-a

ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, May 1990, p. 792-795

Vol. 34, No. 5

0066-4804/90/050792-04$02.00/0 Copyright © 1990, American Society for Microbiology

Vancomycin Pharmacokinetics in Burn Patients and Intravenous Drug Abusers MICHAEL J. RYBAK,12* LISA M. ALBRECHT,'t JULIE R. BERMAN,2 LAWRENCE H. WARBASSE,3 AND CRAIG K. SVENSSON1

College ofPharmacy and Allied Health Professionsl* and School of Medicine,3 Wayne State University, Detroit, Michigan 48202, and Departments of Pharmacy Services2 and Internal Medicine,4 Detroit Receiving Hospital and University Health Center, Detroit, Michigan 48201 Received 6 September 1989/Accepted 15 February 1990

The pharmacokinetics of vancomycin were evaluated in 34 patients (10 burn patients, 14 intravenous drug abusers [IVDAJ, and 10 controLs). Multiple serum samples were drawn following a 1-h vancomycin ion at steady state over an 8- to 12-h dosing interval. Pharmacokinetic parameters were derived by noncompartmental analysis. There were no significant differences among the groups with respect to age, weight, serum creatinine, volume of distribution, or protein binding. Burn patients had a sigificantly higher creatinine clearance than did IVDA or controls. Vancomycin clearances averaged 142.8, 98.0, and 67.7 mI/min in burn patients, IVDA, and controls, respectively. The renal clearance of vancomycin was also higher in burn patients than in the other groups. IVDA tended to have a higher vancomycin clearance (31% higher) than did controls, but the difference was not statistically significant. Vancomycin clearance was much higher in burn patients requiring dosage individualization and close monitoring. A considerable amount of vancomycin was eliminated through renal tubular secretion, making dosage predictions based on creatinine clearance more difficult. Further work with IVDA will be needed to determine if they represent a group requlring aggressive vancomycin dosages.

Renewed interest in vancomycin, resulting from the inin methicillin-resistant Staphylococcus aureus and Staphylococcus epidermidis, has prompted new investigations regarding the pharmacokinetics, potential for toxicity, and dosage adjustment guidelines of vancormycin (8, 9, 11, 16, 17, 21). Altered pharmacokinetics of vancomycin resulting in increased dosage requirements have been investigated in. burn patients. Although there is clinical evidence for increased vancomycin dosage requirements in this patient population, data supporting these observations have been inconclusive (3, 6). A similar phenomenon has been documented to occur with aminoglycoside administration'in burn patients. The mechanism for enhanced aminoglycoside elimination that results in increased dosage requirements'for burn patients appears to be related to an increased glomerular filtration rate (12, 23-25). Intravenous drug abusers (IVDA) have also been reported to eliminate aminoglycosides at a rate which necessitates more aggressive aminoglycoside dosage regimens (10). This observation has not been reported with vancomycin' a'dministration in this patient population. In this inve'stigation, we compared the steadystate pharmacokinetics of vancomycin in burn patients, IVDA, 'and a control population.

participation in the study. Patients who were admitted to our hospital and who were started on vancomycin by their physicians for presumed or documented gram-positive infections were evaluated for the study. Patients were eligible for the burn group if they presented with >10% total body surface area burn injury and for the IVDA group if they had a history of recently injecting intravenous drugs; all other patients were eligible for the control group. Exclusion criteria consisted of significant liver or renal disease (liver transaminase values greater than three times the normal range or serum creatinine of >2 mg/dl), recent surgery or infection of the upper urinary tract, concurrent exposure to drugs which may affect the renal elimination of vancomycin (dopamine, dobutamine, loop diuretics, etc.), or inability to provide informed consent. The dosages of vancomycin ranged from 2,000 to 6,000 mg/day (28.9 to 42.7 mg/kg of actual body weight). All vancomycin doses were administered by intravenous infusion over a 1-h period. The pharmacokinetic evaluation was carried out a minimum of 48 h after the initiation of vancomycin therapy. Sampling for vancomycin occurred over the prescnbed dosing interval of 8 or 12 h. Blood samples were serially collected from a central line catheter or a heparin lock placed in the forearm opposite the infusion site. Samples were obtained before infusion, at the end of infusion, and at 0.17, 0.33, 0.5, 0'.75, 1, 1.5, 2, 3, 4, 6, and 7 h postinfusion for patients on an 8-h dosing interval. Additional samples were obtained at 8 and 11 h for patients receiving vancomycin on a 12-h dosing interval. Specimens were allowed to clot for approximately 45 min before centrifugation. Urine was collected from the start of the vancomycin infusion to the end of the dosing interval for the determination of creatinine clearance (CLCR) and vancomycin renal clearance (CLR). Burn patients were catheterized, assuring accurate urine collection. Nonburn patients were asked to void prior to dose administration. Urine was

crease

MATERIALS AND METHODS A total of 34 patients participated in this study. Ages ranged from 18 to 64 years (mean ± standard deviation, 37 10.9 years). Ten burn patients (5 males) and 14 IVDA (9 males) made up the treatment study groups. The inpatient control group (nonburn, non-IVDA) consisted of 10 patients (8 males). Informed consent was obtained from all patients prior to * Corresponding author. t Present -address: College of Pharmacy, University of Illinois, Chicago, IL 60612.

792

VOL. 34, 1990

VANCOMYCIN ELIMINATION IN VARIOUS PATIENT POPULATIONS

793

TABLE 1. Patient demographics No. of males/

No.tfemales

Group Age (yr)a Group (n)(n) Age (yr)a

Vancomycin dose

(mg/kg)a

CLAlbumin CLCR (ml/min)' (mg/dl)a

IVDA (14) 36.0 t 6.2 9/5 28.9 ± 6.4 85.5 t 23.0 Control (10) 38.0 ± 11.9 8/2 29.4 ± 6.0 68.3 + 30.4 Burn (10b) 36.4 ± 14.9 5/5 42.7 ± 18.5c 111.0 t 28.3c aMean + standard deviation. b 30.9o + 14.4% of total body surface area was burned, and 18.3 + 11.7 days elapsed before treatment with vancomycin was begun. c P < 0.05.

collected over the dosing interval by spontaneous voiding. One of the investigators was present in the subject's room for the majority of the collection period. Serum and urine samples were stored at -20°C until analysis. Samples were assayed for vancomycin content within 1 week from the time of collection by a fluorescence polarization immunoassay (TDx; Diagnostic Division, Abbott Laboratories, Irving, Tex.). Our intra- and interrun coefficients of variation were less than 6% in the concentration range of 0.6 to 100 ,ug/ml. The coefficients of variation for the control standards that were run with each group of subject specimens averaged 4.7% for the high control (75 ,ug/ml), 1.8% for the medium control (35 ,ug/ml), and 3.5% for the low control (7.0 ,ug/ml). Vancomycin serum protein binding was determined by ultrafiltration of the 1-, 6-, and 11-h serum samples. Total vancomycin and unbound vancomycin concentrations were determined in serum with a micropartition system (MPS-1; Amicon Corp., Danvers, Mass.). The free fraction (fu) of vancomycin was expressed as the ratio of the vancomycin concentration in the ultrafiltrate to that in serum multiplied by 100. Minimum vancomycin absorption (mean + SD recovery, 97.2% + 0.8%) was observed when an ultrafiltrate of serum was spiked (24 eg/ml) with vancomycin (five replicates). This observation is consistent with those in previous reports using this method for the determination of vancomycin serum protein binding (1, 16). The serum concentration-versus-time data were fit to polyexponential functions with a nonlinear least-squares regression program (4). The minimum number of exponentials needed to describe the curve was determined with a modified Akaike Information Criteria test with a weighting factor of l/y2 (4). The area under the serum concentrationversus-time curve over the dosing interval (AUCSSOT) was calculated by the linear trapezoidal rule with the LAGRAN program (15). Total body clearance (CL) was calculated as follows: CL = dose/AUCSS-r The steady-state volume of distribution (Vss) was calculated with the formula of Gibaldi and Perrier (7), as follows: vss

=

dose. AUMCi.V.0, 2 (AUCSS-)2

-

dose. t 2. AUCsso5

where AUMCi.v0,. is the area under the first non-normalized moment curve after single-dose administration from zero time to infinity and t is the duration of infusion. AUMC.v.0,. was estimated by the method of Smith and Schentag (20), which provides a means for calculating AUMCi.Vo-. from steady-state data. Vancomycin CLR was calculated as follows:

CLR

=

Ael/AUCss5,

where Aeo., is the amount of unchanged drug excreted in urine over the dosing interval. Clearance of unbound drug and steady-state volume of distribution of unbound drug were determined by dividing

2.6 t 0.3 2.8 ± 0.6 1.7 ± 0.5c

the pharmacokinetic parameter for total drug by the measured fu in serum. The filtration clearance of unbound vancomycin (CLF) was estimated as the product offu and the glomerular filtration rate (assumed to be equal to CLCR). Urinary recovery (percentage) of unchanged drug was calculated by dividing the amount recovered in urine by the dose administered. Vancomycin net renal secretory clearance (CLs) was calculated by subtracting CLF from CLR. The CLR/CLF ratio was calculated to elucidate the mechanisms of vancomycin renal excretion. Differences in patient characteristics and pharmacokinetic parameters for the three groups were determined by analysis of variance. Tukey's test was used to identify differences between burn patients, IVDA, and controls. Differences between trough (i.e., the sample prior to dosing) vancomycin concentrations and C (i.e., the last sample) were compared with a paired Studnt t test to ascertain that patients had indeed reached the steady state. Statistical significance was defined as P < 0.05. Data are presented as means ± SDs unless otherwise indicated. RESULTS The demographic characteristics, including burn patient descriptions, are shown in Table 1. Infection diagnosis consisted of sepsis (burn, n = 10; IVDA, n = 3; control, n = 5), cellulitis (IVDA, n = 6; control, n = 4), osteomyelitis (IVDA, n = 3; control, n = 1), septic arthritis (IVDA, n = 1), and endocarditis (IVDA, n = 1). The time postburn in which patients in the burn group were started on vancomycin therapy ranged from 7 to 45 days. There were no significant differences among groups in age, actual body weight, or serum creatinine. Overall, serum albumin levels were low in all three groups, ranging from 1.0 to 3.8 mg/dl (all groups combined, 2.4 + 0.7 mg/dl) but were significantly lower in the burn patients (1.7 ± 0.5 mg/dl). Burn patients tended to be more aggressively dosed initially, accounting for the significant difference between vancomycin dosages prescribed for burn and nonburn patients. There were no significant differences between Cmin and trough vancomycin concentrations in serum, indicating the achievement of the steady state. Mean values for steady-state vancomycin pharmacokinetic parameters are shown in Table 2. CLCR and vancomycin CL were found to be significantly greater in burn patients than in either IVDA or controls. CLR was significantly greater in burn patients than in controls. Although there was a trend for vancomycin CL and CLR to be greater in the IVDA group than in the control group, it did not reach statistical significance. The proportion of vancomycin eliminated by nonrenal mechanisms was found to be small and not statistically different among the three groups. There were no differences in Vs among the three groups. Vancomycin serum protein binding did not differ among the three groups, withfu ranging from 0.41 to 0.77. Hence, changes in

794

ANTIMICROB. AGENTS CHEMOTHER.

RYBAK ET AL. "

V_< r-

+I +I +I

x) ch ooo W)

tr _ N bb ; | +1+1 +1 rN 6 oreports have indicated that burn patients exhibit altered vancomycin pharmacokinetics as compared with nonburn ,J,+l +l +l patients. Brater et al. (3) found enhanced vancomycin CL in +o a X ci 06 Oburn patients which correlated with an increased CLCR. Garrelts and Peterie (6) found a 78% increase in the vancomycin dosage required to produce therapeutic concentrao .h tions in serum in burn patients as compared with nonburn 4 patients. This increased dosage requirement occurred even _ ,.5 -_ +I+1+I+ U though there was only a 12% increase in CLCR in burn versus o= t c W; Vt k> xnonburn patients. In the present investigation, we found that vancomycin CL was significantly greater in burn patients than in the control group or the IVDA group. This was 00 > N associated with a significant increase in CLCR. 8 oE x c>Several m possible mechanisms for enhanced vancomycin +I +I +I and aminoglycoside elimination in burn patients have been _ ~oproposed, including (i) increased nonrenal elimination (parvl; ticularly through the burn wound), (ii) increased glomerular filtration rate, and (iii) altered protein binding (2, 3, 12, 18, 19). Our data indicate that there is no significant difference in - t'; the urinary recovery of vancomycin in burn patients as m compared with nonburn patients. This would appear to rule +1+1 +1 ^iu w out enhanced nonrenal elimination in burn patients as a mechanism of increased clearance, assuming that the assay used is not interfered with by any as-yet-unidentified metabolites or degradation products. Although there have been 00X m 0 0

.

studies documenting differences in vancomycin concentrations determined by high-pressure liquid chromatography or +++ CZ fluorescence polarization immunoassay in anuric patients undergoing peritoneal dialysis, studies in patients with norE~ CD Co ,0 cl, IC 11* mal renal function (as in this investigation) are lacking (13, _ o 22). The fact that protein binding did not differ among the three groups also indicates that altered binding is not a ZF . N ,plausible J mechanism for the altered elimination observed in _ r^ burn patients. These observations indicate that the probable U nD =Q ch +I +I +I O mechanism for increased vancomycin elimination in burn o patients is secondary to altered renal elimination. ; x~rThere are conflicting data in the literature regarding whether the mechanism(s) of vancomycin renal elimination involves filtration only or filtration and secretion. Golper and associates (8) performed the most rigorous evaluation of 00l0°t+1 renal elimination mechanisms published to vancomycin ++ Based upon their observations, these investigators In oo =date. concluded that vancomycin was eliminated only via filtration and nonrenal mechanisms. Unfortunately, technical difficulties prevented the investigators from determining protein v binding, and they assumed a value of 10 to 20%. The o X,. = E ;> 4 o g assumption of protein binding values similar to those meaUd m sured in the present study would have led Golper et al. to -

E

VOL. 34, 1990

VANCOMYCIN ELIMINATION IN VARIOUS PATIENT POPULATIONS

conclude that the renal mechanisms of vancomycin elimination include both filtration and secretion. Our finding that the CLR/CLF ratio was greater than 1 in all three patient groups supports the conclusion that secretion contributes to the renal elimination of vancomycin. This conclusion is consistent with the observations of other investigators (5, 14, 16; T. A. Golper, L. Elizinga, H. Noonan, J. Anderson, D. N. Gilbert, and W. M. Benner, Program Abstr. 26th Intersci. Conf. Antimicrob. Agents Chemother., abstr. no. 1260, 1986). Thus, burn patients may exhibit an increased filtration fraction, an increased secretory component, or both. The data in Table 2 suggest that both filtration and secretion may be enhanced in burn patients, although because of the large intersubject variability, only the increase in secretory clearance was statistically significant. The fact that the small proportion of drug eliminated by the nonrenal route was not statistically different among the three groups further supports this observation. Further studies are needed to confirm these observations and to probe the physiological and biochemical mechanisms of such changes. The extensive use of vancomycin in IVDA and the report by King et al. (10) of altered aminoglycoside pharmacokinetics in this patient population led us to examine this subgroup for possible altered kinetics. While the mean vancomycin CL was 31% higher in IVDA than in controls, the difference was not statistically significant. We recently reported (M. J. Rybak, S. A. Lerner, D. P. Levine, L. M. Albrecht, P. L. McNeil, G. A. Thompson, and M. T. Kenny, 29th ICAAC, abstr. no. 1343, 1989) increased dosage requirements in IVDA receiving teicoplanin (an investigational glycopeptide antibiotic similar to vancomycin). The trend towards an increased vancomycin CL and the observations with teicoplanin indicate the need for further examinations of IVDA as a patient population which may require higher-than-usual dosages of some antibiotics. There are conflicting data in the literature as to the ability of CLCR to predict vancomycin CL in various patient populations. While we observed a reasonable correlation (all three patient groups combined) between CLCR and vancomycin CL (r2 = 0.77), the limited range of CLCR values in our patient population (which excluded those with renal dysfunction) did not permit us to make a rigorous assessment of the utility of CLCR in guiding dosage regimen design. In summary, burn patients exhibit a higher vancomycin CL than do nonburn patients and, therefore, require more frequent and higher doses. The mechanism of this enhanced clearance appears to be through both increased filtration and increased secretory clearance. A similar mechanism may be responsible for the apparently higher vancomycin CL seen in the IVDA group; however, we were unable to demonstrate a statistically significant change. Therefore, this observation requires confirmation in a larger IVDA population before specific dosage recommendations can be made. Until more information is available, it would seem prudent to monitor this patient population closely, anticipating that IVDA may require more aggressive vancomycin dosing. LITERATURE CITED 1. Ackerman, B. H., H. Taylor, K. M. Olsen, W. Addel-Malak, and A. A. Pappas. 1988. Vancomycin serum protein binding determination by ultrafiltration. Drug Intell. Clin. Pharm. 22:300-303. 2. Bloedow, D. C., J. F. Hansbrough, T. Hardin, and M. Simmons. 1986. Postburn serum drug binding and serum protein concen-

trations. J. Clin. Pharmacol. 26:147-151.

3. Brater, D. C., R. E. Bawdon, S. A. Anderson, G. F. Purdue, and J. L. Hunt. 1986. Vancomycin elimination in patients with burn injury. Clin. Pharmacol. Ther. 39:631-634. 4. Fox, J. L., and M. L. Lamson. 1987. RSTRIP user handbook.

795

MicroMath, Inc., Salt Lake City, Utah. 5. Garaud, J. J., B. Regnier, F. Inglebert, F. Faurisson, J. Bauchet, and F. Vachon. 1984. Vancomycin pharmacokinetics in critically ill patients. J. Antimicrob. Chemother. 14(Suppl.):53-57. 6. Garrelts, J. C., and J. D. Peterie. 1988. Altered vancomycin dose vs. serum concentration relationship in burn patients. Clin. Pharmacol Ther. 44:9-13. 7. Gibaldi, M., and D. Pemer. 1982. Pharmacokinetics, 2nd ed., p. 414. Marcel Dekker, Inc., New York. 8. Golper, T. A., H. M. Noonan, L. Elzinga, D. Gilbert, R. Brummett, J. L. Anderson, and W. M. Bennett. 1988. Vancomycin pharmacokinetics, renal handling and nonrenal clearances in normal human subjects. Clin. Pharmacol. Ther. 43: 565-570. 9. Healy, D. P., R. E. Polk, M. L. Garson, D. T. Rock, and T. J. Comstock. 1987. Comparison of steady-state pharmacokinetics of two dosage regimens of vancomycin in normal volunteers. Antimicrob. Agents Chemother. 31:393-397. 10. King, C. H., R. J. Creger, and J. J. Ellner. 1985. Pharmacokinetics of tobramycin and gentamicin in abusers of intravenous drugs. Antimicrob. Agents Chemother. 27:285-290. 11. Klastersky, J., and P. Van der Auwera. 1986. Cephalosporins, vancomycin, aminoglycosides and other drugs, especially in combination, for the treatment of methicillin-resistant staphylcoccal infections. J. Antimicrob. Chemother. 17(Suppl. A): 19-24. 12. Loirat, P., J. Rohan, A. Baillet, F. Beaufils, R. David, and A. Chapman. 1978. Increased glomerular filtration rate in patients with major burns and its effect on the pharmacokinetics of tobramycin. N. Engl. J. Med. 299:915-919. 13. Morse, G. D., D. K. Nairin, J. S. Bertino, and J. J. Walsh. 1987. Overestimation of vancomycin concentrations utilizing fluorescence polarization immunoassay in patients on peritoneal dialysis. Ther. Drug Monit. 9:212-215. 14. Nivoche, Y., A. Contrepois, C. Cremieus, and C. Carbon. 1982. Vancomycin in rabbits: pharmacokinetics, extravascular diffusion, renal excretion and interactions with furosemide. J. Pharmacol. Exp. Ther. 222:237-240. 15. Rocci, M. L. Jr., and W. J. Jusko. 1983. LAGRAN program for area and moments in pharmacokinetic analysis. Comput. Programs Biomed. 16:203-216. 16. Rodvold, K. A., R. A. Blum, J. H. Fischer, H. Z. Zokufa, J. C. Rotschafer, K. B. Crossley, and L. J. Riff. 1988. Vancomycin pharmacokinetics in patients with various degrees of renal function. Antimicrob. Agents Chemother. 32:848-852. 17. Rybak, M. J., J. J. Frankowsky, D. J. Edwards, and L. M. Albrecht. 1987. Alanine aminopeptidase and P2-microglobulin excretion in patients receiving vancomycin and gentamicin. Antimicrob. Agents Chemother. 31:1461-1464. 18. Sawchuk, R. J. 1984. Drug absorption and disposition in burn patients, p. 333-348. In L. Z. Benent, Massoud, N. and J. G. Gambertoglio (ed.), Pharmacokinetic basis for drug treatment. Raven Press, Publishers, New York. 19. Sawchuk, R. J., and T. S. Rector. 1980. Drug kinetics in burn patients. Clin. Pharmacokinet. 4:548-556. 20. Smith, I. L., and J. J. Schentag. 1984. Noncompartmental determination of the steady-state volume of distribution during

multiple dosing. J. Pharm. Sci. 73:281-282.

21. Sorrel, T. C., D. R. Packham, S. J. Shanker, M. Foldes, and R. Munro. 1982. Vancomycin therapy for methicillin-resistant

Staphylococcus aureus. Ann. Intern. Med. 97:344-350.

22. White, L. O., R. Edwards, H. A. Holt, A. M. Lovering, R. G. Finch, and D. S. Reeves. 1988. The in-vitro degradation at 37°C of vancomycin in serum, CAPD fluid and phosphate-buffered saline. J. Antimicrob. Chemother. 22:739-745. 23. Zaske, D. E., and R. J. Cipolle. 1981. Rapid individualization of gentamicin dosage regimens in 66 burn patients. Burns 7: 215-220. 24. Zaske, D. E., R. J. Sawchuk, D. N. Gerding, and R. G. Strate. 1976. Increased dosage requirements of gentamicin in burn patients. J. Trauma 16:824-828. 25. Zaske, D. E., R. J. Sawchuk, and R. G. Strate. 1978. The necessity of increased doses of amikacin in burn patients. Surgery 84:603-608.

Vancomycin pharmacokinetics in burn patients and intravenous drug abusers.

The pharmacokinetics of vancomycin were evaluated in 34 patients (10 burn patients, 14 intravenous drug abusers [IVDA], and 10 controls). Multiple ser...
849KB Sizes 0 Downloads 0 Views