Activity of Moxifloxacin, Imipenem, and Ertapenem against Escherichia coli, Enterobacter cloacae, Enterococcus faecalis, and Bacteroides fragilis in Monocultures and Mixed Cultures in an In Vitro Pharmacokinetic/Pharmacodynamic Model Simulating Concentrations in the Human Pancreas

Abstract

The activities of moxifloxacin, imipenem, and ertapenem against pathogens causing severe necrotizing pancreatitis were studied in an in vitro pharmacokinetics/pharmacodynamics (PK/PD) model. Escherichia coli, Enterobacter cloacae, Enterococcus faecalis, and Bacteroides fragilis were exposed in monocultures and mixed cultures to concentrations of the three agents comparable to those in the human pancreas. Moxifloxacin was more active than the two carbapenems in monocultures and mixed cultures, reducing the numbers of CFU more drastically and more rapidly.

TEXT

Infectious complications are the major determinant of morbidity and mortality in patients with severe acute pancreatitis (SAP), in particular in patients with necrotizing pancreatitis. Although polymicrobial infections caused by anaerobes and aerobes, including Escherichia coli, Enterobacter cloacae, Enterococcus faecalis, and Bacteroides fragilis are frequent in patients with SAP (2, 4, 13, 19), antibiotic prophylaxis is not recommended; however, in case of infectious complications, treatment with, e.g., a fluoroquinolone or a carbapenem is indicated (1, 4, 12, 16, 21, 23, 25).

Thus, we assessed the pharmacokinetics/pharmacodynamics (PK/PD) of moxifloxacin, imipenem, and ertapenem against SAP pathogens by simulating the regimens that either are recommended for treatment of SAP (carbapenems) or are commercially available (moxifloxacin).

(Part of the data in this publication were presented at the Interscience Conference on Antimicrobial Agents and Chemotherapy, Boston, MA, September 2010, and at the 21st conference of the European Society of Clinical Microbiology and Infectious Diseases, Milan, Italy, May 2011.)

Drug concentrations were deduced by log-linear regression from published data following infusions of 400 mg moxifloxacin (22) and 1,000 mg each of imipenem (3, 18) and ertapenem (24). The following maximal total drug concentrations (C_max), time to C_max (T_max), and elimination half-life (t_1/2) in human pancreatic tissue were deduced from these published data: for moxifloxacin, C_max = 4.6 mg/liter, T_max = 5 h, and t_1/2 = 8; for imipenem, C_max = 16 mg/liter, T_max = 0.25 h, and t_1/2 = 1.3 h; for ertapenem, C_max = 1.3 mg/liter, T_max = 0.5 h, and t_1/2 = 4 h. For calculation of free drug concentrations, the following was considered. First, binding of antibiotics to serum proteins was assessed (moxifloxacin, 40% [20]; imipenem, 20% [11]; ertapenem, 95% [17]). Second, all agents bind predominantly to albumin. Third, albumin concentrations in serum amount to 4.5%, compared to 1.3% in pancreatic juice (7, 8). This difference in albumin content was considered for the calculation of free drug concentrations in serum and pancreatic tissue, respectively.

E. coli ATCC 11775, E. cloacae ATCC 13047, E. faecalis ATCC 19433, and B. fragilis ATCC 25285 served as indicator strains. The Enterobacteriaceae were cultivated aerobically in cation-adjusted Mueller-Hinton broth, and B. fragilis in monoculture as well as B. fragilis and E. coli in mixed culture were grown anaerobically in brucella broth (Oxoid GmbH, Wesel, Germany). The pH was adjusted to 7.2, which corresponds to the physiologically relevant pH value in pancreatic juice (7, 8). Pre- and postexposure MICs (Table 1) were determined according to CLSI standards (9, 10) using the relevant American Type Culture Collection (ATCC) control strains. Moxifloxacin (batch BXO1X6E), imipenem (batch 0885170), and ertapenem (batch 0932340) were provided by the manufacturers. The open one-compartment model according to Grasso et al. (14) was used to simulate the free pancreas concentrations. Samples for drug monitoring and viable count determinations were withdrawn at 0, 0.5, 1, 2, 4, 6, 8, and 10 h and at the end of the study at 24 h. Drug concentrations in the system were monitored in the absence and presence of bacteria; actual drug concentrations were almost identical under both conditions and were on average within 1.2 to 2.4% of the desired profiles. The indicator strains (inoculum, 6.0 to 6.7 log₁₀ CFU/ml) grew equally well in monocultures, with growth rates (k) ranging from 0.19 (E. cloacae) to 0.25 h⁻¹ (B. fragilis), and in mixed cultures of E. coli plus E. cloacae (k = 0.2 + 0.2 h⁻¹), E. coli plus E. faecalis (k = 0.21 + 0.22 h⁻¹), and E. coli plus B. fragilis (k = 0.23 + 0.29 h⁻¹), respectively. The lower limit of detection of viable counts was 2 log₁₀ CFU/ml.

Table 1.

Drug susceptibilities of the four indicator strains

Druga	Pre-exposure/postexposure MIC (mg/liter)
	E. coli	E. cloacae	E. faecalis	B. fragilis
MXF	0.06/0.06	0.06/0.06	0.25/0.25	0.25/0.25
IMI	0.06/0.06	0.25/0.25	0.5/0.5	0.125/0.125
ERTA	0.015/0.03	0.25/0.25	8.0/8.0	0.25/0.25

^aMXF, moxifloxacin; IMI, imipenem; ERTA, ertapenem.

Viable counts of E. coli and E. cloacae growing in monocultures were eliminated by moxifloxacin within 4 and 10 h, respectively; moxifloxacin reduced the inocula of E. faecalis and B. fragilis by 1.9 and 2.7 log₁₀ CFU/ml within 6 h; regrowth was not observed. Imipenem reduced viable counts of E. coli and E. cloacae growing in monocultures by 4 log₁₀ units within 2 h and those of E. faecalis and B. fragilis by 3 log₁₀ units each within 8 h. The latter two species regrew by 5 log₁₀ units. Ertapenem reduced CFU of E. coli, E. cloacae, and B. fragilis in monocultures by 5.8 log₁₀ each in 8 h, whereas viable counts of E. faecalis were not affected. Regrowth was noted for E. coli by 4.5, for E. cloacae by 5.3, and for B. fragilis by 3.8 log₁₀ units.

The activities of the three study drugs against mixed cultures are summarized in Table 2. Moxifloxacin was more active against E. cloacae, E. faecalis, and B. fragilis growing in mixed cultures than against monocultures, reducing the CFU more extensively and more rapidly. However, all species except B. fragilis regrew in mixed cultures, whereas they did not regrow in monocultures. The two carbapenems were less active in mixed cultures than in monocultures; first, viable counts were reduced more slowly, and second, regrowth was recorded in almost every case, except of B. fragilis. MICs of pre- and postexposure isolates were identical despite regrowth of bacteria having been exposed to the three study drugs (Table 1). Moxifloxacin tended to reduce viable counts of all species growing in mixed cultures more effectively and more rapidly than imipenem or ertapenem. The different activities of the agents tested in either monocultures or mixed cultures cannot be due to different growth conditions or competition for nutrient supply, as the growth rates of the indicator strains were almost identical under both conditions.

Table 2.

Antibacterial activities of moxifloxacin (MXF), imipenem (IMI), and ertapenem (ERTA) against mixed bacterial cultures in an in vitro model simulating free drug concentrations in pancreatic tissue

Drug	E. coli + E. cloacae		E. coli + E. faecalis		E. coli + B. fragilis
	log10 ΔCFU (h)a	Regrowth (fold) after 24 h	log10 ΔCFU (h)a	Regrowth (fold) after 24 h	log10 ΔCFU (h)a	Regrowth (fold) after 24 h
MXF	−>6 (4)/−>6 (4)	+2/+6	−>6 (2)/−4 (8)	+2/+2.3	−>6 (2)/−3 (6)	+2/0
IMI	−4 (6)/−3 (6)	+2/+4	−5.5 (10)/−2 (8)	+3/+3	−5 (6)/−2.5 (8)	+2/0
ERTA	−5.3 (4)/−5.3 (4)	+3.1/+6.6	−4.5 (4)/+1.6 (4)	+3.4/+2.5	−5.4 (4)/−2.1 (6)	+2/0

^aMaximal reduction (negative values) or growth (positive values) of viable counts in the indicated time.

Data generated in the experiments described above simulating free concentrations in the human pancreas demonstrate that moxifloxacin and the two carbapenems were active against the major bacterial species causing infectious complications in patients with SAP. There is a 40 to 60% risk of pancreatic infections in patients with necrotizing pancreatitis, so antibacterial treatment is justified. Actually, carbapenems or ciprofloxacin are recommended (1, 5, 12, 21, 25). However, in contrast to ciprofloxacin, moxifloxacin is active against anaerobes; furthermore, its pharmacokinetic profile is more favorable than that of ciprofloxacin, as the concentrations of moxifloxacin in pancreatic tissue exceed the corresponding serum concentrations 3-fold (22), compared to a 1:1 ratio for ciprofloxacin (6, 15). In comparison to ertapenem, which exerts no activity against enterococci, moxifloxacin is variably active against E. faecalis. Furthermore, moxifloxacin was found to affect viable counts of the pathogens in mixed cultures more effectively and more rapidly than the two carbapenems. Thus, moxifloxacin may be superior to imipenem and ertapenem and could be a therapeutic alternative for an antibacterial treatment in patients with severe necrotizing pancreatitis.