might change the pharmacokinetic properties of fluconazole except for rifampicin. They had no known allergy to polyene or azole antifungals. The patients included in the no-rifampicin group had never received rifampicin or had discontinued rifampicin at least 3 months before the start of this study. All patients received the standard treatment for cryptococcal meningitis: amphotericin 0.7 mg/kg/day for 14 days followed by fluconazole 400mg once daily.

To detect a difference of 25% in fluconazole AUC between the two groups with an α level of 0.05 and 80% power, the analysis would require at least 11 patients in each group.[9] Therefore, blood samples were collected from the first 12 patients enrolled in each group. The sample collections were conducted in three periods. In period I, on day 8 of fluconazole 400 mg/day, 5–10mL blood samples were collected before taking fluconazole and 1, 2, 4, 8, 12 and 24 hours after taking fluconazole. In order to study the effect of coadministration interval on the extent of interaction, two blood samples were again collected before and 4 hours after fluconazole administration from the same 12 patients of each group at period II (on day 15 of fluconazole 400 mg/day). Finally, to study the effect of fluconazole dosage, after the patients showed negative CSF culture their fluconazole dosage was reduced to 200 mg/day. Two blood samples were collected at the same times in period III (on day 8 of fluconazole 200 mg/day).

All blood samples were allowed to clot at room temperature, centrifuged at 2400 rev./min for 6 minutes, and then serum samples were separated and kept at −20°C until analysed.

In order to compare the efficacy of fluconazole treatment for cryptococcal meningitis between these two groups, after 1, 2, 4, 6 and 8 weeks of treatment with fluconazole 400 mg/day (or until culture for C. neoformans had become negative), CSF samples were collected by lumbar puncture, monitored for conversion to negative culture and analysed for protein, glucose, white blood cells and number of fungal cells positive for India ink staining.

Drug Assay

Concentrations of fluconazole in serum samples were quantified by high performance liquid chromatography (HPLC). Phenacetin was used as an internal standard. The extraction process was modified from the method reported by Foulds et al.[10] The mobile phase was a mixture of methanol and 10 mmol/L phosphate buffer (pH 7), 1 : 1 v/v, at a flow rate of 1 mL/min. The column was Adsorbosphere C18 (Waters Corporation, Milford, MA, USA). The column effluent was monitored by UV detection at 260nm. The detection limit of fluconazole was 1 mg/L. The intra- and interday coefficients of variation (CV) were <15% (ranging from 1.5% at a concentration of 40 mg/L to 12.1% at a concentration of 1 mg/L).

Data Analysis

A complete concentration-time profile over one administration interval (24 hours) was generated after a 400mg dose of fluconazole, using seven concentrations at 0, 1, 2, 4, 8, 12 and 24 hours. For this period, the pharmacokinetic parameters, including maximum serum concentration (Cmax), time to reach Cmax (tmax), AUC24, absorption rate constant (ka), ke and t½β, were derived by using the pharmacokinetics program RSTRIP II version 2.0, which performs compartmental modelling and kinetic analysis.[11] Apparent clearance (CL/F) was derived from dose/AUC24 and apparent volume of distribution (Vd/F) was calculated as CL/F divided by ke.[12]

During periods II and III, ke was determined from the slope of the semilogarithmic concentration-time curve derived from the two blood samples obtained at trough and at 4 hours after drug administration.

The data were analysed using descriptive statistics. Comparisons were performed by analysis of variance (ANOVA) and/or Student’s t-test for continuous variables and by the chi-squared test for categorical variables.


Demographic Data

Of the 40 subjects with AIDS-associated cryptococcal meningitis, 28 (70%) were male. Their ages ranged from 21 to 48 years (mean ± SD, 31.68 ± 6.31 years). Baseline laboratory results indicated that the mean haematocrit was 32.04 ± 5.94%, lower than the normal level in patients with AIDS. White blood cell count was within the normal range with a mean value of 4600 ± 1998 cells/mm3. Both blood urea nitrogen (BUN) and creatinine were also within the normal range, with mean values of 13.65 ± 6.21 mg/dL and 1.00 ± 0.33 mg/dL, respectively.

Pharmacokinetic Interaction

For period I, the mean (± SD) concentrations of fluconazole at various times in both groups are shown in table I. The concentrations between the groups were significantly different at all times (p < 0.05).
Table I

Comparison of fluconazole concentrations between the two groups at various times after administration of fluconazole 400mg at steady state (period I, day 8 after starting fluconazole 400 mg/day). Values are means ± SD (range) for 12 patients in each group

Data analysis with RSTRIP II indicated that the kinetic profiles of fluconazole could be explained by two exponential terms. The pharmacokinetic parameters for both groups and their comparisons are shown in table II. The concomitant administration of rifampicin with fluconazole resulted in significant changes in the pharmacokinetic parameters of fluconazole, including 39% increase in ke, 28% decrease in t½β, 22% decrease in AUC24, 17% decrease in Cmax and 30% increase in CL/F. The pharmacokinetic parameters reflecting the elimination process, i.e. ke, t½β and CL/F, were significantly different between these two groups (p < 0.001). Therefore, AUC24 and Cmax of fluconazole were also significantly lower in the rifampicin group (p = 0.001 and p = 0.009, respectively). The absorption process, as represented by ka, showed no significant difference between the two groups (p = 0.816). All patients in both groups displayed a tmax within 4 hours after the administration of fluconazole, and the values were not significantly different between groups. Comparison of Vd/F found no significant difference between groups (p = 0.394).
Table II

Comparison of pharmacokinetic parameters for fluconazole between the two groups. Values are means ± SD for 12 patients in each group

Effect of Duration of Coadministration and Dosage on the Pharmacokinetic Interaction

As well as the complete pharmacokinetic profile obtained during period I, two blood samples were also collected at trough (time zero) and at 4 hours after fluconazole administration in periods II and III. Table III shows the concentrations of fluconazole and their corresponding elimination rate constants during each period from both groups.
Table III

Comparison of fluconazole concentrations and elimination rate constants (ke) between the two groups during three different periods. Values are means ± SD for 12 patients in each group

There were no significant differences between fluconazole concentrations in periods I and II. As expected, fluconazole concentrations during period III were approximately one-half of those during periods I and II, since the dosage of fluconazole had been decreased by one-half. In the rifampicin group in period III, concentrations at trough and at 4 hours (approximately Cmax) were lower than in the fluconazole-alone group and close to reported values for the 50% minimum inhibitory concentration (MIC50) of C. neoformans.[13,14] Reported values from the microbiology laboratory at Bamrajnaradura Hospital, Nonthaburi, Thailand, for MIC90 and MIC50 of C. neoformans are approximately 16 and 8 mg/L, respectively (unpublished data). The ke values determined during any of these three periods were not significantly different within the same group. However, ke was consistently increased in the coadministration group compared with the fluconazole-alone group, with increases of 39%, 34% and 37% during periods I, II and III, respectively.

Effect on Clinical Efficacy

Rate of Conversion to Negative CSF Culture

CSF conversion rates are shown in table IV. All patients had negative CSF culture within 6 weeks. There were no significant differences in conversion rates of CSF culture between groups (p = 0.792 by chi-squared test).
Table IV

Comparison of rates of cerebrospinal fluid conversion to negative culture. There was no significant difference between rates (p = 0.792 by chi-squared test)

CSF Components

Comparisons of CSF components between groups are shown in table V. The CSF components of most patients returned to the normal range within 6 weeks of therapy. There were no apparent differences in CSF components between the two groups.
Table V

Comparison of cerebrospinal fluid components. Values are means ± SD for the indicated numbers of patients


The pharmacokinetic parameters of fluconazole presented in this study were derived from steady-state concentrations of fluconazole, in contrast with previous studies, which were performed with a single dose of fluconazole. However, the changes in pharmacokinetic parameters of fluconazole caused by coadministration with rifampicin were not greatly different from those previously reported after single-dose studies in healthy volunteers. The decrements in AUC24 were quite similar, and the increments in ke were slightly higher, after multiple doses.

ke and CL/F were significantly greater in the drug coadministration group, resulting in shorter t½β. These values indicate that the interaction affects the elimination process. Rifampicin most probably enhances the elimination of fluconazole by inducing cytochrome P450 in the liver.[6,15] The pharmacokinetic parameters involved in absorption, i.e. ka and tmax, showed no significant difference between the groups, indicating that the absorption of fluconazole was not affected by concomitant administration of rifampicin.

Different durations of coadministration (periods I, II or III) resulted in similar values for ke within the groups, indicating unchanged increments in ke between the two groups. This implies that enzyme induction by rifampicin had reached its peak during period I and was not affected by continued administration of fluconazole. Different dosages of fluconazole (periods I + II vs period III) also resulted in the same extent of interaction, as shown by similar increases in ke between the two groups. This implies that the metabolism of fluconazole is a first-order reaction and that no saturation of the process occurs in the dosage range used in this study.

There were no obvious differences in CSF components between the two groups. The CSF cultures of all patients became negative within 6 weeks, with median times of 4 weeks in both groups. This median time of conversion is similar to that reported in other patients receiving the same regimen for treatment of cryptococcal meningitis, i.e. amphotericin for 2 weeks followed by fluconazole 400 mg/day.[1619] No significant difference in conversion rates of CSF cultures could be found between the two groups. These results imply that the interaction did not significantly affect the clinical efficacy of this regimen, perhaps because of the use of amphotericin in the first 2 weeks and the fact that fluconazole 400 mg/day still yields concentrations above the MIC of the infecting organism, even with coadministration of rifampicin. However, use of fluconazole 200 mg/day for prophylaxis of cryptococcal meningitis resulted in concentrations of fluconazole lower than the MIC for most of the time in the fluconazole plus rifampicin group. Long-term monitoring for recurrence rates of cryptococcal meningitis is required to assess the significance of this observation.


Coadministration of rifampicin with fluconazole caused significant changes in the pharmacokinetic parameters of fluconazole, including ke, t½β, CL/F, Cmax and AUC. Rifampicin is an inducer of hepatic microsomal enzymes and accelerates the elimination process of fluconazole.

In this study, the drug interaction did not cause any significant changes in the early clinical outcomes of patients with cryptococcal meningitis, as shown by no significant difference in conversion rates of CSF cultures whether or not rifampicin was concomitantly administered with fluconazole. However, when fluconazole 200 mg/day was used as prophylaxis for cryptococcal meningitis, serum concentrations of fluconazole were mostly lower than the MIC when concomitantly administered with rifampicin. This could lead to poor clinical outcomes and may increase the recurrence rate. Consequently, long-term studies are required of the clinical response to fluconazole in fungal infections when rifampicin is coadministered. For treatment of serious infections, a 30% increase in the dosage of fluconazole, especially during the 200 mg/day period, might be considered in the setting of concomitant administration of rifampicin.


The authors are grateful to Chulalongkorn University for financial support, Siam Pharmaceutical Co. Ltd for supplying fluconazole, the Department of Medical Science and the Ministry of Health for providing fluconazole standard and internal standard (phenacetin), and to all nurses at the seventh building and staff at the Laboratory of Bamrajnaradura Hospital for their helpfulness.

Copyright information

© Adis Data Information BV 2004