Characterization of Bactericidal Activity of Clindamycin against ...

4 downloads 0 Views 198KB Size Report
MICHAEL E. KLEPSER,1,2,3* MARY ANNE BANEVICIUS,1 RICHARD QUINTILIANI,2,4 .... Aldridge and Stratton reported results from time-kill studies.
ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Aug. 1996, p. 1941–1944 0066-4804/96/$04.0010 Copyright q 1996, American Society for Microbiology

Vol. 40, No. 8

Characterization of Bactericidal Activity of Clindamycin against Bacteroides fragilis via Kill Curve Methods MICHAEL E. KLEPSER,1,2,3* MARY ANNE BANEVICIUS,1 RICHARD QUINTILIANI,2,4 2,3 AND CHARLES H. NIGHTINGALE Department of Pharmacy1 and Office for Research,2 Hartford Hospital, Hartford, Connecticut 06102; School of Pharmacy, University of Connecticut, Storrs, Connecticut 062683; and School of Medicine, University of Connecticut, Farmington, Connecticut 060324 Received 20 November 1995/Returned for modification 7 March 1996/Accepted 20 May 1996

Kill curves were determined for five isolates of Bacteroides fragilis with clindamycin at concentrations equal to the MIC or to 4, 16, and 64 times the MIC. Examination of plots of log CFU per milliliter versus time revealed no association between the clindamycin concentration and the rate and extent of the bactericidal activity against B. fragilis at or below 64 times the MIC. The pharmacodynamic properties of a xenobiotic help one to characterize the interactions which occur between a pharmacologic agent and its site of action and the resultant effect. For antimicrobial agents, reference to pharmacodynamics generally implies the postantibiotic effect and/or the relationship between antibiotic concentration and the rate and extent of bactericidal activity. Characterization of the pharmacodynamic properties for agents such as the b-lactams and aminoglycosides has had significant ramifications on our utilization of these agents (3, 5, 8, 11, 12–14, 16, 17, 20, 23, 24). Although clindamycin has been used in clinical practice for quite some time, there still appear to be confusion and controversy regarding the appropriate dosing regimen of this agent (2, 4, 6, 10, 19, 22). In part, confusion over the selection of an appropriate dosing regimen stems from a general lack of understanding regarding the bactericidal characteristics of clindamycin. Since few data concerning the relationships between clindamycin concentrations and rate and extent of bactericidal activity exist, we examined these relationships via kill curve methods against Bacteroides fragilis. (This work was presented at the American College of Clinical Pharmacy Winter Practice and Research Forum, Orlando, Fla., 13 to 15 February 1995 [abstract 101] [11a].) Five isolates of B. fragilis, i.e., three clinical isolates (strains 11, 13, and 266) and two reference strains (ATCC 25285 and ATCC 23745), were selected for testing. The MIC of clindamycin against each microorganism was determined. MIC analyses were conducted via broth microdilution techniques according to the National Committee for Clinical Laboratory Standards procedures for anaerobic testing (15). MIC analyses were conducted with an inoculum of approximately 5 3 105 CFU/ml in prereduced enriched soybean-casein digest broth obtained from BACTEC NR7A vials (Becton Dickinson Diagnostic Instrument Systems, Towson, Md.). After 48 h of incubation at 378C under anaerobic conditions, MICs were recorded. Clindamycin powder for all testing procedures was provided by The Upjohn Company, Kalamazoo, Mich. Isolates were stored in skim milk at 2708C until their use. Prior to testing, the organisms were subcultured twice on blood agar plates. Ten milliliters of prereduced enriched soybean-

casein digest broth was then inoculated with approximately five to seven colonies from the second transfer plate and incubated anaerobically for 8 to 12 h at 378C to ensure that bacteria reached logarithmic growth phase prior to use in kill curve procedures. Subsequently, the bacterial suspension was visually adjusted to a 0.5 McFarland turbidity standard and then further diluted 1:100 with prereduced enriched soybean-casein digest broth to yield a starting inoculum of approximately 106 CFU/ml. Thirty milliliters of the adjusted bacterial solution was transferred into evacuated BACTEC NR7A vials. At time zero, a 0.1-ml sample was obtained from the control culture vial, serially diluted, and plated onto blood agar plates to verify the starting inoculum. The appropriate amount of aqueous clindamycin stock solution was then added to the test vials to produce clindamycin concentrations equal to the MIC or to 4, 16, and 64 times the MIC for the test isolate. No clindamycin was added to the control vial. The volume of clindamycin solution added to each vial composed less than 7% of the total volume in the vial. Solutions were then incubated at 378C under anaerobic conditions. At predetermined time points—4, 8, 12, 24, 36, and 48 h following the introduction of clindamycin into the system—0.1-ml samples were aseptically removed from each vial. Tenfold serial dilutions were performed on samples, and a 10-ml aliquot from each dilution was streaked onto a blood agar plate for colony count determination. When the number of CFU per milliliter was expected to be less than 1,000, a 10-ml sample was taken from sample vials and plated directly onto a blood agar plate without dilution. The lower limit of bacterial quantitation was 100 CFU/ml. Following incubation in an anaerobic jar equipped with a BBL GasPak Plus packet (Becton Dickinson Microbiology Systems) at 378C for 48 h, the number of CFU on each plate was determined. Kill curve experiments were run in duplicate. Mean colony count data (log CFU per milliliter) versus time were plotted for each isolate and used for visual comparisons of the rate and extent of bactericidal activity. The time to achieve a 99.9% reduction from the initial inoculum was determined for each organism at the various clindamycin concentrations. The median MICs for the five test isolates for clindamycin ranged from 0.03 mg/ml for ATCC 23745 to 8.0 mg/ml for B. fragilis 266 (Table 1). Kill curve results are presented in Fig. 1. Clindamycin exhibited a marked bactericidal effect, defined by $3-log decrease from the starting inoculum, for ATCC 23745, ATCC 25285, and B. fragilis 11 (Fig. 1A to C). The rates of kill noted for these isolates (assessed by the time to a 99.9%

* Corresponding author. Present address: College of Pharmacy, University of Iowa, S412 Pharmacy Building, Iowa City, IA 522421112. Phone: (319) 335-8861. Fax: (319) 353-5646. Electronic mail address: [email protected]. 1941

1942

NOTES

ANTIMICROB. AGENTS CHEMOTHER.

TABLE 1. Mean time to 99.9% reduction from starting inoculum for test isolates at various multiples of the clindamycin MIC B. fragilis isolate

ATCC 23745 ATCC 25285 11 13 266 a

Mean time (h) to 99.9% reduction from starting inoculum at multiple of MIC

MIC (mg/ml)a

0.03 0.5 0.5 1.0 8.0

1

4

16

64

24 48 24 Not obtained Not obtained

24 48 24 36 Not obtained

24 48 24 12 Not obtained

24 48 24 12 Not obtained

Breakpoints for clindamycin against B. fragilis: #2 mg/ml (susceptible), 4 mg/ml (intermediate), and $8 mg/ml (resistant) (15).

reduction in CFU per milliliter from the starting inoculum [Table 1]) were not influenced by increasing clindamycin concentrations above the MICs for the test isolates. Likewise, the extent of bactericidal activity (i.e., the absolute reduction in CFU per milliliter) was also independent of clindamycin concentration for these three isolates. In contrast, the killing kinetics observed with strain 13 were slightly different from those observed with ATCC 23745, ATCC 25285, and B. fragilis 11. At a concentration equal to the MIC for this isolate (1.0 mg/ml), clindamycin initially produced a bacteriostatic effect. However, by 24 h, the CFU per milliliter had increased by 1 log over the initial inoculum, and at 36 h bacterial counts were equal to the growth control values. At four times the MIC, a slow bactericidal effect was observed, and a 99.9% reduction in CFU per milliliter was achieved at 36 h. Further increasing the concentration of clindamycin to 16 times the MIC again resulted in an increase in the rate of kill; however, increasing the clindamycin concentration from 16 to 64 times the MIC did not result in further enhancement in the rate of kill. Against B. fragilis 266, clindamycin produced roughly a 2-log decrease in CFU and was therefore not considered bactericidal. However, the rate and extent of the decline in viable CFU were not enhanced by antibiotic concentrations greater than the MIC for the isolate (Fig. 1E). Recognition of the importance of pharmacodynamics and incorporation of pharmacodynamics into clinical practice have had a significant impact on our approach to antimicrobial therapy. Antibiotic dosing according to pharmacodynamic principles theoretically allows clinicians to take full advantage of an agent’s bactericidal activity and potentially minimize drug-associated toxicities. Clindamycin is an antimicrobial agent which remains remarkably active against many grampositive aerobic and anaerobic organisms despite extensive clinical use. Additionally, clindamycin has excellent bioavailability and a relatively long half-life, characteristics which make it an attractive agent for oral therapy. However, many clinicians elect to prescribe alternative agents amidst fears of Clostridium difficile superinfections and poor tolerance by patients. In this study, we observed concentration-independent activity of clindamycin against four strains of B. fragilis (ATCC 23745, ATCC 25285, B. fragilis 11, and B. fragilis 266), with MICs ranging from 0.03 to 8.0 mg/ml. Against three susceptible strains, clindamycin demonstrated markedly bactericidal activity; however, bacteriostatic activity against a fourth resistant strain, for which the MIC was 8.0 mg/ml, was observed. In contrast, a slightly different killing profile was noted for B. fragilis 13, for which the MIC was 1.0 mg/ml. Against this isolate, there appeared to be a correlation between clindamycin concentrations and the rate of bactericidal activity up to 16 times the MIC. Above 16 times the MIC, however, the rate of bactericidal activity was not associated with the clindamycin concentration. The reason for this observed difference in bac-

tericidal activity among strains is not known. The MBC for B. fragilis 13 was determined to detect possible tolerance; however, good correlation between the MIC and MBC was observed (1 and 2 mg/ml, respectively). The relationships between the concentration of an antimicrobial agent and the rate and extent of bactericidal activity can be illustrated by a sigmoidal dose-response curve (Emax model). The activities of all antimicrobial agents, regardless of the concentration-killing relationship, can be illustrated by this model. Therefore, all antibacterial agents possess regions in their dose-response curves characterized by concentration dependence and independence; however, the ranges of concentrations over which these regions exist vary among compounds and are manifested by the slope of the curve. Agents which possess concentration-independent activity are characterized by a steeper dose-response curve than those exhibiting concentration-dependent bactericidal activity. In the case of clindamycin, bacterial killing most likely increases as the amount of clindamycin increases until the steep portion of the curve is reached. After this threshold is achieved, no additional killing is observed with increasing drug levels. Similar observations have been made with the b-lactams (7, 9, 18). Previously, Aldridge and Stratton reported results from time-kill studies with various agents, including clindamycin, conducted with members of the B. fragilis group (1, 21). In those reports, the authors hint that the bactericidal activity of clindamycin may be dependent on the concentration over the range of 0.5 to 4 times the MIC. The observations made regarding concentration and bactericidal activity can be explained by the failure to test multiples of the MIC sufficient to reach the transition portion of the dose-response curve. The authors presented similar results for ceftizoxime and cefotetan (1). Our data suggest that clindamycin possesses concentrationindependent killing characteristics. Clinically, these findings may significantly alter our approach to clindamycin dosing. Previously, dosages ranging from 600 mg every 6 or 8 h to 900 mg every 8 h to 1,200 mg every 12 h have been examined as potential dosing regimens for intravenous clindamycin (2, 4, 6, 10, 22). Our findings suggest that the ideal dosing regimen for clindamycin is one that optimizes the amount of time that concentrations in serum remain above the MIC for an organism while simultaneously minimizing the patient’s exposure to the drug. Although each of the above-mentioned dosing regimens provides suprainhibitory concentrations of clindamycin throughout their respective dosing intervals, patients may be exposed to greater quantities of clindamycin than is necessary for a beneficial outcome. We have recently completed an evaluation of intravenous and oral clindamycin at a 300-mg dose administered every 8 and 12 h in a group of healthy volunteers (11b). The duration of antibacterial activity of these regimens was determined by calculating the amount of serum bactericidal activity against multiple isolates of B. fragilis and a variety of gram-positive aerobic microorganisms. Serum bactericidal

FIG. 1. Mean log CFU per milliliter versus time for B. fragilis ATCC 23745 (MIC 5 0.03 mg/ml) (A), B. fragilis ATCC 25285 (MIC 5 0.5 mg/ml) (B), B. fragilis 11 (MIC 5 0.5 mg/ml) (C), B. fragilis 13 (MIC 5 1.0 mg/ml) (D), and B. fragilis 266 (MIC 5 8.0 mg/ml) (E). E, control; }, MIC; w, 4 times the MIC; h, 16 times the MIC; {, 64 times the MIC. Dashed line, limit of quantitation (100 CFU/ml).

1943

1944

NOTES

ANTIMICROB. AGENTS CHEMOTHER.

titers demonstrated measurable bactericidal activity throughout each of the dosing intervals for the anaerobic isolates and for several of the gram-positive isolates tested. These data provide further evidence of the concentration-independent antibacterial activity of clindamycin. In summary, the findings of this study provide clinicians and investigators with much-needed pharmacodynamic data regarding the bactericidal nature of clindamycin against B. fragilis. These data, which point to the existence of concentrationindependent bactericidal activity for clindamycin, may be used as the foundation for the future development of optimal clindamycin dosing regimens. REFERENCES 1. Aldridge, K. E., and C. W. Stratton. 1991. Bactericidal activity of ceftizoxime, cefotetan, and clindamycin against cefoxitin-resistant strains of the Bacteroides fragilis group. J. Antimicrob. Chemother. 28:701–705. 2. Ameer, B., P. Sesin, and A. W. Karchmer. 1987. Selecting clindamycin dosing regimens. Am. J. Hosp. Pharm. 44:2027–2028. 3. Begg, E. J., B. A. Peddie, S. T. Chambers, and D. R. Boswell. 1992. Comparison of gentamicin dosing regimens using an in-vitro model. J. Antimicrob. Chemother. 29:427–433. 4. Buchwald, D., S. B. Soumerai, N. Vandevanter, M. R. Wessels, and J. Avorn. 1989. Effect of hospitalwide change in clindamycin dosing schedule on clinical outcome. Rev. Infect. Dis. 11:619–624. 5. Bundtzen, R. W., A. U. Gerber, D. L. Cohn, and W. A. Craig. 1981. Postantibiotic suppression of bacterial growth. Rev. Infect. Dis. 3:28–37. 6. Chin, A., M. A. Gill, M. K. Ito, A. E. Yellin, T. V. Berne, P. N. R. Heseltine, M. D. Appleman, and F. C. Chenella. 1989. Evaluation of two different dosage regimens of clindamycin and the penetration into human appendix. Ther. Drug Monit. 11:421–424. 7. Craig, W. A., and S. C. Ebert. 1991. Killing and regrowth of bacteria in vitro: a review. Scand. J. Infect. Dis. Suppl. 74:147–154. 8. Davis, B. D. 1987. Mechanism of bactericidal action of aminoglycosides. Microbiol. Rev. 51:341–350. 9. Drusano, G. L. 1991. Human pharmacodynamics of beta-lactams, aminoglycosides and their combinations. Scand. J. Infect. Dis. Suppl. 74:235–248. 10. Flaherty, J. F., L. C. Rodondi, B. J. Guglielmo, J. C. Fleishaker, R. J. Townsend, and J. G. Gambertoglio. 1988. Comparative pharmacokinetics and serum inhibitory activity of clindamycin in different dosing regimens. Antimicrob. Agents Chemother. 32:1825–1829. 11. Isaksson, B., L. Nilsson, R. Maller, and L. So¨re´n. 1988. Postantibiotic effect of aminoglycosides on gram-negative bacteria evaluated by a new method. J. Antimicrob. Chemother. 22:23–33. 11a.Klepser, M. E., M. A. Banevicius, R. Quintiliani, and C. H. Nightingale. 1995. Characterization of the bactericidal activities of clindamycin, metronidazole, and piperacillin/tazobactam against Bacteroides fragilis. Pharmaco-

therapy 15:123. (Abstract 101.) 11b.Klepser, M. E., D. P. Nicolau, R. Quintiliani, and C. H. Nightingale. 1996. Comparison of the bactericidal activity of clindamycin administered every eight and twelve hours against Staphylococcus aureus, Streptococcus pneumoniae, and Bacteroides fragilis. Pharmacotherapy 16:501. (Abstract 74.) 12. Leggett, J. E., B. Fantin, S. Ebert, K. Totsuka, B. Vogelman, W. Calame, H. Mattie, and W. A. Craig. 1989. Comparative antibiotic dose-effect relations at several dosing intervals in murine pneumonitis and thigh-infection models. J. Infect. Dis. 159:281–292. 13. Livingston, D. H., and M. T. Wang. 1993. Continuous infusion of cefazolin is superior to intermittent dosing in decreasing infection after hemorrhagic shock. Am. J. Surg. 165:203–207. 14. Mordenti, J. J., R. Quintiliani, and C. H. Nightingale. 1985. Combination antibiotic therapy: comparison of constant infusion and intermittent bolus dosing in an experimental animal model. J. Antimicrob. Chemother. 15(Suppl. A):313–321. 15. National Committee for Clinical Laboratory Standards. 1990. Methods for antimicrobial susceptibility testing of anaerobic bacteria. Approved standard M11-82. National Committee for Clinical Laboratory Standards, Villanova, Pa. 16. Nicolau, D. P., C. D. Freeman, P. P. Belliveau, C. H. Nightingale, J. W. Ross, and R. Quintiliani. 1995. Experience with a once-daily aminoglycoside program administered to 2,184 adult patients. Antimicrob. Agents Chemother. 39:650–655. 17. Nishida, M., T. Murakawa, T. Kamimura, and N. Okada. 1978. Bactericidal activity of cephalosporins in an in vitro model simulating serum levels. Antimicrob. Agents Chemother. 14:6–12. 18. Onyeji, C. O., D. P. Nicolau, C. H. Nightingale, and R. Quintiliani. 1994. Optimal times above MICs of ceftibuten and cefaclor in experimental intraabdominal infections. Antimicrob. Agents Chemother. 38:1112–1117. 19. Plaisance, K. I., G. L. Drusano, A. Forrest, R. J. Townsend, and H. C. Standiford. 1989. Pharmacokinetic evaluation of two dosage regimens of clindamycin phosphate. Antimicrob. Agents Chemother. 33:618–620. 20. Roosendaal, R., I. A. J. M. Bakker-Woudenburg, J. C. van den Berg, and M. F. Michel. 1985. Therapeutic efficacy of continuous versus intermittent administration of ceftazidime in an experimental Klebsiella pneumoniae pneumonia in rats. J. Infect. Dis. 152:373–378. 21. Stratton, C. W., L. S. Weeks, and K. E. Aldridge. 1987. Comparison of kill-kinetic studies with agar and broth microdilution methods for determination of antimicrobial activity of selected agents against members of the Bacteroides fragilis group. J. Clin. Microbiol. 25:645–649. 22. Townsend, R. J., and R. P. Baker. 1987. Pharmacokinetic comparison of three clindamycin phosphate dosing schedules. Drug Intell. Clin. Pharm. 21:279–281. 23. Vogelman, B., S. Gudmundsson, J. Turnidge, J. Leggett, and W. A. Craig. 1988. In vivo postantibiotic effect in a thigh infection in neutropenic mice. J. Infect. Dis. 157:287–298. 24. Zeisler, J. A., J. D. McCarthy, W. A. Richelieu, and M. B. Nichol. 1992. Cefuroxime by continuous infusion: a new standard of care? Infect. Med. 11:54–60.