Journal of Antimicrobial Chemotherapy Advance Access published March 19, 2014

J Antimicrob Chemother
doi:10.1093/jac/dku065

In vitro activity and post-antibiotic effects of colistin in combination

with other antimicrobials against colistin-resistant KPC-producing
Klebsiella pneumoniae bloodstream isolates
Paolo Gaibani1*, Donatella Lombardo1, Russell E. Lewis2, Marcella Mercuri1, Sonia Bonora1,
Maria Paola Landini1 and Simone Ambretti1

Downloaded from http://jac.oxfordjournals.org/ at TOBB Ekonomi ve Teknoloji Universitesi on May 21, 2014
1
Operative Unit of Clinical Microbiology, S. Orsola-Malpighi University Hospital, Bologna, Italy; 2Department of Medical Sciences and
Surgery, Operative Unit of Infectious Diseases, S. Orsola-Malpighi University Hospital, Bologna, Italy

*Corresponding author. Tel: +39-051-6364316; Fax: +39-051-6363076; E-mail: paolo.gaibani@unibo.it

Received 19 December 2013; returned 22 January 2014; revised 11 February 2014; accepted 12 February 2014

Objectives: Combination therapy is recommended for the treatment of KPC-producing Klebsiella pneumoniae
(KPC-Kp), but the optimal regimen for colistin-resistant strains is unknown. We compared the synergistic activity
and post-antibiotic effect (PAE) of colistin in combination with other antimicrobials against colistin-susceptible
and -resistant KPC-Kp bloodstream isolates.
Methods: The genotypes of nine colistin-susceptible and eight colistin-resistant KPC-Kp bloodstream isolates
were analysed using PCR and amplicon sequencing. Combinations of colistin, meropenem, tigecycline, rifampicin
and teicoplanin were then screened using the Etest, a chequerboard assay and time – kill studies. Synergistic
combinations were also analysed with respect to the PAE in time –kill curves and the PAE at clinically achievable
concentrations.
Results: Insertional inactivation of the PhoQ/PhoB two-component regulatory system by mgrB-IS5 was identified in
6/8 (75%) colistin-resistant KPC-Kp. Colistin/rifampicin combinations resulted in no interactions [fractional inhibi-
tory concentration (FIC) indices 1.5 –2] for colistin-susceptible strains, but were uniformly synergistic (FIC indices
0.1 –0.4) against colistin-resistant KPC-Kp. Time –kill kinetic analysis, at clinically achievable fixed concentrations
of rifampicin and colistin, confirmed synergy and produced persistent growth inhibition (3 h) of colistin-resistant
KPC-Kp strains exposed to colistin/rifampicin or colistin/rifampicin/tigecycline combinations.
Conclusions: Combinations of colistin plus rifampicin, and less frequently tigecycline, exhibited synergistic activity
in vitro against colistin-resistant KPC-Kp strains.

Keywords: colistin resistance, K. pneumoniae, synergy assays, PAEs, bacteraemia

Introduction including carbapenemase and extended-spectrum b-lactamase

Klebsiella pneumoniae is a Gram-negative pathogen implicated in
hospital-acquired urinary tract infections, nosocomial pneumonia
(including ventilator-associated pneumonia), intra-abdominal
infections and bloodstream infections.1 – 3 During the last decade
the increasing spread of carbapenem-resistant Enterobacteriaceae
has been documented worldwide.4 – 7 In particular, K. pneumoniae
producing K. pneumoniae carbapenemase (KPC) has been identified
as a major public health threat because of the rapid plasmid-
mediated spread of resistance and limited available therapeutic
options.8,9
Treatment of KPC-producing K. pneumoniae infections is com-
plicated by: the presence of multiple resistance mechanisms,
production; mutations in DNA gyrase and porins; and the produc-
tion of aminoglycoside-modifying enzymes.10 Consequently, only
a few remaining drugs have reliable activity against the epidemic
KPC-producing K. pneumoniae sequence type (ST) 258 strain,
namely colistin, gentamicin and tigecycline.11 Unfortunately,
these antibiotics have pharmacokinetic and pharmacodynamic
limitations for treating bloodstream infections and pneumonia.2
Moreover, several recent reports have described the emergence
and spread of colistin- and tigecycline-resistant KPC-producing
K. pneumoniae ST258 and ST512 clones.8,12,13
Several studies have aimed to identify the most effective treat-
ment for bloodstream infections caused by these pathogens, by
comparing antibiotic monotherapy and combination therapies.14,15

# The Author 2014. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved.
For Permissions, please e-mail: journals.permissions@oup.com

1 of 10
Gaibani et al.
Combinations of two or more drugs seem to be more effective
than single antibiotic treatment in terms of improved outcome
in KPC K. pneumoniae bloodstream infections.10,14,15 However,
there are only few data about the in vitro activity of different com-
binations of antibiotics against this type of multidrug-resistant
organism. In a recent study, the combination of colistin plus
rifampicin showed synergistic antimicrobial activity against 13
KPC-producing K. pneumoniae colistin-resistant strains isolated
from different pathological samples.16 Based on these findings,
we investigated the activity and post-antibiotic effects (PAEs) of
colistin in combination with meropenem, tigecycline, rifampicin
and teicoplanin against 17 KPC-producing K. pneumoniae strains
isolated from bloodstream infections.
Materials and methods
Bacterial isolates
We tested 17 K. pneumoniae bloodstream isolates cultured from different
hospitalized patients at our institution (S. Orsola-Malpighi Hospital,
Bologna) from 1 March 2011 to 30 April 2012. The strains were categorized
based on their antimicrobial susceptibility (n ¼ 9) or resistance (n ¼ 8) to
colistin. Routine identification and antimicrobial susceptibility testing
were performed using a Vitek2 semi-automated system (bioMérieux,
France). MIC results were interpreted following EUCAST guidelines.17 The
modified Hodge test and a commercial disc diffusion synergy assay
(Rosco Diagnostica, Taastrup, Denmark) were utilized as confirmatory
tests to detect and discriminate carbapenemase producers.18
The presence of the KPC gene was investigated by PCR and the sequen-
cing of corresponding amplicons, as previously described.7,19 The genetic
relationships among the different K. pneumoniae isolates were investi-
gated by multilocus sequence typing (MLST) following the method
described by Diancourt et al. 20
To investigate the mechanism of colistin resistance among the
KPC-producing K. pneumoniae strains, the inactivation or deletion of the
mgrB gene among the colistin-resistant and -susceptible isolates was
evaluated by PCR as previously described.21
Antibiotic testing
Antibiotics were selected for this study based on a review of the literature
on the different synergistic antimicrobial combinations against different
multidrug-resistant Gram-negative bacteria.10,15,16 The following anti-
biotics were used in the chequerboard and Etest synergy assays: rifam-
picin, meropenem, tigecycline, teicoplanin and colistin sulphate.
Antibiotics were obtained from standard commercial sources and pre-
pared as previously described.22 Antibiotic combinations for the time –
kill assays were selected based on clinically achievable concentration
ranges where synergy was observed by the chequerboard dilution and
Etest assays.
Etest synergy assay
MICs were determined using the Etest method according to the manufac-
turer’s instructions (bioMérieux, France). Antibiotic combinations tested
using Etest strips were placed on cation-adjusted Mueller – Hinton agar
plates, crossed with a 908 angle at the intersection between relative MIC
for each K. pneumoniae isolate and incubated at 378C for 24 h.23
Chequerboard assay
The MICs of the selected antibiotics were determined using the broth
microdilution reference method. 22 In brief, 100 mL of serial 2-fold
2 of 10
dilutions of each drug were dispensed in a 96-well microtitre plate
(Sarstedt Inc., Newton, USA) and incubated with each test organism
to yield the appropriate density of 5×10 5 cfu/mL, at 378C for 24 h.
The MIC value was defined as the minimum antibiotic concentration
with no visible growth. The synergy assay combined the two antibiotics
by diluting both in serial 2-fold dilutions, along the ordinate for the drug
A and the abscissa for the drug B. The MICs of single drugs in combin-
ation were determined after 24 h of incubation at 378C as described
earlier. 22 Each drug combination was tested in triplicate on differ-
ent days.
Kill-curve assay
Downloaded from http://jac.oxfordjournals.org/ at TOBB Ekonomi ve Teknoloji Universitesi on May 21, 2014
Time– kill assays were performed by inoculating 5×105 cfu/mL of the test
organism into cation-adjusted Mueller– Hinton broth. The antibiotics were
tested alone and in combination at fixed concentrations (meropenem
8 mg/L, colistin 2 mg/L, tigecycline 0.125 mg/L, rifampicin 2 mg/L, teico-
planin 10 mg/L). Antimicrobial concentrations were selected on the
basis of a review of the literature that defined the clinically achievable
serum levels for each drug.24 – 28 In order to determine the viable bacteria
count during the time – kill experiments, 100 mL of each suspension was
removed, serially diluted 10-fold in PBS and plated on blood agar plates
at defined timepoints (0, 4, 8, 16 and 24 h post-inoculation). The bacterial
load (log10 cfu/mL) was determined by enumerating the colonies after
18– 22 h of incubation at 378C. The lower limit of accurately quantifiable
cfu using this method was 1 log10 of viable bacteria per mL. All experi-
ments were performed in triplicate.
Synergy interpretations
The fractional inhibitory concentration (FIC) index (SFIC) was calculated
both for the Etest and chequerboard assays by using the following for-
mula: SFIC¼FIC of agent A+FIC B, where FIC A is the MIC of the combin-
ation/MIC of drug A alone, and FIC B is the MIC of the combination/MIC of
drug B alone. SFICI results were interpreted with the following criteria, as
previously described:29 synergy, FICI ≤0.5; no interaction, FICI .0.5 – 4;
antagonism, FICI.4.
For the time–kill experiments, synergy was defined as a ≥2 log10 cfu/
mL decrease at 24 h for the antimicrobial combination compared with the
most active single agent. No interaction was defined as a ,2 log10 cfu/mL
increase or decrease at 24 h for the drug combination in comparison with
the most active antibiotic alone. Bactericidal activity was defined as a
.3 log10 cfu/mL reduction for the antibiotic combination treatment com-
pared with the untreated control at the start of each assay from the ori-
ginal inocula, whereas bacteriostatic activity was defined as a ,3 log10
cfu/mL decrease.
PAE
PAE was determined by testing the antibiotic combinations that showed
synergy according to both the chequerboard and Etest methods. Test
isolates were exposed to antibiotics in fixed combinations (2 mg/L rifam-
picin and 2 mg/L colistin; 2 mg/L rifampicin, 2 mg/L colistin and
0.125 mg/L tigecycline) for 2 h at 378C, followed by washout, centrifuga-
tion and resuspension in 10 mL of Mueller – Hinton broth and incubation
at 378C, as described previously.30 At defined timepoints (1, 2, 3, 4, 5, 6,
7, 8 and 24 h after antibiotic exposure) viable bacteria expressed as log10
cfu/mL were counted by plating the serial 10-fold dilutions on Mueller –
Hinton agar. Bacteria control curves were determined with the same
model without antibiotics. The PAE was determined by calculating the
difference between T and C, as previously described. 31 Briefly, T was
determined by calculating the time for the bacterial culture (cfu/mL)
to increase by 1 log10 the number of viable organisms with respect to
Synergy against colistin-resistant K. pneumoniae
the number of live bacteria counted after the removal of antibiotics, and
C was defined as the time of the control culture.
Results
Phenotypic and genotypic characterization
of KPC K. pneumoniae
From March 2011 to April 2012, 17 KPC-producing K. pneumoniae
strains were identified among patients with bacteraemia caused by
KPC-producing K. pneumoniae. All isolates were resistant to
b-lactams, including carbapenems, and to various other antimicro-
JAC
Synergy evaluation
Synergy was determined by comparing the Etest and broth micro-
dilution chequerboard methods and then confirmed by time –kill
analysis. The Etest showed no interactions for any antibiotic com-
binations tested among all colistin-susceptible strains (Table 2).
Similar results were obtained using the chequerboard assay
(Table 3).
When the antibiotic combinations were tested against the
colistin-resistant K. pneumoniae isolates using the Etest, all strains
showed synergy for the combination of colistin sulphate with
rifampicin, while two (25%) strains and one (12.5%) strain
showed synergistic interactions when colistin sulphate was tested

bials (amikacin, fluoroquinolones, trimethoprim/sulfamethoxazole
etc.), while all strains were susceptible or intermediate to gentami-
cin. Eleven percent (2/17) of isolates were also resistant to tigecyc-
line (isolates 43C and 2I). The isolates were grouped based on
colistin susceptibility, with 47% (8/17) resistant to colistin, with
MICs of 16 mg/L (1/8), 32 mg/L (3/8), 64 mg/L (2/8) and 128 mg/
L (2/8), with a mean of 64 and a standard deviation of 41.
Analysis of carbapenemase genetic determinants revealed the
presence of blaKPC-3 genes in all isolates (Table 1). MLST analysis
was performed to determine the strain relationships among the
KPC-producing K. pneumoniae isolates. Interestingly, MLST showed
that most of the colistin-susceptible strains belonged to the epi-
demic K. pneumoniae ST512 clone, while most of the colistin-
resistant strains belonged to the ST258 epidemic clone (Table 1).
Genotypic analysis by both PCR and DNA sequencing revealed
that six of the eight colistin-resistant K. pneumoniae isolates had
an insertional inactivation of the mgrB gene by an IS5-like sequence
(Table 1). However, two colistin-resistant K. pneumoniae isolates
(named 7C and 12C) showed a complete mgrB gene sequence,
suggesting a different mechanism of resistance.
Downloaded from http://jac.oxfordjournals.org/ at TOBB Ekonomi ve Teknoloji Universitesi on May 21, 2014
with tigecycline and meropenem, respectively (Table 2). At the
same time, no synergistic interactions were observed for rifampi-
cin in combination either with tigecycline or meropenem (data not
shown). In addition, colistin sulphate plus teicoplanin showed no
interaction result for any tested strains.
The chequerboard assay showed synergy for colistin sulphate
plus meropenem against three of eight colistin-resistant strains
(Table 3). However, a higher percentage (75%) of colistin-resistant
K. pneumoniae isolates displayed a synergistic (6/8) effect for
colistin sulphate plus tigecycline. Consistent with the Etest results,
synergy was observed for all colistin-resistant strains that were
tested with colistin sulphate plus rifampicin (Table 3).
Additionally, the combination of teicoplanin and colistin sulphate
displayed no interaction against any colistin-resistant K. pneumo-
niae strains.
To better evaluate the pharmacodynamics of antibiotic inter-
actions on the basis of achievable serum levels of each drug,
time – kill kinetic assays were performed for all colistin-resistant
strains at fixed drug concentrations. The results showed a syner-
gistic effect for colistin sulphate plus rifampicin against all the

Table 1. Clinical and molecular data for colistin-susceptible and -resistant KPC-producing K. pneumoniae isolates included in the study

Carbapenemase Colistin resistance

Strain Site of isolation Date of collection gene MLST determinant

27B blood 04/04/2011 KPC-3 512 mgrB a

47B blood 13/03/2011 KPC-3 512 mgrB a
32B blood 18/03/2011 KPC-3 258 mgrB a
56C blood 13/06/2011 KPC-3 512 mgrB a
51H blood 09/01/2012 KPC-3 512 mgrB a
79J blood 22/03/2012 KPC-3 512 mgrB a
36C blood 30/05/2011 KPC-3 258 mgrB a
14K blood 06/04/2012 KPC-3 512 mgrB a
43C blood 08/06/2011 KPC-3 512 mgrB a
26B blood 14/03/2011 KPC-3 258 IS5-like element b
54B blood 09/04/2011 KPC-3 258 IS5-like element b
81B blood 29/04/2011 KPC-3 258 IS5-like element b
2I blood 23/01/2012 KPC-3 512 IS5-like element b
12C blood 07/05/2011 KPC-3 258 mgrB a
2D blood 09/07/2011 KPC-3 258 IS5-like element b
80B blood 29/04/2011 KPC-3 258 IS5-like element b
7C blood 09/05/2011 KPC-3 258 mgrB a

Bold formatting indicates colistin-resistant KPC-producing K. pneumoniae strains.

a
Complete mgrB gene sequence.
b
mgrB gene interrupted by an IS5-like element.

3 of 10
Gaibani et al.

Table 2. Results obtained using the Etest method for antibiotic combinations tested against colistin-susceptible and -resistant KPC-producing K.
pneumoniae

MIC (mg/L) CST/TGC CST/MEM MEM/TGC CST/RIF CST/TEC

Strain MEM CST TGC RIF TEC SFIC activity SFIC activity SFIC activity SFIC activity SFIC activity

27B ≥32 0.125 2 ≥32 ≥256 2 I 2 I 2 I 2 I 2 I

47B ≥32 0.125 2 ≥32 ≥256 2 I 1.8 I 2 I 1.5 I 2 I
32B 24 0.125 2 ≥32 ≥256 2 I 2 I 2 I 2 I 2 I
56C ≥32 0.125 2 ≥32 ≥256 2 I 1.6 I 2 I 2 I 2 I
51H ≥32 0.125 2 ≥32 ≥256 1.6 I 2 I 2 I 1.5 I 2 I
79J ≥32 0.125 2 ≥32 ≥256 2 I 1.6 I 2 I 2 I 2 I

Downloaded from http://jac.oxfordjournals.org/ at TOBB Ekonomi ve Teknoloji Universitesi on May 21, 2014
36C ≥32 0.125 2 ≥32 ≥256 1.6 I 2 I 2 I 1.5 I 2 I
14K ≥32 0.125 2 ≥32 ≥256 2 I 2 I 2 I 2 I 2 I
43C ≥32 0.190 4 ≥32 ≥256 1.6 I 2 I 1.5 I 2 I 2 I
26B ≥32 16 2 ≥32 ≥256 1.4 I 1.25 I 1.5 I 0.3 S 1.2 I
54B ≥32 24 2 ≥32 ≥256 1.3 I 0.8 I 1.5 I 0.2 S 1.5 I
81B ≥32 24 2 ≥32 ≥256 0.4 S 0.6 I 1.5 I 0.3 S 2 I
2I ≥32 24 16 ≥32 ≥256 0.4 S 1.6 I 2 I 0.5 S 1.3 I
12C ≥32 24 2 ≥32 ≥256 1.5 I 0.4 S 1.3 I 0.4 S 2 I
2D ≥32 16 2 ≥32 ≥256 1.4 I 1.5 I 1.5 I 0.1 S 2 I
80B ≥32 32 2 ≥32 ≥256 1.2 I 1 I 2 I 0.2 S 2 I
7C ≥32 32 2 ≥32 ≥256 1 I 1.3 I 1.5 I 0.2 S 2 I

MEM, meropenem; CST, colistin sulphate; TGC, tigecycline; RIF, rifampicin; TEC, teicoplanin; S, synergy; I, no interaction; AN, antagonism.
Bold formatting indicates the colistin-resistant KPC-producing K. pneumoniae strains.

Table 3. Results obtained using the chequerboard method for antibiotic combinations tested against colistin-susceptible and -resistant KPC-producing
K. pneumoniae

MIC (mg/L) CST/TGC CST/MEM MEM/TGC CST/RIF CST/TEC

Strain MEM CST TGC RIF TEC SFIC activity SFIC activity SFIC activity SFIC activity SFIC activity

27B 32 0.064 2 64 .256 2 I 2 I 1.5 I 1.5 I 2 I

47B 128 0.032 2 64 .256 2 I 2 I 2 I 2 I 2 I
32B 32 0.064 2 32 .256 2 I 2 I 1.5 I 1 I 2 I
56C 128 0.064 2 64 .256 2 I 2 I 2 I 1.5 I 2 I
51H 64 0.064 2 64 .256 2 I 2 I 1.5 I 2 I 2 I
79J 128 0.032 2 64 .256 2 I 2 I 1.5 I 1.5 I 2 I
36C 16 0.125 2 64 .256 2 I 2 I 2 I 1.3 I 2 I
14K 128 0.064 2 64 .256 2 I 2 I 0.8 I 1.2 I 2 I
43C 64 0.064 4 64 .256 2 I 2 I 1.5 I 1.3 I 2 I
26B 16 16 2 64 >256 0.2 S 0.6 I 1.5 I 0.1 S 2 I
54B 32 64 2 64 >256 0.2 S 0.6 I 2 I 0.1 S 2 I
81B 16 32 2 64 >256 0.2 S 1.5 I 2 I 0.1 S 2 I
2I 64 32 16 256 >256 0.1 S 1.5 I 2 I 0.2 S 2 I
12C 16 128 2 256 >256 1.2 I 0.3 S 2 I 0.1 S 2 I
2D 32 64 2 32 >256 0.6 I 0.3 S 1.5 I 0.1 S 2 I
80B 16 32 2 32 >256 0.2 S 0.6 I 0.6 I 0.1 S 2 I
7C 64 128 2 32 >256 0.2 S 0.3 S 0.8 I 0.1 S 2 I

MEM, meropenem; CST, colistin sulphate; TGC, tigecycline; RIF, rifampicin; TEC, teicoplanin; S, synergy; I, no interaction; AN, antagonism.
Bold formatting indicates the colistin-resistant KPC-producing K. pneumoniae strains.

colistin-resistant KPC-producing K. pneumoniae strains (Figure 1). the combination of colistin sulphate/rifampicin plus tigecycline
However, none of the other antibiotic combinations showed a syn- yielded a synergistic effect with a similar bacterial reduction count
ergistic effect at the defined antibiotic concentrations. In addition, to that observed for colistin sulphate plus rifampicin (Figure 1).

4 of 10
Synergy against colistin-resistant K. pneumoniae JAC
26B 54B
1011 1011
1010 1010
109 109
Log10 cfu/mL

Log10 cfu/mL
108 108
107 107
106 106
105 105
104 104
103 103
102 102

0 4 8 16 24 0 4 8 16 24
Time (h) Time (h)

Downloaded from http://jac.oxfordjournals.org/ at TOBB Ekonomi ve Teknoloji Universitesi on May 21, 2014
81B 2I
1012 1011
1011 1010
1010 109

Log10 cfu/mL
Log10 cfu/mL

109 108
108 107
107
106
106
105 105
104 104
103 103
102 102

0 4 8 16 24 0 4 8 16 24
Time (h) Time (h)

7C 2D
1011 1011
1010 1010
109 109
Log10 cfu/mL

Log10 cfu/mL

108 108
107 107
106 106
105 105
104 104
103 103
102 102

0 4 8 16 24 0 4 8 16 24
Time (h) Time (h)

1011 80B 1011 12C

1010 1010
109 109
Log10 cfu/mL

108
Log10 cfu/mL

108
107 107
106 106
105 105
104 104
103
103
102
102
0 4 8 16 24 0 4 8 16 24
Time (h) Time (h)

MEM TGC RIF CST CST/MEM CST/RIF

MEM/CST/TGC CST/TGC CST/RIF/TGC Untreated culture

Figure 1. In vitro evaluation of the bactericidal activities of various antibiotic combinations against eight colistin-resistant K. pneumoniae isolates tested
using a kill-curve assay. Time– kill analysis was performed for isolate/drug combinations that displayed synergy according to the Etest or chequerboard
assay. The light grey zone indicates a synergistic effect and the dark grey zone indicates bactericidal activity. MEM, meropenem; CST, colistin sulphate;
TGC, tigecycline; RIF, rifampicin.

5 of 10
Gaibani et al.

26B 54B
1012 Antibiotics removed 1013 Antibiotics removed
1011 1012
1010 1011
109 1010
Log10 cfu/mL

Log10 cfu/mL
108 109
107 108
106 107
106
105 105
104 104
103 PAE = 2 h 103 PAE = 2 h
102 102
101 101
0 2 4 6 8 10 12 14 16 18 20 22 24 0 2 4 6 8 10 12 14 16 18 20 22 24
Time (h) Time (h)

Downloaded from http://jac.oxfordjournals.org/ at TOBB Ekonomi ve Teknoloji Universitesi on May 21, 2014
81B 2I
1013 Antibiotics removed Antibiotics removed
1013
1012 1012
1011 1011
1010 1010
Log10 cfu/mL

Log10 cfu/mL
109 109
108 108
107 107
106 106
105 105
104 104 PAE = 0.1 h
103 PAE = 1.5 h 103
102 102
101 101
0 2 4 6 8 10 12 14 16 18 20 22 24 0 2 4 6 8 10 12 14 16 18 20 22 24
Time (h) Time (h)
7C 2D
1013 Antibiotics removed Antibiotics removed
1012
1012 1011
1011 1010
1010 109
Log10 cfu/mL
Log10 cfu/mL

109 108
108 107
107
106 106
105 105
104 104
PAE = 1.5 h 103 PAE = 2 h
103
102 102
101 101
0 2 4 6 8 10 12 14 16 18 20 22 24 0 2 4 6 8 10 12 14 16 18 20 22 24
Time (h) Time (h)
80B 12C
1013 Antibiotics removed 1013 Antibiotics removed
1012 1012
1011 1011
1010 1010
Log10 cfu/mL

Log10 cfu/mL

109 109
108 108
107 107
106 106
105 105
104 104
103 PAE = 1 h 103 PAE = 4.5 h
102 102
101 101
0 2 4 6 8 10 12 14 16 18 20 22 24 0 2 4 6 8 10 12 14 16 18 20 22 24
Time (h) Time (h)

Figure 2. PAE of the combination of colistin plus rifampicin (triangles) in comparison with untreated control cultures (squares) of the colistin-resistant
KPC-producing K. pneumoniae strains.

Analysis of cfu counts further revealed that colistin sulphate/ resistant isolates this combination had bacteriostatic activity
rifampicin plus tigecycline exhibited bactericidal activity against (Figure 1). At the same time, colistin sulphate plus rifampicin dis-
only a single isolate (7C), whereas for the remaining colistin- played bacteriostatic activity against all colistin-resistant strains.

6 of 10
Synergy against colistin-resistant K. pneumoniae
Table 4. Comparative PAE of colistin/rifampicin and colistin/rifampicin/
tigecycline against colistin-resistant K. pneumoniae strains
PAE (h)
DPAE+tigecycline
Strain CST/RIF CST/RIF/TGC (CST/RIF/TGC2CST/RIF)
26B 2 3 1 pneumoniae isolates.
54B 2 2.5 0.5
81B 1.5 2 0.5
2I 0.1 0.3 0.2
7C 1.5 3 1.5
2D 2 3 1 colistin in combination with meropenem or tigecycline, as we
80B 1 4 3 observed in several colistin-resistant KPC-producing K. pneumo-
12C 4.5 5 0.5
Mean 1.8 2.8 1 Notably, we observed synergistic in vitro activity for colistin plus
CST, colistin sulphate; TGC, tigecycline; RIF, rifampicin.
PAE
PAE was determined based on synergy results obtained with
antibiotic combinations observed in the kill-curve assay. PAE
results for colistin sulphate plus rifampicin for each strain
are shown in Figure 2. A mean PAE of 1.8 h was observed follow-
ing incubation of colistin sulphate plus rifampicin at fixed con-
centrations (2 mg/L for each drug) for 2 h at 378C. In detail, PAE
values ranged from 0.1 h (strain 2I) to 4.5 h (12C), as shown in
Table 4.
PAE values for colistin sulphate/rifampicin plus tigecycline
(0.125 mg/L) against each KPC-producing K. pneumoniae strain
are shown in Figure 3. The mean PAE was 2.8 h, ranging from
0.3 to 5 h. In detail, the PAE for strain 2I (which was tigecycline
resistant) turned out to be much shorter than for the other
colistin-resistant K. pneumoniae isolates, whereas the PAE for
strain 80B was prolonged (Table 4).
Comparison of the PAE of two antibiotic combinations showed
that when tigecycline was added to colistin sulphate plus rifampi-
cin the bacteriostatic effect was longer, with a mean increase of
1 h (Table 4).
Discussion
KPC-producing K. pneumoniae has spread rapidly worldwide over
the last decade.3,4,8 Its broad antibiotic resistance and the limited
available therapeutic options have created a need for new thera-
peutic approaches to treat infections caused by this pathogen.
Polymyxins were considered, until recently, one of the last avail-
able therapeutic options for carbapenem-resistant infections.1
However, colistin resistance has increased among KPC-producing
K. pneumoniae over the last few years.32 – 35 Genotypic analy-
sis conducted among all the colistin-resistant K. pneu-
moniae strains demonstrated that six isolates (75% of the
colistin-resistant strains) carried an insertional inactivation of
the mgrB gene. The disruption of the mgrB gene by an IS5-like
element was recently proposed as an important mechanism
related to the colistin-resistant phenotype in carbapenem-
resistant Enterobacteriaceae.21 However, two isolates (25% of
the colistin-resistant strains) carried a complete sequence of the
JAC
mgrB gene, suggesting that an alternative mechanism is respon-
sible for colistin resistance. In a previous in silico study, we pro-
posed three other candidate genes involved in the biogenesis
and modification of the outer membrane (waaL, rfbA, vacJ) that
could implicated in colistin resistance in K. pneumoniae.35
Further studies will be performed to completely characterize the
molecular mechanism of colistin resistance in KPC-producing K.
Several studies have proposed that a combination of colistin
plus a carbapenem or tigecycline are the most active treatment
for colistin-resistant KPC-producing K. pneumoniae bacter-
aemia.10,15,36 Similarly, our data suggest synergistic activity for
Downloaded from http://jac.oxfordjournals.org/ at TOBB Ekonomi ve Teknoloji Universitesi on May 21, 2014
niae using the Etest and chequerboard assays.
rifampicin in all colistin-resistant KPC-producing K. pneumoniae
strains. At the same time, no synergistic interactions for colistin in
combination with teicoplanin were observed, either in colistin-
susceptible or -resistant KPC-producing K. pneumoniae. These find-
ings are in agreement with previous studies that demonstrated the
effectiveness of a combination of polymyxins with rifampicin
against colistin-resistant KPC-producing K. pneumoniae.16,37 A pos-
sible mechanism for the synergy observed between colistin and
rifampicin could be related to the activity of each drug being via dif-
ferent cellular pathways.38 Specifically, colistin-mediated damage
to the bacterial cell membrane likely enhanced the penetration
and accumulation of rifampicin in Gram-negative bacteria. In this
regard, the combination of rifampicin with tigecycline or merope-
nem showed no synergistic activity, supporting that it is the ability
of colistin to disrupt the bacterial membrane integrity that allows
the entry and accumulation of rifampicin. Rifampicin may have
some beneficial activity for biofilm-embedded organisms in the
gut and central venous catheters,38 although this mechanism
was not specifically tested in our experiments.
Combinations of polymyxins with several agents against
multidrug-resistant pathogens (Acinetobacter baumannii,
Pseudomonas aeruginosa, Serratia marcescens and K. pneumo-
niae) have been extensively investigated by in vitro studies.37 – 40
Synergistic inhibition was generally observed for the colistin/
rifampicin combinations against most multidrug-resistant patho-
gens, including carbapenem- and colistin-resistant strains.38,40
The results presented in the present study are hence consistent
with previous in vitro studies that demonstrated the effectiveness
of the combination of colistin and rifampicin.16,38,40 To best of our
knowledge, we have described here for the first time an in vitro PAE
for the combination of colistin and rifampicin against various
carbapenemase-producing Enterobacteriaceae (colistin-resistant
KPC-producing K. pneumoniae).
The findings of this study indicate that colistin and rifampicin in
combination show an evident and potent in vitro PAE activity
against colistin-resistant KPC-producing K. pneumoniae. In add-
ition, the PAE activity of the colistin and rifampicin combination
was further prolonged by tigecycline.
The clinical benefit of colistin and rifampicin combination are
still controversial.38,40 Our data suggest that a colistin plus rifam-
picin and tigecycline regimen could represent an effective option
for restoring some colistin activity against multidrug-resistant
strains. Use of this combination may also reduce the risk of select-
ing colistin heteroresistant populations.41,42
7 of 10
Gaibani et al.

26B 54B
1013 Antibiotics removed 1013 Antibiotics removed
1012 1012
1011 1011
1010 1010
Log10 cfu/mL

Log10 cfu/mL
109 109
108 108
107 107
106 106
105 105
104 104
103 PAE = 3 h 103 PAE = 2.5 h
102 102
101 101
0 2 4 6 8 10 12 14 16 18 20 22 24 0 2 4 6 8 10 12 14 16 18 20 22 24
Time (h) Time (h)

Downloaded from http://jac.oxfordjournals.org/ at TOBB Ekonomi ve Teknoloji Universitesi on May 21, 2014
81B 2I
1013 Antibiotics removed 1012 Antibiotics removed
1012 1011
1011 1010
1010 109
Log10 cfu/mL

Log10 cfu/mL
109 108
108 107
107 106
106 105
105
104 104 PAE = 0.3 h
103 PAE = 2 h 103
102 102
101 101
0 2 4 6 8 10 12 14 16 18 20 22 24 0 2 4 6 8 10 12 14 16 18 20 22 24
Time (h) Time (h)
7C 2D
1013 Antibiotics removed 1013 Antibiotics removed
1012 1012
1011 1011
1010 1010
Log10 cfu/mL
Log10 cfu/mL

109 109
108 108
107 107
106 106
105 105
104 104
103 PAE = 3 h 103 PAE = 3 h
102 102
101 101
0 2 4 6 8 10 12 14 16 18 20 22 24 0 2 4 6 8 10 12 14 16 18 20 22 24
Time (h) Time (h)
80B Antibiotics removed 12C
1013 Antibiotics removed 1013
1012 1012
1011 1011
1010 1010
Log10 cfu/mL

Log10 cfu/mL

109 109
108 108
107 107
106 106
105 105
104 104
103 103
102 PAE = 4 h 102 PAE = 5 h
101 101
100 100
0 2 4 6 8 10 12 14 16 18 20 22 24 0 2 4 6 8 10 12 14 16 18 20 22 24
Time (h) Time (h)

Figure 3. PAE of the combination of colistin plus rifampicin and tigecycline (triangles) in comparison with untreated control cultures (squares) of the
colistin-resistant KPC-producing K. pneumoniae strains.

Conclusions Further studies will be performed to identify the molecular mech-

Considering the potent and prolonged PAE activity of colistin, anism responsible for the synergistic interactions. Pharmacokinetic
rifampicin and tigecycline in combination, this synergistic associ- and pharmacodynamic data for the synergistic association are
ation appears to offer important implications for the dosing inter- needed to confirm our in vitro results and to better predict the in
val in the treatment of KPC-producing K. pneumoniae. vivo efficacy of these synergistic combinations in clinical practice.

8 of 10
Synergy against colistin-resistant K. pneumoniae
Acknowledgements
We are grateful to Dr Nicolò Girometti and Dr Michele Bartoletti for useful
discussions.
Funding
This work was partially supported by the Emilia-Romagna region and the
University of Bologna.
Transparency declarations
None to declare.
References
1 Kontopidou F, Giamarellou H, Katerelos P et al. Infections caused by
carbapenem-resistant Klebsiella pneumoniae among patients in
intensive care units in Greece: a multi-centre study on clinical outcome
and therapeutic options. Clin Microbiol Infect 2013; 20: O117–23.
2 Daikos GL, Markogiannakis A, Souli M et al. Bloodstream infections
caused by carbapenemase-producing Klebsiella pneumoniae: a clinical
perspective. Expert Rev Anti Infect Ther 2012; 10: 1393– 404.
3 Munoz-Price LS, Poirel L, Bonomo RA et al. Clinical epidemiology of the
global expansion of Klebsiella pneumoniae carbapenemases. Lancet
Infect Dis 2013; 13: 785–96.
4 Nordmann P, Naas T, Poirel L. Global spread of carbapenemase
producing Enterobacteriaceae. Emerg Infect Dis 2011; 17: 1791 –8.
5 Canton R, Akova M, Carmeli Y et al. Rapid evolution and spread of
carbapenemases among Enterobacteriaceae in Europe. Clin Microbiol
Infect 2012; 18: 413– 31.
6 Glasner C, Albiger B, Buist G et al. Carbapenemase-producing
Enterobacteriaceae in Europe: a survey among national experts from 39
countries, February 2013. Euro Surveill 2013; 18: pii¼20525.
7 Gaibani P, Ambretti S, Berlingeri A et al. Rapid increase of
carbapenemase-producing Klebsiella pneumoniae strains in a large
Italian hospital: surveillance period 1 March – 30 September 2010. Euro
Surveill 2011; 16: pii¼19800.
8 Giani T, Pini B, Arena F et al. Epidemic diffusion of KPC carbapenemase-
producing Klebsiella pneumoniae in Italy: results of the first countrywide
survey, 15 May to 30 June 2011. Euro Surveill 2013; 18: pii¼20489.
9 Lee GC, Burgess DS. Treatment of Klebsiella pneumoniae
carbapenemase (KPC) infections: a review of published case series and
case reports. Ann Clin Microbiol Antimicrob 2012; 11: 32.
10 Petrosillo N, Giannella M, Lewis R et al. Treatment of carbapenem-
resistant Klebsiella pneumoniae: the state of the art. Expert Rev Anti
Infect Ther 2013; 11: 159– 77.
11 Hirsch EB, Tam VH. Detection and treatment options for Klebsiella
pneumoniae carbapenemases (KPC): an emerging cause of multidrug-
resistant infection. J Antimicrob Chemother 2010; 65: 119–25.
12 Comandatore F, Sassera D, Ambretti S et al. Draft genome sequences
of two multidrug resistant Klebsiella pneumoniae ST258 isolates resistant
to colistin. Genome Announc 2013; 1: pii¼e00113–12.
13 Comandatore F, Gaibani P, Ambretti S et al. Draft genome of Klebsiella
pneumoniae sequence type 512, a multidrug-resistant strain isolated
during a recent KPC outbreak in Italy. Genome Announc 2013; 1:
pii¼e00035–12.
14 Zarkotou O, Pournaras S, Tselioti P et al. Predictors of mortality in
patients with bloodstream infections caused by KPC-producing Klebsiella
JAC
pneumoniae and impact of appropriate antimicrobial treatment. Clin
Microbiol Infect 2011; 17: 1798 –803.
15 Tumbarello M, Viale P, Viscoli C et al. Predictors of mortality in
bloodstream infections caused by Klebsiella pneumoniae carbapenemase-
producing K. pneumoniae: importance of combination therapy. Clin Infect
Dis 2012; 55: 943–50.
16 Tascini C, Tagliaferri E, Giani T et al. Synergistic activity of colistin plus
rifampicin against colistin-resistant KPC-producing Klebsiella pneumoniae.
Antimicrob Agents Chemother 2013; 57: 3990–3.
17 EUCAST. Breakpoint Tables for Interpretation for MICs and
Zone Diameters, Version 4.0. http://www.eucast.org/fileadmin/src/
media/PDFs/EUCAST_files/Breakpoint_tables/Breakpoint-table_v_4.0.pdf
(1 March 2014, date last accessed).
Downloaded from http://jac.oxfordjournals.org/ at TOBB Ekonomi ve Teknoloji Universitesi on May 21, 2014
18 Ambretti S, Gaibani P, Berlingeri A et al. Evaluation of phenotypic and
genotypic approaches for the detection of class A and class B
carbapenemases in Enterobacteriaceae. Microb Drug Resist 2013; 19: 212–5.
19 Ambretti S, Gaibani P, Caroli F et al. A carbapenem-resistant Klebsiella
pneumoniae isolate harboring KPC-1 from Italy. New Microbiol 2010;
33: 281– 2.
20 Diancourt L, Passet V, Verhoef J et al. Multilocus sequence typing of
Klebsiella pneumoniae nosocomial isolates. J Clin Microbiol 2005; 43:
4178– 82.
21 Cannatelli A, D’Andrea MM, Giani T et al. In vivo emergence of colistin
resistance in Klebsiella pneumoniae producing KPC-type carbapenemase
mediated by insertional inactivation of the PhoQ/PhoP mgrB regulator.
Antimicrob Agents Chemother 2013; 57: 5521– 6.
22 Andrews JM. Determination of minimum inhibitory concentrations. J
Antimicrob Chemother 2001; 48 Suppl 1: 5 – 16. Erratum in: J Antimicrob
Chemother 2002; 49: 1049.
23 Sopirala MM, Mangino JE, Gebreyes WA et al. Synergy testing by Etest,
microdilution checkerboard, and time– kill methods for pan-drug-resistant
Acinetobacter baumannii. Antimicrob Agents Chemother 2010; 54:
4678– 83.
24 Markou N, Markantonis SL, Dimitrakis E et al. Colistin serum
concentrations after intravenous administration in critically ill patients
with serious multidrug-resistant, Gram-negative bacilli infections: a
prospective, open-label, uncontrolled study. Clin Ther 2008; 30: 143–51.
25 Rodvold KA, Gotfried MH, Cwik M et al. Serum, tissue and body fluid
concentrations of tigecycline after a single 100 mg dose. J Antimicrob
Chemother 2006; 58: 1221– 9.
26 Hirsch EB, Guo B, Chang KT et al. Assessment of antimicrobial
combinations for Klebsiella pneumoniae carbapenemase-producing K.
pneumoniae. J Infect Dis 2013; 207: 786– 93.
27 Albur M, Noel A, Bowker K et al. Bactericidal activity of multiple
combinations of tigecycline and colistin against NDM-1-producing
Enterobacteriaceae. Antimicrob Agents Chemother 2012; 56: 3441– 3.
28 Nicolau DP. Pharmacokinetic and pharmacodynamic properties of
meropenem. Clin Infect Dis 2008; 47: 32 –40.
29 Odds FC. Synergy, antagonism, and what the chequerboard puts
between them. J Antimicrob Chemother 2003; 52: 1.
30 Ozbek B, Sentürk A. Postantibiotic effects of tigecycline, colistin sulfate,
and levofloxacin alone or tigecycline-colistin sulfate and tigecycline-
levofloxacin combinations against Acinetobacter baumannii.
Chemotherapy 2010; 56: 466–71.
31 Lavigne JP, Bonnet R, Michaux-Charachon S et al. Post-antibiotic and
post-b-lactamase inhibitor effects of ceftazidime plus sulbactam on
extended-spectrum b-lactamase-producing Gram-negative bacteria.
J Antimicrob Chemother 2004; 53: 616–9.
32 Capone A, Giannella M, Fortini D et al. High rate of colistin resistance
among patients with carbapenem-resistant Klebsiella pneumoniae
9 of 10
Gaibani et al.
infection accounts for an excess of mortality. Clin Microbiol Infect 2013;
19: 23 –30.
33 Halaby T, Al Naiemi N, Kluytmans J et al. Emergence of colistin
resistance in Enterobacteriaceae after the introduction of selective
digestive tract decontamination in an intensive care unit. Antimicrob
Agents Chemother 2013; 57: 3224 –9.
34 Mammina C, Bonura C, Di Bernardo F et al. Ongoing spread of
colistin-resistant Klebsiella pneumoniae in different wards of an acute
general hospital, Italy, June to December 2011. Euro Surveill 2012; 17:
pii:20248.
35 Sassera D, Comandatore F, Gaibani P et al. Comparative genomics of
closely related strains of Klebsiella pneumoniae reveals genes possibly
involved in colistin resistance. Ann Microbiol 2013; in press.
36 Qureshi ZA, Paterson DL, Potoski BA et al. Treatment outcome of
bacteremia due to KPC-producing Klebsiella pneumoniae: superiority of
combination antimicrobial regimens. Antimicrob Agents Chemother
2012; 56: 2108– 13.
10 of 10
37 Elemam A, Rahimian J, Doymaz M. In vitro evaluation of antibiotic
synergy for polymyxin B-resistant carbapenemase-producing Klebsiella
pneumoniae. J Clin Microbiol 2010; 48: 3558– 62.
38 Landman D, Georgescu C, Martin DA et al. Polymyxins revisited. Clin
Microbiol Rev 2008; 21: 449– 65.
39 Jamal MA, Rosenblatt JS, Hachem RY et al. Prevention of
Gram-negative bacterial biofilm on minocycline/rifampicin impregnated
catheters sequentially coated with chlorhexidine. Antimicrob Agents
Chemother 2013; 58: 1179– 82.
40 Yahav D, Farbman L, Leibovici L et al. Colistin: new lessons on an old
antibiotic. Clin Microbiol Infect 2012; 18: 18–29.
41 Cai Y, Chai D, Wang R et al. Colistin resistance of Acinetobacter
Downloaded from http://jac.oxfordjournals.org/ at TOBB Ekonomi ve Teknoloji Universitesi on May 21, 2014
baumannii: clinical reports, mechanisms and antimicrobial strategies. J
Antimicrob Chemother 2012; 67: 1607 –15.
42 Li J, Rayner CR, Nation RL. Hetero-resistance to colistin in
multidrug-resistant Acinetobacter baumannii. Antimicrob Agents
Chemother 2006; 50: 2946– 50.