Successful Ertapenem-Doripenem Combination Treatment of Bacteremic Ventilator-Associated Pneumonia Due to Colistin-Resistant KPC-Producing Klebsiella pneumoniae

LETTER

Severe infections caused by carbapenemase-producing Klebsiella pneumoniae are becoming a significant problem worldwide and are associated with high morbidity and mortality rates (1–3). Recently, treatments based on therapies with combinations of colistin, tigecycline, meropenem, fosfomycin, and/or aminoglycoside have been suggested (4–6). However, the emergence of strains resistant to almost all of the antibiotics listed above has further complicated the possibility of treating these infections (7–9).

A 65-year-old male was admitted to the Neurosurgical Intensive Care Unit of Azienda Policlinico Umberto I in Rome for treatment of cerebral hemorrhage and hydrocephalus that occurred 3 days after a surgical excision of a subependymoma. The patient was intubated and mechanically ventilated.

On day 25 of admission, the clinical course was complicated by the development of a bacteremia with severe sepsis due to Enterobacter aerogenes that was successfully treated with meropenem (1 g every 8 h [q8h]). On day 43, the patient once again developed a high-grade fever with multiple pulmonary bilateral infiltrates. Blood cultures and semiquantitative cultures of endotracheal aspirates yielded colistin/tigecycline-resistant, multidrug-resistant (MDR) K. pneumoniae isolates according to the bioMérieux Vitek-2 automated system.

Despite the antibiotic treatment with colistin (loading dose of 9 MU followed by 4.5 MU q12h), meropenem (2 g q8h), and rifampin (300 mg q8h) for 6 days and afterward with colistin plus fosfomycin (3 g q8h) for 5 days, high fever and bacteremia persisted, with an increase of procalcitonin levels (to 140 ng/ml), development of multiple-organ-dysfunction syndrome (total bilirubin, 14.9 mg/dl; creatinine, 3 mg/dl; platelets, 28,000/μl, PaO²/FiO² < 300) and the need for inotropic drug support. A subsequent laboratory study, in which both broth microdilution (BMD) analysis and an Etest were performed, confirmed that these isolates (4 isolates collected since day 47 of hospitalization from 3 blood cultures and 1 endotracheal aspirate) were resistant to ertapenem, meropenem, imipenem, doripenem, amikacin, colistin, and fosfomycin but evidenced that they were susceptible to tigecycline with both methods, confirming the overestimation of the MIC for this drug if performed with the Vitek2 system (Table 1) (6, 10). The same clinical isolates, genotyped by pulsed-field gel electrophoresis (PFGE) and multilocus sequence typing (MLST), belonged to the same clone and were sequence type (ST) 512. PCR detection showed that all isolates harbored the bla_KPCsub> gene (11).

Table 1.

Antibiotic susceptibility comparison by Vitek 2, broth microdilution, and Etest methods against 4 K. pneumoniae isolates^a

Isolate no. (day of hospitalization)	Specimen	Antibiotic	MIC (mg/liter)b
			Vitek 2	Etest	BMD
1 (48)	Endotracheal aspirate	IPM	≥16	>32	32
		MEM	≥16	>32	64
		ERTA	≥8	>32	256
		DOR	n.t	n.t	64
		AK	≥64	48	32
		COL	≥16	2	32
		FOSFO	≥128	32	64
		TGC	≥8	0.38	0.5
2 (48)	Blood	IPM	≥16	>32	32
		MEM	≥16	>32	64
		ERTA	≥8	>32	512
		DOR	n.t	n.t	32
		AK	≥64	48	32
		COL	≥16	4	16
		FOSFO	≥128	64	128
		TGC	≥8	1	0.5
3 (53)	Blood	IPM	≥16	>32	32
		MEM	≥16	>32	64
		ERTA	≥8	>32	64
		DOR	n.t	n.t	64
		AK	≥64	48	32
		COL	≥16	6	16
		FOSFO	≥128	32	64
		TGC	≥8	0.38	0.5
4 (59)	Blood	IPM	≥16	>32	32
		MEM	≥16	>32	64
		ERTA	≥8	>32	512
		DOR	n.t	n.t	64
		AK	≥64	64	32
		COL	≥16	4	16
		FOSFO	≥128	32	64
		TGC	≥8	0.75	0.5

^aAbbreviations: BMD, broth microdilution; IPM, imipenem; MEM, meropenem; ERTA, ertapenem; DOR, doripenem; AK, amikacin; COL, colistin; FOSFO, fosfomycin; TGC, tigecycline; n.t, not tested.

^bData represent 2013 EUCAST breakpoints.

On day 52, therapy was modified to ertapenem (500 mg q24h) plus doripenem (250 mg q8h by a 4-h-extended infusion) according to the initial values of creatinine clearance. Tigecycline therapy was not considered because, at the moment when we choose therapy, the only available data were those obtained with the Vitek-2 automated system, which showed tigecycline-resistant isolates. The patient remained febrile for a further 4 days, whereas the bacteremia cleared after 8 days. Subsequently, inotropic support was discontinued, and the PaO2/FiO2 ratio and platelet count values returned gradually to normal within 2 weeks, creatinine and total bilirubin values to normal in 3 weeks, and the procaciltonin value to normal in 4 weeks. With the improvement of renal function, ertapenem was administered at a dose of 1,000 mg q24h and doripenem at a dose of 500 mg q8h (day 67) up to 1,000 g q8h (day 73). Eventually, the patient completed a 4-week dual-carbapenem treatment course. No relapse was observed after 1 month of follow-up after discontinuation of the antibiotics.

The activity of the carbapenem combination was also confirmed in vitro with the striking synergy that was observed in the studies of the killing curves. In fact, in these experiments, the combination of ertapenem plus doripenem at 1 × MIC was strongly synergic after 4 h, achieving 99.9% killing, as was ertapenem plus meropenem, maintaining this behavior until 24 h. The value for ertapenem alone showed an increase of 1 log after 24 h, while those for doripenem and meropenem alone showed an increase of 3 log (Fig. 1).

Fig 1.

Time-kill curves for K. pneumoniae with ertapenem (Ert) at 1 × MIC (512 mg/liter), doripenem (Dor) at 1 × MIC (64 mg/liter), meropenem (Mer) at 1 × MIC (64 mg/liter), and the combinations of ertapenem plus doripenem at 1 × MIC and ertapenem plus meropenem at 1 × MIC.

Our case report on the result obtained in vitro and in vivo with a KPC-3-producing K. pneumoniae seems to corroborate experiments performed by Bulik et al. (12), who recently postulated that the enhanced efficacy of this dual-carbapenem therapy against KPC-2-producing K. pneumoniae may be related to the KPC enzyme's preferential affinity for ertapenem.