TABLE 5. Monte Carlo simulations to predict probability of microbiologic treatment success for suspected or documented anthrax meningitis for antimicrobial drugs at specific intravenous dosages.

Antimicrobial drug	Drug class	Mechanism of action	IV antimicrobial drug dosage for 70-kg adult	MIC90 (µg/mL)	PK/PD breakpoint (µg/mL)*,†
High probability§
Ciprofloxacin	Fluoroquinolone	C	400 mg every 8 hours	0.03	0.06
Levofloxacin	Fluoroquinolone	C	500 mg every 12 hours	0.25	0.25¶
Meropenem	Carbapenem	C	2 g every 8 hours in a 30-minute infusion	0.12	0.25**
Imipenem/cilastatin	Carbapenem/dehydropeptidase inhibitor	C	1 g every 6 hours in a 1-hour infusion	0.12	1**
Penicillin G	Penicillin	C	4 million units every 4 hours	0.03††	0.25**
Ampicillin	Penicillin	C	2 g every 4 hours	0.03††	0.5**
Ampicillin/sulbactam	Penicillin/ß-lactamase inhibitor	C	3 g every 6 hours	0.03††	0.25**
Doxycycline	Tetracycline	S	100 mg every 12 hours	0.03	0.25¶
Minocycline	Tetracycline	S	100 mg every 12 hours	≤0.06	0.125¶
Acceptable probability§§
Piperacillin/tazobactam	Penicillin/ß-lactamase inhibitor	C	3.375 g every 4 hours	1††	1
Levofloxacin	Fluoroquinolone	C	750 mg every 24 hours	0.25	0.125¶
Low probability¶¶
Vancomycin	Glycopeptide	C	15 mg/kg every 12 hours	2	0.06***
Amikacin	Aminoglycoside	C	25 mg/kg every 24 hours	1	0.125**
Clindamycin†††	Lincosamide	S	900 mg every 8 hours	0.25	0.0313§§§
Linezolid†††	Oxazolidinone	S	600 mg every 12 hours	2	0.25***

Source: CDC unpublished study.

Abbreviations: C = bactericidal; CSF = cerebrospinal fluid; fAUC/MIC = area under the unbound (i.e. free) concentration time curve divided by the minimal inhibitory concentration; fT>MIC = duration of the unbound (i.e. free) concentration above the minimum inhibitory concentration (expressed as a fraction of the dosing interval); IV = intravenous; MIC = minimum inhibitory concentration; MIC₉₀ = minimum inhibitory concentration required to inhibit the growth of 90% of organisms; PD = pharmacodynamic; PK = pharmacokinetic; S = bacteriostatic.

* Highest MIC (µg/mL) with a >95% probability of successful CSF treatment.

^† PK/PD targets used were fAUC/MIC ≥18 for doxycycline and minocycline, fAUC/MIC ≥75 for quinolones, fT>MIC ≥40% for carbapenems, fT>MIC ≥50% for penicillins, fAUC/MIC ≥90 for amikacin, fAUC/MIC ≥27 for clindamycin, AUC/MIC ≥110 for linezolid, and AUC/MIC ≥400 for vancomycin.

^§ High probability is defined as a ≥95% probability of treatment success.

^¶ PK/PD target achievement and the highest treatable MICs were similar (i.e., within twofold) in CSF and plasma for ciprofloxacin, levofloxacin, doxycycline, and minocycline.

** For the studied ß-lactams (i.e., meropenem, imipenem, penicillin G, ampicillin, and piperacillin) and for amikacin, the highest treatable MICs based on the plasma targets were twofold to thirty-two-fold higher compared with those in CSF.

^†† Previous susceptibility testing to infecting strain is recommended. Bacillus anthracis naturally carries penicillin-resistance genes for a ß-lactamase that is expressed in up to 10% of naturally occurring strains, creating resistance to penicillin G, ampicillin, and piperacillin.

^§§ Acceptable probability is defined as a ≥90% to <95% probability of treatment success; these antimicrobial drugs should be used only if those that provide high probability of success are not available.

^¶¶ Low probability is defined as a ˂90% probability of treatment success.

*** For linezolid and vancomycin, the highest treatable MICs based on plasma drug exposures were fourfold to eightfold below the MIC₉₀.

^††† Clindamycin and linezolid did not achieve sufficient drug exposures in CSF and plasma as monotherapy to achieve clinically relevant bacterial killing. However, these protein synthesis inhibitors might be useful in combination therapy to inhibit toxin production.

^§§§ For clindamycin, the highest treatable MIC based on plasma drug exposures was 0.25 µg/mL (MIC₉₀).