Gentamicin-Resistant Enterococcus faecalis Sequence Type 6 with Reduced Penicillin Susceptibility: Diagnostic and Therapeutic Implications

In 2007, 20 Enterococcus faecalis isolates from bloodstream infections diagnosed at seven hospitals in the North Denmark region were recorded by the regional clinical microbiology laboratory as being penicillin resistant (MICs ≥ 16 μg/ml) and ampicillin susceptible (MICs ≤ 4 μg/ml) using Etest (AB bioMérieux, Solna, Sweden). Seventeen and two isolates were additionally resistant to gentamicin (MICs ≥ 1,024 μg/ml) and imipenem (MIC = 32 μg/ml), respectively (Table 1).

TABLE 1.

Origin, MICs determined by Etest, and genotypes based on partial (gdh gene) or full multilocus sequence typing (MLST) and pulsed-field gel electrophoresis (PFGE) of 33 E. faecalis isolates from bloodstream infections in the North Denmark region in 2007

Strain ID	Site of infectiona	Originb	Hospitalc	Etest MIC (μg/ml) ford:					MLST		PFGE profile
				AMP	PEN	IPM	GEN	VAN	gdh	ST
Isolates with reduced susceptibility to penicillin
B69486	AVP	C	4	1.5	>32	6	>1024	3	12	6	A1
B135456	Urinary tract	N	6	4	>32	8	96	2	12	6	A1
B33621	Urinary tract	N	7	1	16	8	1024	3	12	6	A1
B37814	Undetermined	N	8	1.5	>32	8	>1024	3	12	6	A1
B63792	Urinary tract	NH	8	1.5	>32	6	>1024	4	12	6	A1
B71580	HBP tract	HC	8	1.5	>32	8	>1024	4	12	6	A1
B98380	Undetermined	C	1	3	16	4	>1024	2	12	6	A2
B15725	Undetermined	N	8	1.5	32	32	>1024	3	12	6	A2
B72715	Urinary tract	N	4	2	>32	6	1024	3	12	6	A3
B37093	Undetermined	N	8	1	>32	4	>1024	3	12	6	A3
B33672	Perirenal abscess	N	3	1	32	4	1024	3	12	6	A4
B54027	IV catheter	N	8	3	>32	6	>1024	3	12	6	A4
B147794	Diabetic gangrene	N	8	0.75	>32	4	>1024	2	12	6	A4
B110274	Undetermined	N	8	3	>32	4	>1024	3	12	6	A5
B85040	Urinary tract	HC	8	1.5	16	4	>1024	3	12	6	A6
B67010	Endocarditis	N	8	1.5	>32	8	>1024	3	12	6	A7
B37957	Surgical wound	N	8	1.5	>32	6	>1024	3	12	6	A8
B84847	HBP tract	N	5	1.5	>32	2	16	3	2	23	NDe
B56765	Undetermined	N	8	4	>32	32	>1024	4	4	28	ND
B16457	Undetermined	N	8	4	32	8	96	4	New	New	ND
Isolates with full susceptibility to penicillin
B84828	HBP tract	N	2	2	4	2	16	3	12	New	ND
B115329	Abdomen	N	4	2	3	2	24	2	12	New	ND
B104838	Urinary tract	N	8	2	2	2	12	3	1	ND	ND
B56478	Urinary tract	C	3	1.5	4	1.5	64	3	3	ND	ND
B96280	Urinary tract	N	8	1	1	2	12	3	3	ND	ND
B30537	Undetermined	HC	8	1.5	4	2	>1024	3	5	ND	ND
B150692	Endocarditis	N	8	0.5	1.5	3	16	3	5	ND	ND
B126859	Urinary tract	N	4	3	6	2	16	3	7	ND	ND
B70401	Biliary tract	N	8	2	12	2	48	4	7	ND	ND
B123830	IV catheter	N	8	3	3	2	24	4	7	ND	ND
B75968	Undetermined	HC	7	1.5	2	2	48	3	16	ND	ND
B28054	Pancreas	N	8	1	6	2	48	3	20	ND	ND
B128229	Surgical wound	N	8	3	12	1.5	24	4	27	ND	ND

^aAVP, aorthic vascular prosthesis; HBP, hepato-biliary-pancreatic.

^bN, nosocomial onset; C, community onset; HC, healthcare associated; NH, nursing home associated.

^c1, Brovst Hospital; 2, Dronninglund Hospital; 3, Farsø Hospital; 4, Frederikshavn Hospital; 5, Hjørring Hospital; 6, Hobro Hospital; 7, Aalborg Hospital section North; 8, Aalborg Hospital section South.

^dAMP, ampicillin; PEN, penicillin; IPM, imipenem; GEN, gentamicin; VAN, vancomycin.

^eND, not determined.

Due to the unusual susceptibility pattern obtained by Etest, the penicillin MICs were measured by broth microdilution (BMD) according to the Clinical and Laboratory Standards Institute (CLSI) (1). Twofold dilutions of penicillin G sodium salt (Sigma-Aldrich Denmark A/S, Copenhagen, Denmark) in the range between 0.5 and 32 μg/ml were prepared using Mueller-Hinton broth (MHB, Oxoid A/S, Greve, Denmark). Thirteen E. faecalis blood culture isolates from hospitalized patients obtained during the same year were randomly selected for comparative purposes based on their susceptibility to penicillin determined by Etest (MICs ≤ 12 μg/ml) (Table 1). All 20 isolates were reported as being penicillin resistant according to Etest and the 13 control isolates were found to be penicillin susceptible by BMD based on the CLSI interpretive MIC breakpoint (S ≤ 8 μg/ml) and fell into the wild-type E. faecalis MIC distribution (1 to 16 μg/ml), provided by the European Committee on Antimicrobial Susceptibility Testing (EUCAST). However, using BMD according to CLSI standards, the penicillin MICs for the isolates that were reported to be penicillin resistant according to Etest were two- to eightfold higher (4 to 8 μg/ml) than those for the control isolates reported to be susceptible by Etest (1 to 2 μg/ml), with the exception of a single isolate displaying a MIC of 2 μg/ml. Gentamicin resistance (MIC ≥ 1,024 μg/ml) was confirmed in 17 of the 20 isolates displaying reduced susceptibility to penicillin, whereas the two isolates displaying imipenem resistance were susceptible to this agent (MICs ≤ 2 μg/ml).

Multilocus sequence typing (MLST) (2) showed that most (17/20) E. faecalis isolates with reduced penicillin susceptibility belonged to sequence type 6 (ST6), whereas none of the control isolates belonged to ST6, as shown by partial (gdh gene) MLST typing. Pulsed-field gel electrophoresis (PFGE) (3) demonstrated closely related patterns (≤5-band difference) among the 17 ST6 isolates with reduced penicillin susceptibility (data not shown). Eight PFGE profiles (A1 to A8) were designated on the basis of minor band differences, and the most common profile (A1) was found in six isolates originating from four hospitals (Table 1).

We report the occurrence in several Danish hospitals of a distinctive penicillin susceptibility phenotype in E. faecalis consisting of reduced penicillin susceptibility and full ampicillin susceptibility. Most isolates displaying this phenotype were additionally resistant to gentamicin, belonged to type ST6, and had indistinguishable or closely related PFGE patterns, indicating the occurrence of a clonal outbreak. Despite the increased MICs of penicillin and imipenem obtained by Etest, the ST6 isolates were susceptible according to CLSI standards. The results draw attention to the ambiguity of penicillin and imipenem MICs determined by concentration gradient diffusion assays such as Etest. In laboratories where these assays are routinely employed for susceptibility testing of E. faecalis, resistant isolates should be confirmed by broth dilution. As long as the mechanism leading to in vitro-reduced penicillin susceptibility in E. faecalis ST6 remains unknown, the use of penicillin should be considered with caution for treatment of infections caused by strains displaying this atypical phenotype.