Activity of Antimicrobial Combinations against KPC-2-Producing Klebsiella pneumoniae in a Rat Model and Time-Kill Assay

Abstract

This study evaluated the efficacy of tigecycline (TIG), polymyxin B (PMB), and meropenem (MER) in 80 rats challenged with Klebsiella pneumoniae carbapenemase (KPC)-producing K. pneumoniae infection. A time-kill assay was performed with the same strain. Triple therapy and PMB+TIG were synergistic, promoted 100% survival, and produced negative peritoneal cultures, while MER+TIG showed lower survival and higher culture positivity than other regimens (P = 0.018) and was antagonistic. In vivo and in vitro studies showed that combined regimens, except MER+TIG, were more effective than monotherapies for this KPC-producing strain.

TEXT

Infections due to Klebsiella pneumoniae carbapenemase (KPC)-producing Enterobacteriaceae are associated with therapeutic failure and increased mortality (1–5). It has been suggested that antibiotic combination might be a better alternative compared to monotherapy for treatment of KPC-producing isolates (4–9); however, further investigation on this therapeutic strategy is required (10, 11).

In the present study, we evaluated the efficacy of regimes for a KPC-producing K. pneumoniae strain in an experimental model of systemic infection and in a time-kill assay (TKA).

A KPC-producing K. pneumoniae strain, coded RM-1209, was isolated from a patient's blood and identified by Vitek 2 (bioMérieux, Craponne, France) in 2012. Antibiotic MICs were determined by agar dilution and interpreted according to CLSI and EUCAST for tigecycline (TIG) (12, 13). The strain presented MICs of >32 mg/liter, 1 mg/liter, and 0.5 mg/liter for meropenem (MER), tigecycline, and polymyxin B (PMB), respectively. Detection of the bla_KPC gene was performed with BigDye v1.1 Sequencing kits (Applied Biosystems, Foster City, CA, USA), and KPC-2 was confirmed at databases queried by NCBI BLAST (6).

The study was approved by the ethical committee, according to the Protocol for the Protection and Welfare of Animals (European Union). The animal model has been previously described (13). In brief, male and female immunocompetent Wistar rats weighing between 190 and 300 g were randomized into each treatment group. Absence of immunosuppression demanded administration of highly concentrated inoculums with 1.5 × 10¹⁰ CFU/ml. This solution, when diluted 1:20 presents an absorbance of 0.546 at 625 nm wavelength on spectrophotometry (corresponding to tube 50 of nephelometric scale). Seven groups of 10 rats were treated with the following regimens: MER, TIG, PMB, MER+PMB, MER+TIG, PMB+TIG, or PMB+MER+TIG. Ten rats were untreated (control group). The sample size (80 rats) was calculated using the formula n = z2PQ/d2 based on the expected proportion of deaths (50%) and 8% standard error with an 80% statistical power.

Animals were injected with a 0.7-ml intraperitoneal aliquot of 1.5 × 10¹⁰ CFU/ml KPC-producing K. pneumoniae inoculum in the log growth state (14). After infection, rats received antimicrobials intraperitoneally, TIG (Tygacil) at 7 mg/kg of body weight every 12 h (15, 16), PMB (polymyxin B) at 2 mg/kg every 12 h (17, 18), or MER (Meronem) at 50 mg/kg every 8 h (19), or they remained untreated. No pharmacokinetic (PK) data were obtained. Length of survival was observed for 24 h, and euthanasia was performed on rats surviving at 24 h. Blood samples were collected by aseptic cardiac puncture, and peritoneal fluid samples were obtained through direct observation after incision by an aseptic technique. Blood samples were incubated in broth, and peritoneal fluid was incubated on MacConkey agar plates for quantitative cultures with a 1-μl loop.

In vitro TKA was performed by inoculating 5 × 10⁶ CFU/ml of the clinical strain into 10 ml of fresh cation-adjusted Mueller-Hinton broth (Oxoid, Basingstoke, United Kingdom) and incubating it at 35°C. MER at 4 mg/liter, TIG at 1 mg/liter, and PMB at 0.25 mg/liter were tested alone and in the same combinations performed in vivo. Aliquots were removed at 1, 6, 12, and 24 h after inoculation. Samples were serially diluted (10⁻¹ to 10⁻⁸) and plated in duplicate on blood agar plates for colony count. Antimicrobial carryover was controlled by streaking the transferred aliquot over the agar plate and observing possible inhibition of growth at the site of the initial streak. Potential in vitro MER hydrolysis was not assessed.

Time-kill curves were constructed by plotting mean colony counts versus time. The results were interpreted after 24 h of incubation.

Survival curves were constructed, and a Gehan-Breslow-Wilcoxon test was performed. Mann-Whitney U and Kruskal-Wallis tests evaluated differences in blood, peritoneal cultures, and mortality between groups. Alpha adjustment was performed with Dunn's multiple comparison tests. SPSS 16 (IBM, Armonk, NY, USA) software was used for statistical analysis, and Prism (GraphPad, La Jolla, CA) was used for graph construction. A P value of <0.05 was considered statistically significant.

In Kaplan-Meier survival curves, untreated rats presented with 80% mortality, a proportion similar to that of rats treated with TIG and MER monotherapies, whose mortality was 60.0% in the two groups (P = 0.061 and P = 0.114). All animals treated with combinations including PMB survived (PMB+TIG, PMB+TIG+MER, PMB+MER), whereas a 70.0% survival rate was observed with MER+TIG (P = 0.018). The MER+TIG combination did not produce significantly different survival from PMB monotherapy (P = 0.901) (Fig. 1A and 2).

FIG 1.

Animal model of sepsis of KPC-2-producing K. pneumoniae and effects of different antibiotic combinations. (A) Mortality (*, P < 0.05 compared with PMB, MER+TIG, MER, TIG, and control; **, P < 0.05 compared with control); (B) peritoneal culture (*, P < 0.05 compared with control, MER, TIG, and MER+TIG); (C) blood culture (***, P < 0.05 compared with control, MER, TIG, and MER+TIG).

FIG 2.

Survival curves of a rat model of sepsis infected with KPC-2-producing Klebsiella pneumoniae and response to treatment by antibiotic combinations or monotherapies. *, P value of <0.05 compared with other groups; **, P value of <0.05 compared with control.

PMB monotherapy and combinations including PMB significantly sterilized more peritoneal cultures than MER+TIG (P = 0.001) (Fig. 1B). The PMB+TIG combination determined the lowest positivity of the blood cultures, statistically significantly different from the control, MER, TIG, and MER+TIG groups (P < 0.001) (Fig. 1C).

The PMB+TIG combination and triple therapy were synergistic, whereas MER+TIG showed an antagonistic effect in the TKA (Fig. 3).

FIG 3.

Time-kill assay of KPC-2-producing K. pneumoniae using antibiotic combinations: (A) PMB+TIG; (B) PMB+MER; (C) PMB+MER+TIG; (D) MER+TIG. Bactericidal activity was defined as a ≥3-log₁₀ reduction in the total number of CFU per milliliter from the original inoculum at 24 h; bacteriostatic activity was defined as a <3-log₁₀ reduction in the total number of CFU per milliliter from the original inoculum at 24 h; synergism was defined as a difference of ≥2 log₁₀ in the reduction of the number of CFU per milliliter between the combination and the most active agent at 24 h; antagonism was defined as a ≥2-log₁₀ increase of the number of CFU per milliliter between the combination and the most active agent at 24 h.

Evaluating survival and overall culture sterilization, better performance occurred with PMB+TIG, followed by triple therapy, PMB+MER, and PMB. Monotherapies did not present the same efficacy observed in the better performing combinations, PMB+TIG and PMB+TIG+MER, corroborating previous studies, which observed lowered mortality with antimicrobial combinations (4–9).

It has been previously described, in retrospective data, that triple therapy might lower mortality in patients infected with KPC-producing K. pneumoniae; however, the MER MIC was not mentioned (4). In the present study, triple therapy was not superior to PMB+TIG; however, this finding should be confirmed using a lower MER MIC strain.

Moreover, the combination of PMB+MER demonstrated an impact on survival and peritoneal cultures but was not superior to PMB monotherapy for treating bacteremia. Despite the lower mortality observed with the PMB+MER combination, no synergism was observed with PMB+MER in the TKA.

Of interest, we found an antagonistic effect of TIG+MER, which caused worse outcomes than PMB monotherapy, as previously suggested by our group (20). Pournaras et al. also showed that MER+TIG is not synergic in vitro (21). Previous studies of nonlethal and nonsystemic infections were discordant about the antagonistic effects of TIG+MER in vivo (20, 22) and in vitro (23). Further investigation using this combination is required, but we believe that TIG+MER should be used with caution for treatment of infections due to KPC-producing isolates. On the contrary, the activity of TIG+PMB showed favorable results in our study, corroborating previous in vitro (21) and in vivo (24) findings. Tigecycline monotherapy did not perform better than MER alone, potentially because of a bacteriostatic effect and inadequate PK for the treatment of bloodstream infections, where TIG should be avoided (25).

A strength of this study is that the in vivo and in vitro experiments showed similar findings of antimicrobial activity and sample size with an adequate statistical power to detect relevant differences in clinical and microbiological outcomes. A limitation of the present study was the evaluation of only one strain, as differences among strains can affect study results despite similar MICs (26). We also did not perform a PK evaluation of the antibiotics in rats, which might contribute to the interpretation of our findings (15, 17–19), but previous in vitro studies have demonstrated that combinations of TIG and MER and TIG and PMB present favorable PK results (27, 28).

In summary, a combined regimen including PMB resulted in improved outcomes in this experimental study. Triple therapy was not superior to dual combinations with PMB. In vitro antagonism of TIG+MER was correlated with similar outcomes and resulted in worse microbiological findings compared to MER, TIG, and PMB monotherapies.