Use of the Hollow-Fiber Infection Model to Measure the Effect of Combination Therapy of Septic Shock Exposures of Meropenem and Ciprofloxacin against Intermediate and Resistant Pseudomonas aeruginosa Clinical Isolates

ABSTRACT

Meropenem-ciprofloxacin combination therapy was compared to the respective monotherapy in a Hollow-Fiber Infection Model against two Pseudomonas aeruginosa isolates. Following initial kill of ∼ 5-logs by each monotherapy, rapid regrowth occurred within 24 h, reaching 10⁸ – 10¹⁰ CFU/mL at 120 h. In contrast, combination therapy achieved > 5-log kill within 6 h and suppressed bacterial regrowth throughout. The results suggest that meropenem-ciprofloxacin combination may provide significantly enhanced bacterial killing and resistance suppression against P. aeruginosa.

INTRODUCTION

Critically ill patients with septic shock are at a higher risk of infection with resistant organisms (1). Pseudomonas aeruginosa is a typical multidrug-resistant pathogen, often associated with high mortality (2). It has intrinsic resistance to many antibiotics and the ability to mutate and acquire extensive resistance (2, 3). Many clinicians elect to treat P. aeruginosa with combination therapy to minimize risk of treatment failure (2, 4). However, the clinical evidence supporting combination therapy is insufficient and remains extensively debated (4, 5). Therefore, targeted studies are warranted to explore scenarios where combination therapy with existing antibiotics may have a benefit over monotherapy. Given the challenge of addressing this in clinical studies, we applied the Hollow-Fiber Infection Model (HFIM) to investigate whether a meropenem-ciprofloxacin combination therapy is superior to monotherapy in terms of bacterial killing and suppression of resistance emergence against clinical isolates of P. aeruginosa, using simulated antibiotic pharmacokinetics from septic shock patients.

Two isolates of P. aeruginosa sourced from critically ill patients (isolates #27 and #40) were examined. MICs for meropenem and ciprofloxacin were determined in quadruplicates using the broth microdilution method (6). Spontaneous mutant frequency studies were performed with initial inoculum of 1 × 10² CFU/mL and drug containing agar plates at concentrations 32 mg/L for meropenem, and 8 mg/L for ciprofloxacin.

Static-concentration time-kill studies (initial inoculum of 1 × 10⁵ CFU/mL) were performed for a preliminary characterization of the effect of meropenem and ciprofloxacin alone and in combination at concentrations of 0.5 × MIC, 2 × MIC and 4 × MIC. Synergy was defined as > 2 log bacterial killing by the combination compared to that achieved by each drug alone after 24 h of exposure (7).

A HFIM was set up as previously described (8), using FiberCell Systems C3008 cartridge. The initial inoculum was 1 × 10⁷ CFU/mL. Three treatment regimens were evaluated: 2 g meropenem every 8 h, 600 mg ciprofloxacin every 8 h and combination of the two. The dosing regimens were based on recent studies on optimal antibiotic dosing in patients with septic shock (9, 10). The simulated pharmacokinetic concentration-time profiles were based on half-life estimations (1.6 h for meropenem and 3.3 h for ciprofloxacin) from previously published data sets of critically ill patients (9, 10) using non-compartmental analysis in R with Pmetrics. Samples for pharmacokinetic and pharmacodynamic analysis were collected over 5 days. The antibiotic agar concentrations for quantitative culture were 8 mg/L (8×MIC) for ciprofloxacin, and 32 mg/L (8×MIC for #27, 2× MIC for #40) for meropenem. For any growth on drug-containing agar plates, MICs were determined on day 5 of the experiment to confirm resistance.

Concentrations of ciprofloxacin and meropenem were measured by a Shimadzu Nexera2 UHPLC-PDA system. The methods were validated in accordance with the FDA criteria for bioanalysis. For ciprofloxacin, precision and accuracy were within 6.2% and 0.4% respectively. For meropenem, precision and accuracy were within 3.4%.

The meropenem MICs for isolates #27 and #40 were 4 mg/L and 16 mg/L, respectively. The MIC of ciprofloxacin for both isolates was 1 mg/L. The spontaneous mutant frequencies of isolate #27 were <9.00 × 10⁻¹¹ for meropenem and <9.00 × 10⁻¹¹ for ciprofloxacin. The mutant frequencies for isolate #40 were 6.00 × 10⁻⁶ for meropenem and <6.00 × 10⁻⁹ for ciprofloxacin.

Fig. 1 presents the total viable counts from static time-kill studies. At 0.5 x MIC, all monotherapies failed to supress bacterial growth, however a potential synergy was observed for the combination therapy. These findings are supported by previous time-kill studies reporting synergism between meropenem and ciprofloxacin against P. aeruginosa isolates (11–14).

FIG 1.

Time-kill curves for P. aeruginosa isolates at 0.5 ×, 2 × and 4 × MIC for isolate #27 and #40. MIC²⁷_MEM = 4 mg/L, MIC²⁷_CIP = 1 mg/L. MIC⁴⁰_MEM = 16 mg/L, MIC⁴⁰_CIP = 1 mg/L. MEM = Meropenem. CIP = Ciprofloxacin.

Fig. 2 shows results from the HFIM. Both the meropenem and ciprofloxacin monotherapy failed to sustain bacterial killing. In contrast, the combination achieved rapid bacterial killing within 6 h (>10⁵ decrease) with no regrowth. Similar results were observed in a recent HFIM-study evaluating meropenem-ciprofloxacin combination against P. aeruginosa in respiratory infections (12). However, to our knowledge, our study is the first to evaluate the meropenem-ciprofloxacin combination in the context of altered pharmacokinetics in septic shock-patients and against already resistant P. aeruginosa isolates. The r² for observed-versus-simulated meropenem and ciprofloxacin plots were 0.9897 and 0.8339, respectively. Mean observed steady state ciprofloxacin AUC was 33 ± 2.9 mg*h/L: average AUC/MIC ratio of 33 for both isolate #27 and #40. Mean ± SD observed meropenem C_min at steady state was 3.3 ± 0.9 mg/L: average C_min/MIC ratio of 0.8 and 0.2 for isolate #27 and #40, respectively.

FIG 2.

The effect of meropenem-ciprofloxacin combination therapy versus monotherapy with each antibiotic on the total bacterial population count of P. aeruginosa clinical isolates in a HFIM.

Fig. 3A to C shows effect of the combination versus the monotherapies on suppression of resistant sub-populations. Both monotherapies accelerated growth of non-susceptible subpopulations while the combination therapy inhibited all resistance development. Ciprofloxacin-resistant subpopulations (Fig. 3B) emerged only during ciprofloxacin monotherapy. The MIC for ciprofloxacin increased from 1 mg/L to 16 mg/L in isolate #27 and to 32 mg/L in #40 (Table 1). A subpopulation resistant to the combination of meropenem and ciprofloxacin was observed only in isolate #40 following ciprofloxacin monotherapy (Fig. 3C). These likely represent the fraction that can be spontaneously resistant to meropenem among the population that is totally resistant to ciprofloxacin. Furthermore, for isolate #40, colonies grown on meropenem plates after ciprofloxacin treatment had an elevated MIC of 64 mg/L for meropenem, compared to 16 mg/L originally. These could be caused by cross-resistance; fluoroquinolones have been shown to increase the carbapenem resistance in P. aeruginosa, possibly due to inhibition of the production of OprD (outer membrane porin channel) and/or induction of overexpression of MexAB-OprM (efflux pump) (15–18).

FIG 3.

A to C: Bacterial counts of resistant subpopulations of P. aeruginosa isolate #27 and #40 from the HFIM, grown on agar plates containing 32 mg/L meropenem (A), 8 mg/L ciprofloxacin (B) or the combination of the two (C). Baseline MICs against meropenem were 4 mg/L and 16 mg/L respectively and against ciprofloxacin 1 mg/L for both isolates. Limit of quantification = 100 CFU/mL. MEM = meropenem, CIP = ciprofloxacin.

TABLE 1.

Post HFIM-treatment MICs in comparison with baseline MIC for P. aeruginosa isolates. The MIC was performed on colonies picked from drug-plates^a

Isolate no.	Regimen	Drug plate	Baseline MIC (mg/L)		Post-treatment MIC (mg/L)
			MEM	CIP	MEM	CIP
# 27	2 g MEM + 600 mg CIP	Drug-free	4	1	4	1
	2 g MEM	MEM	4	1	32	0.5
	600 mg CIP	CIP	4	1	1	16
	Control	MEM	4	1	8	1
		CIP	4	1	2	16
		Drug-free	4	1	2	2
# 40	2 g MEM	MEM	16	1	16	1
		Drug-free	16	1	8	2
	600 mg CIP	MEM	16	1	64	32
		CIP	16	1	8	32
		MEM+CIP	16	1	16	32
	Control	MEM	16	1	16	8
		Drug-free	16	1	8	0.25

^aDrug plate concentrations were 32 mg/L for meropenem and 8 mg/L for ciprofloxacin. MEM = meropenem, CIP = ciprofloxacin.

The clinical use of combination therapy is still under debate (19). Several meta-analyses and single studies have found no mortality difference between monotherapy and combination therapy in patients severely ill with bacteremia (20–23). There is a discrepancy between in vitro studies, often supporting synergy in combination therapy, and clinical studies, failing to do the same. An explanation could be that the benefit from antibiotic therapy in sepsis depends on early administration as well as reaching therapeutic drug concentrations (1). Further, there are other aspects of sepsis treatment that influence outcome, including fluid resuscitation, vasopressors, mechanical ventilation and dialysis (1). Another factor is that a synergistic acitivity is most pronounced in intermediate to resistant isolates. In susceptible isolates, an appropriately dosed monotherapy could be effective enough and a synergistic effect would be harder to detect.

There are some limitations to this study. Firstly, all HFIM experiments were performed in singlicate using only two isolates, similar to other HIFM studies in the literature. Another limitation is the lack of immune cells in the HFIM, which neglects the effect of the immune system in combating a bacterial infection when simulating bacteremia. However, this could also be considered an advantage to allow investigation of the effects of the antibiotics alone, without interference by immune cells.

In conclusion, this study indicates a synergistic benefit of combination therapy with meropenem and ciprofloxacin against intermediate-to-resistant P. aeruginosa in a HFIM simulating critically ill patients with altered pharmacokinetics. Optimized combination therapy resulted in rapid and sustained bacterial killing and resistance suppression, while both monotherapies led to bacterial regrowth and enhanced resistance. Our findings support further clinical investigation regarding combination therapy against P. aeruginosa in septic shock-patients with altered pharmacokinetics.