Previous Article | Next Article ![]()
Antimicrobial Agents and Chemotherapy, September 2003, p. 2775-2780, Vol. 47, No. 9
0066-4804/03/$08.00+0 DOI: 10.1128/AAC.47.9.2775-2780.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.
University of Louisville,1 Department of Veterans Affairs, Louisville, Kentucky,2 Pharmacia and Upjohn Company, Kalamazoo, Michigan3
Received 30 September 2002/ Returned for modification 20 February 2003/ Accepted 30 May 2003
|
|
|---|
|
|
|---|
Linezolid is dosed intravenously or orally at 400 or 600 mg twice a day. Because the bioavailability of linezolid is approximately 100%, no dosage adjustment is needed when therapy is changed from the intravenous to the oral route (12). After the administration of oral 600-mg doses, steady-state peak concentrations in plasma (Cmax) of 21.2 ± 5.78 µg/ml are obtained at a time to Cmax (Tmax) of 1.03 ± 0.62 h. The plasma elimination half-life is 5.40 ± 2.06 h. Clearance occurs by both renal and nonrenal (65%) mechanisms and is moderately variable (80 ± 29 ml/min) at steady state with a dose of 600 mg twice daily. Linezolid is neutral in the physiological pH range and undergoes renal tubular reabsorption. The level of plasma protein binding is 31%, and the volume of distribution approximates the total body water content (40 to 50 liters) (9, 10; Zyvox package insert, Pharmacia Corporation, 2001). In a single-dose study, female human subjects had about 20% lower body weight-normalized clearances than males (Pawsey et al., Abstr. Eur. Congr. Antimicrob. Chemother., 1996). Linezolid is metabolized to two inactive metabolites, an aminoethoxyacetic acid metabolite (metabolite A) and a hydroxyethyl glycine metabolite (metabolite B).
Because a significant component of linezolid elimination is as unchanged drug in the urine, this study examined the single-dose pharmacokinetics of linezolid in male and female volunteers with various degrees of renal function to determine if dose adjustment is needed as a function of renal impairment. The renal functions of the subjects ranged from normal to severe chronic impairment, including end-stage renal disease (ESRD). Those with ESRD were maintained on hemodialysis.
|
|
|---|
In the week prior to receiving the study medication, the subjects were screened for enrollment into the study. A physical examination and laboratory tests were performed to ensure that the subject met the inclusion and exclusion criteria. In those subjects that produced urine, a 24-h urine collection was obtained for determination of the subject's CLCR. Subjects were enrolled in one of the study groups on the basis of this 24-h CLCR. The results of the 24-h CLCR were compared to an estimate of CLCR by use of the equation of Cockcroft and Gault (2). No measurement of CLCR was sufficiently different from the estimated CLCR to prompt a remeasurement of the CLCR.
On the day prior to the study, subjects were admitted to the residence associated with the test facility for an overnight stay and were fed a tyramine-free diet. On the next morning they reported to the clinic in a fasted state. An intravenous catheter was placed in a forearm vein, and a single 600-mg tablet of linezolid was taken by mouth with water. Venous blood samples were obtained at 0.5, 1.0, 1.5, 2, 3, 4, 6, 8,10, 12, 16, 24, 36, and 48 h. Additional blood samples were obtained at 72 h for the subjects in groups 2 and 3 and at 96 h for the subjects in group 3. Hemodialysis patients were studied twice: once while they were on dialysis and once between dialysis sessions. During the dialysis session, hemodialysis was begun 3 h after administration of the oral dose. Additional arterial and venous blood sample pairs were obtained 30 min into the session, at the midpoint, and 30 min prior to the end of the session. Four hours of dialysis was prescribed for all subjects, and all subjects were dialyzed with a CA210 dialyzer (cellulose acetate; surface area, 2.1 m2; Kuf, 9.3) or an F80 dialyzer (polysulfone; surface area, 1.8 m2; Kuf, 65) with a Baxter 550 hemodialysis machine. Blood flow rates were between 400 and 450 ml/min. All dialyzers were processed for reuse. The hematocrit was determined for each of the arterial blood samples. Postdialyzer dialysate was collected by a continuous sampling technique and placed on ice. A t-adaptor was placed in the dialysate effluent line, in which one of the openings was attached to a sampling pump calibrated to remove 1 ml/min. The remainder of the dialysate went to waste. The amount of dialysate processed was recorded, and the sample was frozen for linezolid assay. All samples were frozen and stored at -70°C until they were shipped in dry ice to the sponsor for assay.
Urine was collected from all subjects who made urine. Subjects collected a urine specimen on the day prior to the study and during the study. During the study urine was collected from all groups over the following time intervals: 0 to 4, 4 to 8, 8 to 12, 12 to 24, and 24 to 48 h. For subjects in groups 2 and 3, urine was also collected from 48 to 72 h; and for subjects in group 3, urine was also collected from 72 to 96 h. The hemodialysis patients did not produce appreciable amounts of urine, and none was collected. Urine samples were split into two fractions and frozen at -70°C until they were shipped to the sponsor. One sample was sent to the sponsor for linezolid assay, and the other was assayed for creatinine at the study laboratory. The CLCRs for these samples were calculated as the amount of creatinine recovered in 24 h divided by 1,440 times the concentration in plasma at 12 h.
All plasma, urine, and dialysate samples were shipped on dry ice to the Pharmacia and Upjohn Company for assay by a validated chromatographic method (8). The concentrations of linezolid in human plasma were generated by a sensitive and selective high-performance liquid chromatographic (HPLC) method that was validated at AvTech Laboratories, Inc. (Kalamazoo, Mich.). Mean recoveries for linezolid and the internal standard were approximately 95.4 and 95.8%, respectively. Calibration standard responses were linear over the range of 0.0100 to 20.0 µg/ml by use of a weighted (1/concentration) linear least-squares regression. The lower limit of quantitation (LLOQ) was 0.0100 µg/ml.
Nine analytical runs were required to report the linezolid concentrations found in the 524 plasma specimens. Correlation coefficients were all equal to 1.000. Coefficients of variation (CV) were used to express the precision of the back-calculated calibration standard. The eight calibration standards had CVs that were
4.0%, with mean accuracies between 97.6 and 106%. The interday accuracy and precision were further monitored by analysis of three linezolid quality control (QC) standards with target concentrations of 0.0400, 4.00, and 15.0 µg/ml. The accuracies for the QC standards ranged from 98.5 to 99.7%, with interday precisions (CVs) for all QC standards being
4.3%. Any specimen with a concentration above the concentrations on the calibration standard curve was diluted and reassayed along with an appropriately diluted QC standard prepared with a target concentration of 35.0 µg/ml. Linezolid is stable in human urine when it is stored at
-20°C for 1 year.
Calibration standard responses were linear over the range of 0.200 to 100 µg/ml by use of a weighted (1/concentration2) linear least-squares regression. Results below the LLOQ were reported as <0.200 µg/ml and set to missing for the pharmacokinetic analysis.
When sufficient sample volume was available (which was the case for 463 of 468 total samples), the concentrations of the two major metabolites of linezolid in plasma and urine were determined by a sensitive and specific liquid chromatography-mass spectrometry-mass spectrometry method developed and conducted by Pharmacia and Upjohn Company. The assay was performed as described above for parent linezolid, and acceptable interassay precision and variability parameters were observed for the analytical runs required to generate concentration data for the metabolites.
A cross-validation assay that determined the acceptability of using the HPLC procedure with urine to also accurately quantitate the linezolid in the dialysate samples was conducted. Dialysate samples were included in two of the analytical runs with urine, and the results were successfully reported in the data report for urine.
Pharmacokinetic analysis of the data was performed by noncompartmental techniques. Plots of the data for all subjects were made by using Sigma-Plot software, with the data plotted as the log of the concentration versus time. All calculations were done with an Excel spreadsheet and validated by use of the Clinical Pharmacokinetics Analysis Package (version 1.0.39; Triology Consulting Corporation Clinical Pharmacokinetics Analysis Program, Version 1 Users's Manual, Triology Consulting Corporation, Kalamazoo, Mich., 1996). The terminal elimination phase of the plot was identified visually, and the elimination rate constant (kel) was calculated by linear regression of these datum points. The elimination half-life was calculated as ln 2/kel. The area under the curve (AUC) was calculated by using the trapezoidal rule, and a tail correction was added to this area and was calculated as the last concentration above 0.2 µg/ml (LLOQ) divided by kel. The total apparent oral clearance was calculated as the dose divided by the AUC. The renal clearance was calculated as the amount of parent drug recovered in the urine divided by the AUC. The nonrenal clearance was calculated as the difference between the apparent oral clearance and the renal clearance. The apparent volume of distribution was calculated as the oral dose times the area under the moment curve divided by AUC2. Tmax and Cmax were determined by inspection of the data. We calculated the extraction ratio across the dialyzer as the concentration in the venous blood sample divided by the concentration in the arterial blood sample for each of the additional arterial blood-venous blood sample pairs. The amount of drug removed during dialysis was determined by collecting a continuous sample of dialysate during the treatment. The total volume of dialysate processed was determined from the dialysate flow rate (600 ml/min) and the length of the procedure. The amount removed was then the concentration in the dialysate times the volume processed.
Metabolite pharmacokinetic parameter estimates were calculated by noncompartmental methods. AUC was not extrapolated to infinity in patients with ESRD due to the lack of data for a sufficiently long period of time for accurate determination of a kel (the sampling time was limited to a total of 48 h postdosing due to the necessary dialysis schedule). Half-life data were estimated, but for the subjects with ESRD, these estimates have limitations due to the short sampling duration relative to the estimated half-life and should be interpreted with caution.
Statistical analysis of the data was performed with SPSS software (version 8.0). The data were analyzed by linear regression with analysis of variance by using the CLCR measured during the pharmacokinetic study as the independent variable. The data for the pharmacokinetic parameters for subjects in group 4 were tested for differences between the days off and on dialysis by a paired t test.
|
|
|---|
|
View this table: [in a new window] |
TABLE 1. Demographics of the study groupsa
|
![]() View larger version (27K): [in a new window] |
FIG. 1. Plots of the mean serum linezolid concentration over time for the subjects, by group. Group 1, CLCR, >80 ml/min; group 2, CLCR, 40 to 80 ml/min; group 3, CLCR, 10 to 39 ml/min; group 4, hemodialysis patients.
|
![]() View larger version (15K): [in a new window] |
FIG. 2. Plots of mean serum linezolid concentrations over time for hemodialysis subjects while off dialysis and on dialysis.
|
|
View this table: [in a new window] |
TABLE 2. Linezolid pharmacokinetics in patients with different renal functions after administration of a 600-mg oral dose of linezolida
|
![]() View larger version (14K): [in a new window] |
FIG. 3. Regression plot of the oral linezolid clearance versus the measured CLCR at the time of dosing for all subjects in the study. The line represents the regression line of those data.
|
|
View this table: [in a new window] |
TABLE 3. Metabolite A pharmacokinetics in patients with different renal functions after administration of a 600-mg oral dose of linezolida
|
|
View this table: [in a new window] |
TABLE 4. Metabolite B pharmacokinetics in patients with different renal functions after administration of a 600-mg oral dose of linezolida
|
|
|
|---|
We studied the elimination kinetics of linezolid in patients with renal insufficiency and on dialysis. We calculated linezolid clearance following oral administration of the drug. We are therefore reporting the apparent oral clearance of the drug that may be confounded with a bioavailability term; however, this is unlikely to be a significant factor, considering the absorption properties of linezolid (complete absorption). As shown in Fig. 3, renal function has little effect on the overall clearance of the drug. In fact, despite a strong relationship between the renal clearance of the drug and CLCR (r2 = 0.60; P < 0.001), total clearance did not fall as CLCR fell. This observation is accounted for by the increase in the nonrenal clearance seen as CLCR fell. In fact, some of the highest clearances occurred in subjects with CLCRs below 40 ml/min. This apparent compensatory increase in nonrenal linezolid clearance could be the result of enhanced drug biotransformation, biliary excretion, decreased absolute bioavailability, or changes in drug distribution. For orally administered drugs, the volume of distribution is confounded by bioavailability and therefore cannot be accounted for separately.
Linezolid is a dialyzable drug, and a significant portion of the dose could potentially be removed during a dialysis session, depending on the time of administration of the last dose in relation to the time of the hemodialysis session. A typical hemodialysis session occurs at a frequency of three times per week, with 48 to 72 h between the start of sessions. Therefore, there exists the potential for the elimination of a significant portion of the dose in one of every four to six doses. Overall, linezolid elimination is not affected by renal function, and no dosage adjustment is warranted for patients with renal impairment. However, during the first dialysis session of the treatment course, a supplemental dose of linezolid may be given if necessary to keep levels above the MIC for the organism causing the infection being treated.
The two major metabolites of linezolid were found to accumulate in patients with renal impairment, especially those with severe impairment or ESRD. This would not be unexpected, as both metabolites are almost exclusively eliminated in urine (10). The significance of the metabolite accumulation is not fully understood at this time. However, it is known that the metabolites are qualitatively the same as those in preclinical safety evaluations of linezolid, although they are present at lower concentrations than those expected with multiple dosing in humans. The metabolites also do not possess significant antimicrobial activity or a significant potential to inhibit various cytochrome P450 enzymes or monoamine oxidase. In addition, none of the more than 300 patients with severe renal impairment who have received linezolid either in phase III clinical trials or on a compassionate-use basis has shown any evidence of new types of adverse reactions or unacceptable tolerance, with the exception that the incidence of thrombocytopenia may be higher in patients with severe renal impairment. This is most likely due to the patients' underlying characteristics (low baseline blood counts and longer-term treatments) and not due to the metabolites themselves. The clinical significance of metabolite accumulation is not yet fully elucidated and requires additional study with multiple doses in a controlled manner.
Hemodialysis is a significant source of elimination of linezolid and its two major metabolites in patients with end-stage renal failure. In this study, approximately one-third of the administered dose of the drug was removed by hemodialysis. In the experiments described here we used a high-flux or high-efficiency dialysis membrane comparable to others used in hemodialysis centers in the United States. Therefore, others should have the same experience that we did. Specifically, they should see an extraction of approximately 37% of the drug as it passes through the dialyzer. The amount removed depends on the proximity of the dialysis session in time to the time of dose administration. In our specific case, one-third of the dose was removed when dialysis was started 3 h following administration of the dose. A typical dialysis schedule of three treatments weekly would lead to an average weekly clearance of 100 ml/min, which is approximately the average for the other three renal function groups.
On the basis of these data, the dose of linezolid does not need to be altered for patients with impaired renal function. Because the drug is substantially removed by hemodialysis and to avoid potentially ineffective therapy, one of the twice-daily doses should be administered after the dialysis treatment on days when dialysis is performed.
Jack E. Brewer is acknowledged for conducting the metabolite assay work presented in this paper.
This work was performed at the Kidney Disease Program, Department of Medicine, University of Louisville, Louisville, Ky.
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»