Pharmacokinetics of Rifampin Under Fasting Conditions, With Food, and With Antacids – )
Charles A. Peloquin
Study objectives: Determine the intrasubject and intersubject variability in, and the effects of food or antacids on, the pharmacokinetics of rifampin (RIF).
Design: Randomized, four-period crossover phase I study.
Subjects: Fourteen healthy male and female volunteers.
Interventions: Subjects ingested single doses of RIF, 600 mg, under fasting conditions twice, with a high-fat meal, and with aluminum-magnesium antacid. They also received standard doses of isoniazid, pyrazinamide, and ethambutol.
Measurements and main results: Serum was collected for 48 h and assayed by high-pressure liquid chromatography. Data were analyzed using noncompartmental methods and a compartmental analysis using nonparametric expectation maximization. Both fasting conditions produced similar results: a mean RIF maximal serum concentration (Cmax) of 10.54 [+ or -] 3.18 [micro]g/mL, the time at which it occurred (Tmax) of 2.42 [+ or -] 1.32 h, and the area under the curve from time zero to infinity ([AUC.sub.0-[infinity]]) of 57.15 [+ or -] 13.41 [micro]g [multiplied by] h/mL. These findings are similar to those reported previously. Antacids did not alter these parameters (Cmax of 10.89 [+ or -] 5.22 [micro]g/mL, Tmax of 2.36 [+ or -] 1.28 h, and [AUC.sub.0-[infinity]] of 58.37 [+ or -] 18.49 [micro]g [multiplied by] h/mL). In contrast, the Food and Drug Administration high-fat meal reduced RIF Cmax by 36% (7.27 [+ or -] 2.29 [micro]g/mL), nearly doubled Tmax (4.43 [+ or -] 1.09 h), but reduced [AUC.sub.0-[infinity]] by only 6% (55.20 [+ or -] 14.48 [micro]g [multiplied by] h/mL).
Conclusions: These changes in Cmax, Tmax, and [AUC.sub.0-[infinity] can be avoided by giving RIF on an empty stomach whenever possible. (CHEST 1999; 115:12-18)
Key words: antacids; bioavailability; food; Mycobacterium avium complex; Mycobacterium tuberculosis; pharmacokinetics; rifampin
Abbreviations: ANOVA = analysis of variance; AUC = area under the curve; [AUC.sub.0-[infinity]] = area under curve from time zero to infinity; C = clearance; CI = confidence interval; ClCr = creatine clearance; Clr = renal clearance; Cmax = maximal serum concentration; CV = coefficient of variation; EMB = ethambutol; HPLC = high-pressure liquid chromatography; INH = isoniazid; K = elimination rate construct; ka = absorption rate constant; MIC = minimal inhibitory concentration; NPEM = nonparametric expectation maximization; PZA = pyrazinamide; QC = quality control; RIF = rifampin; Tmax = time at which maximal serum concentration occurred; V = volume of distribution
Rifampin (RIF) is one of the three most important drugs used for the treatment of tuberculosis.[1] The standard “short-course” treatment of tuberculosis consists of isoniazid (INH), RIF, pyrazinamide (PZA), plus either ethambutol (EMB) or streptomycin until susceptibility data are available.[1] Limited information exists regarding the pharmacokinetics of these drugs, in particular population modeling and the effects of food or antacids on the GI absorption of the drug.[2-6]
We examined the pharmacokinetics of RIF in healthy volunteers under fasting conditions (two replicates), with food, and with an aluminum/magnesium hydroxide antacid. This study describes the serum concentrations and the pharmacokinetic behavior under optimal conditions, and can be used as benchmarks for comparison with samples obtained in other clinical settings.
MATERIALS AND METHODS
We conduced a four-period, randomized crossover study of RIF. The study protocol followed the guidelines of the Helsinki Declaration of 1975 and its amendments, and was approved by the institutional review board at Millard Fillmore Hospital, Buffalo, NY. Written informed consent was obtained from each subject before the study. Sixteen normal, healthy female and male volunteers were scheduled to participate. Subjects were eligible to participate if they were [is greater than] 18 years of age and were determined to be in good health. Health was assessed by history, physical examination, and laboratory studies, including serum chemistries, CBC count with differential, and 12-lead ECG. Individuals were excluded if they had histories of a major disease of the kidneys or an estimated creatinine clearance [is less than] 50 mL/min, the liver (transaminases, alkaline phosphatase, or bilirubin [is greater than or equal to] 2 times normal), the cardiovascular system (New York class I to IV heart failure), or a hematocrit [is less than] 36% at screening. They also were excluded if they had known GI diseases that might affect the absorption of the drugs, known positive HIV serology, AIDS, or histories of adverse reactions to INH, RIF, PZA, EMB, or related drugs. They also were excluded if they weighed [is greater than] 130% of ideal body weight, were pregnant or nursing, or donated blood within 30 days prior to study.7 The subjects agreed to refrain from the use of other prescription or nonprescription drugs (including vitamins) and alcohol during the entire study period. Women who were taking oral contraceptives at the start of the study were allowed to continue taking these during the study. They were required to agree to use additional contraceptive methods during the study period and for a week after the last dose of RIF. At the conclusion of the study, each subject underwent a brief physical examination and had blood drawn for serum chemistry and hematology study, and female subjects had repeat pregnancy testing.
Experimental Design
Sixteen subjects were randomized in four blocks of four subjects. The four treatments were fasting conditions (twice, to determine the intrasubject variability), with a high-fat meal, and with aluminum-magnesium antacid. They were housed at the study center from 10 h before to 24 h after dosing, and returned for the 36-h and 48-h collections. After receiving a light snack prior to 11 PM, they fasted overnight. For three of the treatments, they continued to fast for 4 h after the dose. On one of these three fasting occasions, they also took 30 mL of aluminum/ magnesium hydroxide (Mylanta) 9 h before dosing, at the time of dose, after meals, and at bedtime postdose. For the fourth treatment, they consumed the standard Food and Drug Administration high-fat breakfast beginning 0.25 h before dosing. This meal consisted of 8 oz of whole milk, two scrambled eggs, two strips of bacon, two slices of toast with two butter pads, and one hash brown potato patty. The meal provided an estimated 53 g of carbohydrate, 33 g of protein, and 51 g of fat, for 792 Kcal, 57% as fat. Subjects received single oral doses of RIF, 600 mg. They also received 300 mg INH, 30 mg/kg PZA (median, 2,386 mg), and 25 mg/kg EMB (median, 1,950 mg). Doses for all treatment periods were based on the subjects’ prestudy weights. The subjects were allowed to ingest water ad libitum after the doses were given, and identical, nutritionally balanced meals were provided to all subjects during the remainder of the study period. There was a 14-day washout between each study period.
Sample Collection
A 20-gauge angiocatheter was inserted into a forearm vein for the collection of blood samples, and was maintained patent using a dilute heparin solution (10 to 15 U/mL). Two milliliters of blood was withdrawn and discarded prior to collecting each blood sample (12 mL) into plain red-top vacuum tubes. Serial blood samples for serum drug concentration analyses were collected at 0, 0.25, 0.5, 0.75, 1, 1.5, 2, 2.5, 3, 4, 6, 8, 10, 12, 16, 24, 36, and 48 h after the doses. Samples were allowed to clot for 30 min, then centrifuged at 2,500 to 3,000g for 10 min. Serum samples were then harvested and frozen at [is less than or equal to] -70 [degrees] C for 5 months until assay. We have determined previously that RIF is stable in human serum for [is greater than] 24 h at room temperature, and it is stable for periods of [is greater than] 1 year when frozen at [is less than or equal to] -70 [degrees] C.[8]
Urine samples were collected within 30 min of dosing (baseline). Subsequently, all urine was collected from 0 to 12 h and from 12 to 24 h. Subjects were asked to void near the end of each collection period. Samples were kept refrigerated during the period of collection. The total volume was measured at the end of the collection period, and 10-mL aliquots from each collection were frozen at -70 [degrees] C until assay.
Sample Analysis
All high-pressure liquid chromatography (HPLC) assays were performed using a validated HPLC assay on a pump (Waters 510 pump; Milford, MA) and model 680 gradient controller with a solvent select valve, a model 8875 fixed-volume autosampler (Spectra Physics; San Jose, CA), a model 486 ultraviolet detector (Waters), a computer (Macintosh IIci; Apple Computers Inc; Cupertino, CA), and an HPLC data management system (Rainin Dynamax; Woburn, MA). The six-point standard curves for the RIF ranged from 0.5 to 50 [micro]g/mL, with linearity extending well above this range. The absolute recovery of RIF from serum was 95.5%, as determined by comparing peak height counts across four serum curves to an unextracted solvent curve. The within-day precision (percent coefficient of variation [CV]) of validation quality control (QC) samples was 2.4 to 4.6%, and the overall validation precision was 6.3 to 7.1%. QC sample concentrations were 26, 8, and 3 [micro]g/mL. The urine method standard curves for RIF ranged from 0.5 to 200 [micro]g/mL, with similar recovery and reproducibility. The assay error pattern was determined from serum standard curve samples assayed over the course of the validation. A second-order polynomial was fit to the plot of the QC standard deviations (Y) vs their means (X). The assay error pattern used for the subsequent pharmacokinetic analysis was y = 0.0350 + 0.0046x – 0.0001[x.sup.2], [R.sup.2] = 0.992.[9]
Pharmacokinetic Analysis
Serum concentrations below the quantification lower limit were treated as zeros in averaging the concentrations at a given collection time. Data were initially analyzed using noncompartmental methods. The observed maximal serum concentration (Cmax) and the time at which it occurred (Tmax) were determined for each subject by inspection of the serum concentration-vs-time graphs. The area under the serum concentration-vs-time curve (AUC) from time zero to the time of the last quantifiable concentration ([AUC.sub.0-t*]) was determined by the linear trapezoidal rule. The last quantifiable concentration was denoted C*. The AUC from time zero to infinity ([AUC.sub.0-[infinity]]) was determined as [AUC.sub.0-t*] + C*/K, with the elimination rate constant (K) determined using nonparametric expectation maximization (NPEM) (see below). The potential for accumulation of RIF was evaluated using the principle of superposition.[10] The accumulation of RIF with eight daily doses was simulated using the median serum concentration data from 0 to 24 h (first fasting treatment), and extrapolated from 24 h to day 8 using the median NPEM K.
Population pharmacokinetic models were made using software (NPEM2, USC*PACK v. 10.6; Laboratory of Applied Pharmacokinetics, USC; Los Angeles, CA).[11] F, the fraction of the dose absorbed, was arbitrarily fixed at 1. Based on our previous work with RIF, and based on the log-linear decline of serum concentrations post Cmax, a one-compartment open model with first-order absorption and elimination was used, weighted by the inverse assay variance error pattern described above. NPEM obtained the joint probability density functions of the final pharmacokinetic parameters. Three parameters were fit in the initial analyses (absorption rate constant [ka], volume of distribution [V], and clearance [C]). The loglikelihood criterion was used to determine the best fit among candidate models. A second analysis was done using ka, V, and K for fasting 1 state to verify the results and to address problems that may arise with one set of parameters but not the other, such as the “flip-flop” problem of structural identifiability. The analyses were refined by restricting the initial estimates of V to [is greater than or equal to] 0.4 L/kg. In addition, individual subject Bayesian posterior parameter joint densities were estimated starting from the population parameter joint density, and continuing to analyze each subject’s individual data to obtain the individual parameter joint densities (post hoc analysis). These values allow for the calculation of rate constants and half-lives across all subjects. The absorption and elimination t1/2’s were calculated as ln(2)/ka and ln(2)/K, respectively.
D-optimal sampling time analysis was performed using software (ADAPT II) and the NPEM2 parameter estimates.[12] The assay error pattern described above was used. Sampling times were analyzed using the parameters ka, V, and K over the period 0.5 to 24.0 h, with various initial sampling times and sampling time constraints. A two-sample strategy (achieved by fixing ka and fitting only V and K) and a three-sample strategy (achieved by fitting all three parameters) were tested. In addition, an analysis of Cmax was performed over the period 0.5 to 4.0 h, calculating the maximum, median, and minimum percentage for the measured concentration divided by Cmax.
Creatinine clearance (ClCr) was calculated by the method of Cockroft and Gault.[13] The amount of RIF recovered in the urine was calculated as the measured volume of urine multiplied by the corresponding RIF concentration. Total recovery (mg) was calculated as the sum of the recoveries from the collection periods 0 to 12 h and 12 to 24 h, and the percent dose recovered was calculated as total recovery divided by dose multiplied by 100%. Renal clearance (Clr) was calculated as total recovery divided by [AUC.sub.0-24].
Statistical Analysis
Data analysis was performed using software (JMP version 3.1.6; SAS Institute; Cary, NC), with supplemental analyses done with other software (Excel version 4.0; Microsoft; Seattle, WA). Frequency distributions (JMP) included plots of the data, distribution curves to test for normality, parametric and nonparametric measures of central tendency and dispersion, as well as the Shapiro-Wilk W test for normality. Means are reported [+ or -] the SD. The percent CV was calculated as (SD/mean) multiplied by 100%. Differences among the treatment groups were determined using an analysis of variance (ANOVA) model that tested differences based on period, treatment, sequence, and subject (sequence). Pairwise differences across the four treatments were evaluated using individual linear contrasts. Bioequivalence criteria were tested according to the 1992 Food and Drug Administration guidelines.[14] Cmax and [AUC.sub.0-[infinity]] were log transformation, and were analyzed using the ANOVA model described above. Mean estimates and SEs were obtained from the linear contrasts, and these were used to calculate the geometric means and the lower and upper 90% confidence limits. Comparison treatments were considered bioequivalent to the reference treatment (fasting treatment 2) if the comparison parameter 90% lower limit was [is greater than or equal to] 80% and the upper limit was [is less than or equal to] 125%.
Correlation analysis (JMP) was performed across the subject and outcome variables using nonparametric techniques (Spearman rho). The dependence of outcome variables (the pharmacokinetic parameters) on subject characteristics (demographic data such as age, weight, ClCr, etc) was determined using Y by X analyses, one parameter at a time (JMP). Subsequently, models with multiple X variables were constructed using forward addition and backward deletion. Differences between groups (JMP) were determined using the analysis of log likelihood with the Pearson [chi square] statistic (contingency tables), Student’s t test, or ANOVA (three or more groups) of normally distributed data (one-way layouts and linear regression), the Wilcoxon or the Kruskal-Wallis tests (rank sums) for nonnormally distributed data (one-way layouts), and the whole-model test table with [chi square] statistic (logistic regression). Differences between groups or correlations between parameters and covariates were considered statistically significant at p [is less than] 0.05.
RESULTS
Fourteen subjects completed all four treatments: six white women, three black men, and five white men. The remaining subjects dropped out for personal reasons. The mean age was 39.1 [+ or -] 7.4 years and the mean weight was 79.3 [+ or -] 13.2 kg, representing 89 to 130% of ideal body weight. The subjects received RIF dose of 600 mg daily (7.95 [+ or -] 1.45 mg/kg). All subjects denied the use of any nonprotocol medications during the study period. ClCr estimates were a mean 103 [+ or -] 25 mL/min. There was no RIF in any time zero serum or urine samples. The absorption characteristics for RIF with the four treatments are described in Table 1, and the corresponding mean RIF serum concentration-vs-time profiles across the 14 subjects are shown in Figure 1. Under fasting conditions, variability in absorption of RIF was small (Table 1) and the individual results quite reproducible (Fig 2). the mean BIF Cmax was unaffected by antacids, but significantly decreased by food (-36%) (p [is less than] 0.0017). The mean BIF Tmax was unaffected by antacids, while substantially increased by food (+103%) (p [is less than] 0.0001). The Tmax was [is greater than or equal to] 3 h for 12 of 14 subjects when fed, compared with 2 subjects in the other groups. The mean RIF [AUC.sub.0-infinity] was unaffected by antacids, and showed a small reduction with food (-6%) (p = 0.76). Using the bioequivalence criteria, food reduced the Cmax beyond the lower bounds (90% confidence interval [CI], 55.8 to 78.9,%), but not the [AUC.sub.0-[infinity]] (90% CI, 83.5 to 104.4%); and antacids did not significantly affect the Cmax (90% CI, 80.3 to 119,.5%) or [AUC.sub.0-[infinity]] (90% CI, 87.7 to 109.6%).
Table 1–The Absorption Characteristics for RIF for
14 Subjects Across the Four Treatments
Cmax, [micro]g/mL Tmax, h
Group Mean %CV Mean %CV
Fast 1 10.54 30 2.43 55
Fast 2 11.32 27 2.18 66
Antacid 10.89 48 2.36 54
Fed 7.27 31 4.43 25
[AUC.sub.0-C*] ([micro]g
[multiplied by] h/mL)
Group Mean %CV
Fast 1 54.69 24
Fast 2 57.09 29
Antacid 55.01 32
Fed 50.97 28
[AUC.sub.0-[infinity]]
([micro]g [multiplied by] h/mL)
Group Mean %CV
Fast 1 57.15 23
Fast 2 58.98 27
Antacid 58.37 32
Fed 55.20 26
[Figure 1-2 ILLUSTRATION OMITTED]
The median 24-h concentration was zero; therefore, simulation of multiple daily doses did not show any accumulation. Consistent with our previous findings, the serum BIF concentration data and NPEM models revealed two groups of RIF absorbers: smooth absorbers and those who had early low concentrations followed by rapid absorption (low absorbers).[5] Most smooth absorbers (8 of 14) had no measurable concentration at 0.5 h postdose. In contrast, the six low absorbers had concentrations of RIF of [is less than] 2.5 mg/mL at 0.5 h postdose and much larger concentrations ([is greater than] 9 [micro]g/mL) at [is greater than] 1 h. Review of the first and second measurable concentrations in plasma (C1 and C2, respectively) revealed that smooth absorbers had C2/C1 ratios of [is less than] 4, with a median ratio of 1.46 and a range of 1.01 to 3.27. Low absorbers had C2/C1 ratios of [is greater than] 4, with a median ratio of 8.59 and a range of 4.14 to 15.20. Tmax did not occur significantly later in low absorbers when compared with the smooth absorbers (p = 0.544; Wilcoxon).
During the analysis of the first fasting, fed, and antacid treatments, the NPEM program assigned very low volumes of distribution to the low absorbers, specifically those patients who had large C1 values. To overcome this problem, the data were reanalyzed while restricting the lower range of V to 0.4 L/kg. This value was the lowest within the post hoc NPEM2 individual parameter estimates for smooth absorbers, and was consistent with the lowest values seen in a previous study.[5] The problem of very small values for V was not encountered with the second fasting data set. Reanalysis of the second fasting data set while restricting the lower range of V to 0.4 L/kg produced little change in the parameter estimates.
Table 2 shows the parameter estimates for RIF following the 600-mg dose as calculated using post hoc NPEM2 individual parameter estimates (second fasting treatment, unrestricted V). The various parameter estimates were not significantly different across the fasting and antacid treatments. However, the V was larger (p = 0.0015) and K was longer (p = 0.0017) in the fed group.
Table 2–NPEM2 Post Hoc Individual Parameter Estimates for 14 Subjects From the Second Fasting Treatment
Parameter Median Range
Ka, 1/h 0.33 0.21-5.22
abs [t.sub.1/2] h 2.10 0.13-3.30
V, L/kg 0.74 0.38-1.15
Cl, L/h 13.9 8.70-18.1
Cl, L/h/kg 0.18 0.10-0.27
K, 1/h 0.18 0.12-0.30
[t.sub.1/2], h 3.95 2.34-5.98The D-optimal sampling times for all subjects over the period 0.5 to 24.0 h were 3.1 h and 14.9 h for the two-sample strategy, and 2.07, 9.47, and 22.32 h for the three-sample strategy if the entire interval was available for sampling. D-optimal sampling times were not affected by changes in the initial sampling times chosen. Table 3 shows that the samples collected from 2 to 3 h came closest to Cmax for greatest number of the 14 subjects.
Table 3–Concentrations Collected From 0.25 to 4.0 h
Expressed as a Percentage of Cmax
Time Postdose, h Highest % Median % Lowest %
0.25 0 0 0
0.50 9 0 0
0.75 76 13 0
1.00 100 18 0
1.50 100 56 0
2.00 100 74 6
2.50 100 69 14
3.00 100 77 55
4.00 100 64 37
The recovery of RIF in the urine is shown in Table 4. Most of the urinary excretion of RIF occurred during the first 12 h postdose, with about 10% of the dose recovered unchanged in the urine over 24 h. The total recovery of unchanged RIF decreased in the fed group when compared with the fasting state (p = 0.002). The percent of the dose recovered unchanged in the urine was also decreased in the fed group; however, the RIF Clr was not different across the four treatments.
Table 4–Recovery of RIF in Urine Over 24 h
Parameter Median Range
Xu 0-12 h, mg 59.6 37.1-81.7
Xu 12-24 h, mg 3.73 0.99-12.7
Xu total, mg 65.5 41.7-82.7
% dose recovered 10.9 6.95-13.8
Clr, L/h 1.14 0.79-2.25
The RIF results were analyzed with JMP, and the nonparametric measures of association are reported (Spearman rho). There was a trend for older subjects to have lower Cmax and later Tmax values, and they were more likely to have small C2/C1 ratios (ie, smooth absorbers) (r = -0.6637, p = 0.0096). Five of the six smooth absorbers were [is greater than] 40 years old. [AUC.sub.0-[infinity]] and Cmax were lower in heavier patients (r = -0.5396, p = 0.0464 and r = -0.4846, p = 0.0791), respectively. The RIF excretion 0 to 12 h did not correlate with RIF C1 or RIF Clr. In addition, there was no correlation between CrCl and either RIF Cl or RIF Clr.
DISCUSSION
To our knowledge, determinations of the absolute bioavailability of RIF from the tablets vs an IV dosage form in adults have not been performed. Therefore, all parameters were estimated assuming F = 1. Parameter estimates should be interpreted as the value divided by F (V/F, Cl/F, etc).
RIF was well absorbed and most Tmax values were near 2 h. The D-optimal sampling times for the two-sample strategy were 3.1 h, and as late as 14.9 h after the dose. This strategy assumes a low level of detection for the assay. Samples drawn from 1 h to 6 h postdose approached Cmax, although the 2- to 3-h samples were closest overall. The individual serum concentration-vs-time graphs displayed both the smooth and the low accumulation patterns. Under fasting conditions, variability across our 14 subjects was small and values were quite reproducible between the two fasting treatments for nearly all subjects.
Antacids did not effect the absorption of RIF. Therefore, antacids could be taken near the time of RIF dosing, provided that concurrent medications are not affected by the antacids. Based on our MEDLINE search, the effects of antacids on the absorption of RIF have not been described previously. We did find that the histamine-2 antagonist ranitidine has been shown not to effect RIF absorption.[15]
The high-fat meal had significant effects on the RIF, reducing Cmax (-36%) and increasing Tmax (+103%). Food affected [AUC.sub.0-[infinity]] to a lesser extent (-6%). These results are similar to those demonstrated by Siegler et al,[16] who studied 17 patients with active pulmonary tuberculosis; blood was collected at baseline and five time points over 12 h postdose. They showed a 25% reduction in Cmax, 100% increase in Tmax, and 23% reduction in AUC when RIF was administered with a high-fat meal.[16] Zent and Smith[17] administered RIF to 27 patients with active tuberculosis, and blood was collected at baseline and 12 time points over 8 h, plus a 24-h postdose. When RIF was given with a carbohydrate meal, a 15% reduction in Cmax, 19% increase in Tmax, and 4% reduction in AUC were shown. In contrast, when RIF was administered with a high-fat meal, these authors showed no significant effect of RIF’s Cmax, Tmax, and AUC.[17]
Polasa and Krishnaswamy[18] studied six healthy men, dosing them with 10 mg/kg of RIF. Blood was collected at seven time points over 8 h postdose. A wheat-based breakfast consisting of 565 calories, including 9 g protein, 109 g carbohydrate, and 11 g fat was administered on one of two occasions. Compared with the fasting treatment, food reduced the mean Cmax by 30%, doubled the Tmax to 4 h, and reduced the AUC by 26%. Finally, Hagelund et al[19] studied six patients with tuberculosis, in addition to six gastrectomized patients, collecting blood at 1, 3, 5, and 7 h postdose. They compared fasting conditions with a breakfast of bread with butter and marmalade, meat, cheese, one egg, and coffee or tea with milk.[19] This meal delayed absorption, but showed only minor effects on Cmax and AUC in the nongastrectomized patients. Those with a history of GI surgery showed modest effects from food, but significant intersubjeet variability in RIF absorption, regardless of fed or fasting condition. The above articles show that RIF absorption can be affected by various types of meals. The precise content of each meal differed from study to study. It is possible that the specific foods consumed, as much as their content of carbohydrate, protein, or fat, played some role in the changes seen.
Narang et al[20] studied the rifamycin derivative rifabutin, and showed that food had less of an effect on the Cmax (-17%) than we have shown for rifampin’s Cmax (-36%). Their study also showed effects of food on the Tmax (+80%) and AUC (-5%) of rifabutin similar to our results for rifampin. Another rifamycin derivative, rifapentine, actually shows improved absorption with food. Owens et al[21] demonstrated that food increased rifapentine’s Cmax 50%, increased the Tmax by only 11%, and increased the AUC by 46%.
The correlations of Cmax and Tmax with age were relatively weak, although older subjects were significantly more likely to be smooth absorbers. The apparent difference between smooth and low absorbers is due, in part, to the blood sampling schedule. Such differences may not have been apparent with more frequent early blood samples. The negative correlation between weight and both the RIF AUC and the Cmax suggests that RIF should be dosed on a milligram per kilogram basis to avoid underdosing large patients.
The NPEM2 analysis produced parameter estimates consistent with those from our previous investigation.[5] RIF displayed median values for K and [t.sub.1/2] values similar to those from the previous study. The median values for V and Cl from this study were slightly larger. The reasons for these differences were not apparent, other than a different set of subjects were studied.
RIF is cleared predominantly through nonrenal mechanisms, with only 10% of the drug reported to be cleared unchanged in the urine over 24 h.[20] We have reproduced those findings. RIF is converted to 25-desacetylrifampin and other, less abundant metabolites, which are subsequently cleared through nonrenal and, to a lesser extent, renal mechanisms. 25-desacetylrifampin displays microbiological activity approaching that of RIF, while displaying serum concentrations approximately 10% of those for RIF. We did not assay the metabolite in our study.
RIF has good activity against Mycobacterium tuberculosis and modest activity against Mycobacterium avium. Using radiometric techniques, Heifets[22] determined the minimal inhibitory concentration (MIC) of RIF to be 0.25 [micro]g/mL against M tuberculosis, and 4 [micro]g/mL against M avium. Against an isolate of M tuberculosis with an MIC of 0.25 [micro]g/mL, the RIF Cmax: MIC ratio is 42:1, and serum concentrations would remain above MIC for about 15 h. In contrast, against an isolate of M avium with an MIC of 4 [micro]g/mL, the RIF Cmax: MIC ratio is [is less than] 3:1, and serum concentrations would remain above MIC for about 4.5 h. This analysis is consistent with RIF’s superior activity against M tuberculosis compared with M avium.
CONCLUSIONS
The RIF serum concentrations found in this study were consistent with those previously described. In addition, the kinetic behavior of RIF was consistent between the two fasting treatments. Samples drawn between 2 and 3 h postdose approach Cmax for most subjects. Antacids had no effect on the absorption of RIF; however, food significantly decreased Cmax and increased Tmax. Therefore, RIF should be taken on an empty stomach whenever possible.
REFERENCES
[1] American Thoracic Society/Centers for Disease Control. Treatment of tuberculosis and tuberculosis infection in adults and children. Am J Respir Crit Care Med 1994; 149:1359-1374
[2] Acocella G, Conti R, Luisetti M, et al. Pharmacokinetic studies on antituberculosis regimens in humans. Am Rev Respir Dis 1985; 132:510-515
[3] Peloquin CA. Antituberculosis drugs: pharmacokinetics. In: Heifets LB, ed. Drug susceptibility in the chemotherapy of mycobacterial infections. Boca Raton, FL: CRC Press, 1991; 59-88
[4] Peloquin CA. Using therapeutic drug monitoring to dose the antimycobacterial drags. Clin Chest Med 1997; 18:79-87
[5] Peloquin CA, Jaresko GS, Yong CL, et al. Population pharmacokinetic modeling of isoniazid, rifampin, and pyrazinamide. Antimicrob Agent Chemother 1997; 41:2670-2679
[6] Peloquin CA, Namdar R, Nix DE. Effect of food and antacids on the pharmacokinetics (PK) of isoniazid (INH) and (RIF) [abstract]. Am J Respir Crit Care Med 1998; 157:A493
[7] Anonymous. 1983 Metropolitan height and weight tables. Statistical Bulletin–Metropolitan Life Foundation. 1983; 64:3-9
[8.] Peloquin CA. Rifampin stability [letter]. Ther Drug Monit 1998; 20:450-451
[9] Jelliffe RW. Explicit determination of laboratory assay error patterns: a useful aid in therapeutic drug monitoring. ASCP clinical pharmacology check sample 10: No. DM 89-4 (DM 56). Chicago, IL: American Society of Clinical Pathologists, 1989; 1-6
[10] Gibaldi M, Perrier D. Pharmacokinetics. 2nd ed. New York, NY: Marcel Dekker, 1982
[11] Jelliffe RW, Schumitzky A, Van Guilder M. User manual for version 10.6 USC*PACK collection of PC programs. Los Angeles, CA: Laboratory of Applied Pharmacokinetics, USC, 1996
[12] D’Argenio D, Schumitzky A. ADAPT II user’s guide, biomedical simulations resource. Los Angeles, CA: University of Southern California, 1992
[13] Cockroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron 1976; 10:31-41
[14] Food and Drug Administration. Bioavailability and bioequivalence requirements. Fed Register 1992; 57:17997-18001
[15] Purohit SD, Johri SC, Gupta PR, et al. Ranitidine-rifampin interaction. J Assoc Physicians India 1992; 40:308-310
[16] Siegler DI, Bryant M, Burley DM, et al. Effect of meals on rifampicin absorption. Lancet 1974; 2:197-198
[17] Zent C, Smith P. Study of the effect of concomitant food on the bioavailability of rifampicin, isoniazid, and pyrazinamide. Tubercle Lung Dis 1995; 76:109-113
[18] Polasa K, Krishnaswamy K. Effect of food on bioavailability of rifampin. J Clin Pharmacol 1983; 23:433-437
[19] Hagelund CHH, Waahlen P, Eidsaunet W. Absorption of rifampin in gastrectomized patients: effect of meals. Scand J Respir Dis 1977; 58:241-246
[20] Narang PK, Lewis RC, Blanchine JR. Rifabutin, absorption in humans: relative bioavailability of food effect. Clin Pharmacol Ther 1992; 52:335-341
[21] Owens RC, Keung CF, Gardner S, et al. Pharmacokinetic and food effect evaluation of rifapentine in subjects seropositive for the human immunodeficiency virus [abstract A2]. In: Program and abstracts of the 37th Interscience Conference on Antimicrobial Agents and Chemotherapy. Washington DC: American Society for Microbiology, 1997; 1
[22] Heifers LB. Antituberculosis drugs: anti-microbial activity in vitro. In: Heifets LB, ed. Drug susceptibility in the chemotherapy of mycobacterial infections. Boca Raton, FL: CRC Press, 1991; 13-58
(*)From the Department of Medicine (Drs. Peloquin and Namdar and Mr. Singleton), National Jewish Medical and Research Center, Denver, CO; School of Pharmacy (Dr. Peloquin) and School of Medicine (Dr. Peloquin), University of Colorado Denver; and School of Pharmacy (Dr. Nix), University of Arizona, Tucson, AZ.
Presented in part at the American Lung Association/American Thoracic Society International Conference, Chicago, IL, April 24 to 29, 1998.
This study was supported, in part, by NIH grant 1 RO1 AI37845. Correspondence to: Charles A. Peloquin, PharmD, Director, Infectious Disease Pharmacokinetics Laboratory, National Jewish Medical and Research Center, 1400 Jackson St, Denver, CO 80206; e-mail: peloquinc@njc.org
COPYRIGHT 1999 American College of Chest Physicians
COPYRIGHT 2000 Gale Group
