Postantibiotic effect of trovafloxacin against Streptococcus pneumoniae, Haemophilus influenzae, and Neisseria meningitidis in cerebrospinal fluid and broth culture media

Diagnostic Microbiology and Infectious Disease 36 (2000) 241–247

Antimicrobial susceptibility studies

Postantibiotic effect of trovafloxacin against Streptococcus pneumoniae, Haemophilus influenzae, and Neisseria meningitidis in cerebrospinal fluid and broth culture media Pamela R. Tessiera, Charles H. Nightingalec,d, David P. Nicolaua,b,d,* a

Department of Pharmacy Research b Division of Infectious Diseases c Office for Research, Hartford Hospital, Hartford, CT 06102 d School of Pharmacy, University of Connecticut, Storrs, CT 06268 Received 30 September 1999; revised and accepted 6 December 1999

Abstract Trovafloxacin displays exceptional antimicrobial potency against pathogens associated with community-acquired meningitis. The postantibiotic effect (PAE) of trovafloxacin was assessed against two clinical strains each of Streptococcus pneumoniae, Haemophilus influenzae Type B, and Neisseria meningitidis Group B. Testing was performed simultaneously in broth culture media (broth) and pooled CSF at concentrations equivalent to 0.5X and 4X MIC. Mean PAEs in broth at 0.5X and 4X MIC ranged between 0.07 to 1.10 and 0.57 to 5.83 h and between 0.07 to 1.67 and 0.47 to 6.00 h in CSF for all organisms. Overall, the incorporation of CSF did not augment or diminish the duration of the trovafloxacin-induced PAE. These data, together with its pharmacokinetic profile in CSF and antimicrobial potency against these isolates, make trovafloxacin an agent of interest for the treatment of meningitis. © 2000 Elsevier Science Inc. All rights reserved.

1. Introduction The postantibiotic effect (PAE) describes the suppression of bacterial growth after a brief exposure to an antimicrobial agent after removal of the agent. Although the nature of the PAE is still not understood completely, its largest effect is on the design of appropriate dosing regimens (Craig and Gudmundsson, 1996; Gould, 1997). In addition, this effect may be of great importance in meningitis, a site where decreased antimicrobial penetration and rapid efflux hamper optimal drug exposure. At present, many antimicrobials demonstrate in vitro PAEs and the vast majority of these experiments have been conducted in commercially marketed broth culture media (Boswell et al., 1997a and b; Pankuch et al., 1998; Spangler

* Corresponding author. Tel.: ⫹1-860-545-3941; fax: ⫹1-860-5453992. E-mail address: [email protected] (D.P. Nicolau).

et al., 1998). Whereas the standardization and availability of broth media make them a frequently used component of PAE investigations, the incorporation of biological fluids such as serum, urine, and cerebrospinal fluid (CSF) have been reported to markedly alter PAE duration (Karlowsky et al., 1993; Zhanel et al., 1992 and 1991). Trovafloxacin displays an exceptional spectrum of in vitro activity and seems to penetrate well into the CSF (Cutler et al., 1997). As a result, preliminary animal and human studies suggest that this agent may be an important option in the wake of changing antimicrobial susceptibilities to commonly used agents in the treatment of meningitis (Hopkins et al., 1996; Kim et al., 1997). Although trovafloxacin has demonstrated a PAE (Boswell et al., 1997a; Pankuch et al., 1998), studies have not been presented that investigate the influence of CSF on the measurement of this microbiologic phenomenon. The purpose of this study was to assess the influence of CSF on the PAE of trovafloxacin for several pathogens, which may be present in infections of the central nervous system.

0732-8893/00/$ – see front matter © 2000 Elsevier Science Inc. All rights reserved. PII: S 0 7 3 2 - 8 8 9 3 ( 9 9 ) 0 0 1 4 8 - 0

242

P.R. Tessier et al. / Diagnostic Microbiology and Infectious Disease 36 (2000) 241–247

2. Materials and Methods 2.1. Antimicrobials Analytical grade trovafloxacin mesylate powder (Lot No. 31075-199-1F, potency 798 ␮g/mg, expiration 3/2000) was obtained from Pfizer Inc., Groton CT. Once prepared, solutions of trovafloxacin were stored at ⫺80°C until use. 2.2. Bacterial strains and test media Two clinical strains each of Streptococcus pneumoniae, Haemophilus influenzae Type B, and Neisseria meningitidis Group B were used. The S. pneumoniae and N. meningitidis were clinical isolates obtained from patients at our institution, whereas the H. influenzae isolates were acquired from the State of Connecticut Department of Public Health Laboratory. Broth media used for the S. pneumoniae isolates was 5% lysed horse blood in cation-adjusted Mueller-Hinton broth (LHB-CAMHB, Becton Dickinson, Cockeysville, MD); Haemophilus test medium (HTM, Becton Dickinson) for H. influenzae, and HTM with LHB for the N. meningitidis isolates. Discarded CSF was acquired from patients with venticulostomies in our neurosurgical intensive care unit. Over a 3-week period CSF was obtained from eight individuals who were not being treated for either a central nervous or systemic infection. Each CSF sample, ranging from 50 to 450 mL, was treated aseptically and processed individually. After centrifugation, the supernatant was removed and tested for pH, total protein, sterility, antimicrobial activity, and its ability to inhibit or support growth of test isolates in varying percent CSF solutions (vol/vol with broth). The pH was determined using paper test strips (pHydrion, Micro Essential Laboratory, Inc., Brooklyn, NY 11210) having a range of 1–12 and using aqueous pH standards of 4.0 and 7.0 (Lots 1811551 and 1901418; expiration 11/2000 and 1/2001, respectively, Ricca Chemical Co, Arlington, TX 76012). The clinical laboratory assessed total protein levels. For sterility, each CSF was directly plated onto a Trypticase Soy agar plate with 5% sheep blood (BAP, Becton Dickinson) and chocolate agar plate (CHOC, Becton Dickinson), and 1 mL incubated in Haemophilus Test medium (HTM, Becton Dickinson) for 48 h. Visual turbidity in broth or CFU growth on agar was considered positive for bacterial growth and the CSF sample deemed contaminated. Each CSF sample was examined for the presence of antimicrobial activity against each test organism by the following procedure. A 0.5 McFarland standard inoculum of each isolate in the appropriate broth media was prepared and applied onto a BAP (S. pneumoniae and N. meningitidis isolates) or CHOC (H. influenzae isolates) with a sterile cotton swab. A total 20 ␮L of each CSF was dispensed onto a sterilized 1⁄4 inch paper disk (Lot F853, No. 740-E, Schleicher & Schuell Inc., Keene, NH) and placed onto the surface of the agar plate along with a 20-␮L drop of each CSF. All agar plates were incubated

overnight at 35°C in 5% CO2 and then examined for inhibition of organism growth. The ability of the test isolate to grow in individual CSF sample-containing solutions was also ascertained. A culture of the test isolates in logarithmic growth phase was inoculated into 100% broth, and 50% and 90% CSF-containing solutions (vol/vol, 50% and 10% broth, respectively) of each CSF at a starting inoculum of 105 CFU/mL. The CFU/mL in each solution was assessed initially, then serially over 24 h. Adequate growth of test isolates in a CSF sample was defined as a return to logarithmic growth (i.e., a growth pattern similar to that in 100% broth) over the sampling time. CSF that did not support the growth of any one test isolate was excluded from use in PAE experiments. Based on these analyses, four samples were not used because of nonsterility or inhibitory effects on the growth of a test isolate. The remaining four CSF samples from four different patients were combined. The combined CSF had a pH of 7.5 and total protein 68.0 mg/dL. Additionally, it was determined that the S. pneumoniae and N. meningitidis grew in a 90% CSF solution (vol/vol in broth), whereas the maximal concentration, which supported the growth of the H. influenzae isolates, was a 50% solution. As a result, these CSF concentrations were used throughout the PAE experiments. 2.3. Susceptibility determinations Microdilution (S. pneumoniae isolates) and E-test (all eight study isolates) MICs were performed in duplicate for each organism according to current NCCLS guidelines. MICs were incubated at 35°C for 18 –24 h in 5% CO2 for E-tests and in ambient air for microdilutions. Detection of ␤-lactamase enzyme production from H. influenzae and N. meningitidis isolates was preformed using nitrocefincontaining disks (Cefinase, Lot 1006903537, expiration 4/1/2000, Becton Dickinson). 2.4. PAE determinations The PAEs were determined for each organism in both broth and CSF on three or more separate occasions using conventional methods (Craig and Gudmundsson, 1996; MacKenzie and Gould, 1993). Briefly, a trovafloxacin stock solution was added to prewarmed tubes (35°C incubator with 5% CO2) containing broth or CSF and the test isolate to achieve concentrations equivalent to 0.5 and 4 X MIC of the isolate. The final inoculum in each tube was 105 CFU/ mL. The singular exposure to trovafloxacin lasted one hour, then each culture was diluted 1:100 into either fresh broth or CSF to effectively eliminate the antimicrobial. This represented the time zero for PAE calculations. Broth and CSF controls consisting of tubes containing no antibiotic and 4 X MIC diluted 1:100 (to simulate residual antibiotic concentrations) were also included and were handled in a similar fashion to those exposed to trovafloxacin. PAE cultures were maintained at 35°C in a 5% CO2 incubator over the

P.R. Tessier et al. / Diagnostic Microbiology and Infectious Disease 36 (2000) 241–247 Table 1 Mean duration of postantibiotic effect (⫾SD) of trovafloxacin against Streptococcus pneumoniae, Haemophilus influenzae, and Neisseria meningitidis in broth culture media and CSF Isolated

MIC (␮g/mL)

Matrix

0.5 X MIC

4 X MIC

S. pneumoniae 8

0.5

S. pneumoniae 66

0.06

H. influenzae 44

0.016

H. influenzae 46

0.012

N. meningitidis 1

0.003

N. meningitidis 2

0.003

Broth CSF Broth CSF Broth CSF Broth CSF Broth CSF Broth CSF

0.57 (0.49) 0.30 (0.30) 0.37 (0.40) 0.07 (0.12) 0.87 (0.81) 1.67 (0.85) 0.43 (0.40) 1.40 (1.21) 1.10 (1.14) 0.47 (0.81) 0.07 (0.12) 0.97 (0.91)

2.00 (0.62)a 1.80 (0.95)b 0.57 (0.74) 0.47 (0.32) 3.47 (0.59)a 6.00 (0.98)a,c 3.77 (0.49)a 2.97 (2.26) 5.83 (0.75)a 1.63 (1.37)c 4.23 (1.29)a 3.70 (1.41)a

Significant difference versus 0.5 X MIC, p ⬍ 0.05. Nonsignificant difference versus 0.5 X MIC, p ⫽ 0.06. c Significant difference versus broth, p ⬍ 0.05. d Internal strain designation number. a

b

sampling period and viable bacterial counts were assessed at hourly intervals for 8 h and again at 12 and 24 h. At each sampling time, 10 to 100 ␮L of two or three dilutions of each sample were cultured onto either BAP (S. pneumoniae and N. meningitidis) or CHOC (H. influenzae) agar media (Becton Dickinson). PAEs were calculated as follows: PAE ⫽ T ⫺ C, where T is the time (in hours) required for the CFU/mL in the test culture to increase one log10 above the count observed immediately after dilution (time zero) and C is the time required for the CFU/mL in the untreated control culture to increase one log10 above the count observed immediately after dilution. 2.5. Statistical analysis The PAE in each test media was compared between MIC exposures using an unpaired Student’s t-test. The PAE between test media, the time required for the growth control (no antibiotic) to increase one log10 in the test media and the difference in the rate of growth between control (no antibiotic) and the residual antibiotic control were evaluated using the same statistical test. A p value of ⬍0.05 was considered significant. All experiments are reported as the mean value ⫾ SD in hours.

3. Results The median MIC of each test isolate is reported in Table 1. These data reveal that both H. influenzae and N. meningitidis isolates were highly susceptible to trovafloxacin. Although considered susceptible, the MICs of the S. pneumoniae isolates were 100-fold higher than that of the H. influenzae and N. meningitidis. Both N. meningitidis isolates

243

were ␤-lactamase negative as well as H. influenzae 44; H. influenzae 46 was positive. Representative growth curves for each of the test species are displayed in Figs. 1–3. No significant difference was detected between bacterial growth in the non-antibiotic growth controls versus the residual antibiotic growth controls for either broth or CSF for all isolates. These data substantiate the adequacy of the selected dilution technique in minimizing residual trovafloxacin and confirm a lack of interference of sub-MIC effects on the length of PAE in these experiments. The mean duration of the PAE (⫾SD) in hours for each of the six isolates in both matrix tested are displayed in Table 1. Each value represents three or more replicates. Mean PAEs in either matrix at 0.5 X and 4 X MIC ranged between 0.07 to 0.57 and 0.47 to 2.00 h; 0.43 to 1.67 and 2.97 to 6.00 h; and 0.07 to 1.10 and 1.63 to 5.83 h, respectively, for S. pneumoniae, H. influenzae, and N. meningitidis isolates. Trovafloxacin PAEs produced by a concentration of 0.5 X MIC was less than that observed after exposure to 4 X MIC for all isolates. This observation would be expected when testing antimicrobial with concentration-dependent killing such as fluoroquinolones, and the fact that this was observed tends to verify our study results. The higher exposure generally produced a PAE that was more than four times longer irrespective of the test media. For all isolates except S. pneumoniae 66, the PAE at 4 X MIC was significantly different than 0.5 X MIC when tested in broth. In contrast, only H. influenzae 44 and N. meningitidis 2 displayed a significant difference in PAE between 0.5 X versus 4 X MIC in CSF; however, S. pneumoniae 8 resulted in p ⫽ 0.06. None of the PAE values from 0.5 X MIC between broth and CSF were significantly different; and in only two instances were differences found at 4 X MIC. For one isolate, H. influenzae 44, the CSF PAE was longer than that in broth media; for N. meningitidis 1 the reverse was true. Proliferation of all isolates in control tubes containing broth media was more rapid than in CSF with the exception of N. meningitidis 2, with significant differences noted for both S. pneumoniae and H. influenzae isolates (data not shown). It is conceivable that the chemical properties of CSF, possibly a lack of sustaining nutrients, may have contributed to these observations of slowed rate of growth, especially for the H. influenzae isolates. In contrast, N. meningitidis 2, displayed an exceptionally rapid and significantly faster growth rate in CSF than in broth media ( p ⫽ 0.016). Although this isolate grew similarly to N. meningitidis 1 in broth media over several hours (data not shown), growth was obviously enhanced by the presence of CSF.

4. Discussion Recently, trovafloxacin has been shown to elicit a PAE from Gram-negative organisms such as Pseudomonas aeruginosa and Escherichia coli (Boswell et al., 1997a;

244

P.R. Tessier et al. / Diagnostic Microbiology and Infectious Disease 36 (2000) 241–247

Fig. 1. Growth of S. pneumoniae 66 in broth (A) or CSF (B) for growth controls (no antibiotic ⽧, residual antibiotic control Œ) and after trovafloxacin exposure at 0.5X MIC (췦) and 4X MIC (G).

Pankuch et al., 1998), and against Gram-positive organisms including S. pneumoniae and Staphylococcus aureus (Giamarellou-Bourboulis et al., 1999; Pankuch et al., 1998). Other fluoroquinolone antimicrobial agents have likewise demonstrated PAEs against Gram-negative and positive organisms (Boswell et al., 1997b; Fuchs et al., 1997; Licata et

al., 1997; Spangler et al., 1998). However, these PAE investigations have been performed using traditional culture media such as Mu¨eller Hinton broth. This current investigation was undertaken to assess the effect CSF on the PAE induced by trovafloxacin against bacterial isolates frequently associated with community-acquired meningitis.

P.R. Tessier et al. / Diagnostic Microbiology and Infectious Disease 36 (2000) 241–247

245

Fig. 2. Growth of H. influenzae 44 in broth (A) or CSF (B) for growth controls (no antibiotic ⽧, residual antibiotic control Œ) and after trovafloxacin exposure at 0.5X MIC (췦) and 4X MIC (G).

In earlier investigations, the addition of biological fluids such as urine and CSF into in vitro PAE testing resulted in a prolongation of PAE on E. coli (Karlowsky et al., 1993; Zhanel et al., 1992 and 1991). Although limited in number, these studies demonstrated that these fluids produced significant differences between the PAEs obtained for this Gram-negative organism. Our results with H. influenzae at 0.5X MIC seemed to corroborate these findings, although statistical differences were not found. At 4X MIC, one H. influenzae isolate demonstrated an augmented PAE in CSF

that was statistically different versus broth; the PAEs for the other H. influenzae isolate were similar between matrix. This may, as mentioned above, be attributable to the inhibitory effect of CSF on the growth of this organism. S. pneumoniae and N. meningitidis offer contrasting results. No differentiation between PAEs in CSF and broth at either concentration were present with one exception for N. meningitidis 1 at 4X MIC, where broth PAEs were three- to four-fold higher than CSF. The majority of S. pneumoniae and N. meningitidis CSF PAEs were of less duration than in broth, indi-

246

P.R. Tessier et al. / Diagnostic Microbiology and Infectious Disease 36 (2000) 241–247

Fig. 3. Growth of N. meningitidis 1 in broth (A) or CSF (B) for growth controls (no antibiotic ⽧, residual antibiotic control Œ) and after trovafloxacin exposure at 0.5X MIC (췦) and 4X MIC (G).

cating a possible protein-drug binding consequence. Also, because both H. influenzae isolates were tested in a lesser concentration of CSF, the resultant higher PAEs in CSF indicates that this is probable and may clarify the dissimilarity between H. influenzae and both S. pneumoniae and N. meningitidis results. Fluoroquinolone antibiotics, because of their lipophilic na-

ture, diffuse well into the CNS. The penetration of trovafloxacin into CSF of healthy volunteers after a 400-mg infusion was rapid and sustained concentrations of ⱖ0.2 ␮g/mL over 24 h in a recent study by Cutler et al., (1997). Quinolones are generally accepted as having concentration-dependent bacterial killing, therefore a large CPEAK/MIC ratio correlates with bactericidal effect. Because typical MIC90 values for H. influenzae and N.

P.R. Tessier et al. / Diagnostic Microbiology and Infectious Disease 36 (2000) 241–247

meningitidis are considerably lower at 0.03 and 0.004 ␮g/mL (Doern 1998), respectively, a 7- to 50-fold multiple of the MIC would be maintained throughout a 24-h dosing interval and a favorable treatment outcome might be predicted. Current S. pneumoniae isolates, on the contrary, possesses an MIC90 of 0.12– 0.19 ␮g/mL (Doern 1998; Jones et al., 1998) and diminished CPEAK/MIC ratios are noted. In this later situation, the role of the PAE is expected to play a more important role in therapeutic outcome. Information gathered from recent in vivo investigations indicates that perhaps, in meningitis, trovafloxacin exhibits adequate bactericidal activity. In animal models of experimental pneumococcal meningitis, treatment with trovafloxacin was found to be comparable with vancomycin and ceftriaxone (Paris et al., 1995). Nau et al. (1995) observed a correlation between trovafloxacin bactericidal rate and concentration in CSF. In comparison with ceftriaxone, trovafloxacin was found to be as effective, with the calculated ratio of mean trovafloxacin concentration in CSF to minimal bactericidal concentration of 6.8. In a similar experimental meningitis model caused by high-level penicillin resistant S. pneumoniae as investigated by Kim et al. (1997), doses of trovafloxacin ranging from 10 to 30 mg/kg also demonstrated an augmented rate of kill versus ceftriaxone (10 mg/kg). Ratios of concentration in CSF to serum ranged from 2 to 15-fold. Finally, the pharmacodynamic profile in combination with PAE and favorable CSF pharmacokinetics were suggested as an explanation for the effectiveness of trovafloxacin treatment of experimental meningitis (Lutsar et al. 1998). In summary, our data reveal that sustained PAEs are observed with trovafloxacin in CSF and that these values were not substantially different from that determined in standard broth culture media. These data together with the pharmacokinetic profile in CSF and the antimicrobial potency against the isolates most commonly associated with community-acquired bacterial meningitis make trovafloxacin an agent, which warrants further investigation for this life-threatening disease. Acknowledgment We thank Dr. Yvette McCarter (Hartford Hospital Microbiology Laboratory) and Mr. Robert Howard (CT State Laboratory) for their contribution of the test isolates, and Mingkang Zhong, Christina Turley, and Myo Kim for their technical support. This study was supported by a grant from Pfizer Inc., New York, NY. References Boswell, F. J., Andrews, J. M., & Wise, R. (1997a). Postantibiotic effect of trovafloxacin on Pseudomonas aeruginosa. J Antimicrob Chemother 39, 811– 814. Boswell, F. J., Andrews, J. M., & Wise, R. (1997b). Pharmacodynamic properties of BAY 12-8039 on Gram-positive and Gram-negative organisms as demonstrated by studies of time-kill kinetics and postantibiotic effect. Antimicrob Agents Chemother 41(6), 1377–1379. Craig, W. A. & Gudmundsson, S. (1996). Postantibiotic Effect. In: Antibiotics in Laboratory Medicine, 4th ed. Ed, Lorian V., Baltimore: Williams & Wilkins, pp. 296 –329.

247

Cutler, N. R., Vincent, J., Jhee, S. S., Teng, R., Wardle, T., Lucas, G., Dogolo, L. C., & Sramek, J. J. (1997). Penetration of trovafloxacin into cerebral spinal fluid in humans following intravenous infusion of alatrovafloxacin. Antimicrob Agents Chemother 41, 1298 –1300. Doern, G. V. (1998). Trovafloxacin-a new extended-spectrum fluoroquinolone with real potential for the management of infections due to Streptococcus pneumoniae. Formulary 33(suppl 3), S21–S26. Fuchs, P. C., Barry, A. L., & Brown, S. D. (1997). Streptococcus pneumoniae killing rate and post-antibiotic effect of levofloxacin and ciprofloxacin. J Chemother 9(6), 391–393. Giamarellou-Bourboulis, E. J., Sambatakou, H., Grecka, P., Chryssouli, Z., & Giamarellou, H. (1999). Sitafloxacin (DU-6859a) and Trovafloxacin: Postantibiotic effect and in vitro interactions with rifampin a methicillinresistant Staphylococcus aureus. Diagn Microbiol Infect Dis 34, 301–307. Gould, I. M. (1997). Pharmacodynamics and the relationship between in vitro and in vivo activity of antimicrobial agents. J Chemother 9(suppl 1), 74 – 83. Hopkins, S., Williams, D., Dunne, M., Marinovich, L., Edeline, M., Utt, E., & Dutse, A. I. (1996). A randomized, controlled trial of oral and IV trovafloxacin vs. ceftriaxone in the treatment of epidemic meningococcal meningitis. Proceedings of the 37th Interscience Conference on Antimicrob Agents and Chemother (LB21), New Orleans, LA. Jones, R. N., Pfaller, M. A., & Doern, G. V. (1998). Comparative antimicrobial activity of trovafloxacin tested against 3049 Streptococcus pneumoniae isolates from the 1997–1998 respiratory infection season. Diagn Microbiol Infect Dis 32, 119 –126. Karlowsky, J. A., Zhanel, G. G., Davidson, R. J., Zieroth, S. R., & Hoban, D. J. (1993). In vitro postantibiotic effects following multiple exposures of cefotaxime, ciprofloxacin and gentamicin against Escherichia coli in pooled human cerebrospinal fluid and mueller-hinton broth. Antimicrob Agents Chemother 37, 1154 –1157. Kim, Y. S., Liu, Q., Chow, L. L., & Tauber, M. G. (1997). Trovafloxacin in treatment of rabbits with experimental meningitis caused by highlevel penicillin-resistant Streptococcus pneumoniae. Antimicrob Agents Chemother 41, 1186 –1189. Licata, L., Smith, C. E., Goldschmidt, R. M., Barrett, J. F., & Frosco, M. (1997). Comparison of the postantibiotic and postantibiotic sub-MIC effects of levofloxacin and ciprofloxacin on Staphylococcus aureus and Streptococcus pneumoniae. Antimicrob Agents Chemother 41, 950 –955. Lutsar, I., McCracken, J. M. Jr. & Friedland, I. R. (1998). Antibiotic pharmacodynamics in cerebrospinal fluid. Clinical Inf Dis 27, 1117–1129. MacKenzie, F. M. & Gould, I. M. (1993). The post-antibiotic effect. J Antimicrob Chemother 32, 519 –537. Nau, R., Schmidt, T., Kaye, F., Froula, J. L., & Tauber, M. G. (1995). Quinolone antibiotics in therapy of experimental pneumococcal meningitis in rabbits. Antimicrob Agents Chemother 39, 593–597. Pankuch, G. A., Jacobs, M. R., & Appelbaum, P. C. (1998). Postantibiotic effect of trovafloxacin against Gram-positive and -negative organisms. Antimicrob Agents Chemother 42, 1503–1505. Paris, M. M., Hickey, S. M., Trujillo, M., Shelton, S., & McCracken, G. H. Jr. (1995). Evaluation of CP-99,219, a new fluoroquinolone, for treatment of experimental penicillin- and cephalosporin-resistant pneumococcal meningitis. Antimicrob Agents Chemother 39, 1243–1246. Spangler, S. K., Lin, G., Jacobs, M. R., & Appelbaum, P. C. (1998). Postantibiotic effect and postantibiotic sub-MIC effect of levofloxacin compared to those of ofloxacin, ciprofloxacin, erythromycin, azithromycin, and clarithromycin against 20 Pneumococci. Antimicrob Agents Chemother 42, 1253–1255. Zhanel, G. G., Karlowsky, J. A., Davidson, R. J., & Hoban, D. J. (1992). Effect of pooled human cerebrospinal fluid on the postantibiotic effects of cefotaxime, ciprofloxacin and gentamicin against Escherichia coli. Antimicrob Agents Chemother 36, 1136 –1139. Zhanel, G. G., Karlowsky, J. A., Davidson, R. J., & Hoban, D. J. (1991). Influence of human urine on the postantibiotic effect (PAE) of ciprofloxacin against Escherichia coli. Antimicrob Agents Chemother 37, 218 –223.