Stenotrophomonas maltophilia - Antimicrobial Agents and ...

CONTENT ALERTS
In vitro activities of beta-lactam-beta-lactamase
inhibitor combinations against
Stenotrophomonas maltophilia: correlation
between methods for testing inhibitory activity,
time-kill curves, and bactericidal activity.
J L Muñoz Bellido, S Muñoz Criado, I García García, M A Alonso
Manzanares, M N Gutiérrez Zufiaurre and J A García-Rodríguez
Antimicrob. Agents Chemother. 1997, 41(12):2612.
Updated information and services can be found at:
http://aac.asm.org/content/41/12/2612
These include:
Receive: RSS Feeds, eTOCs, free email alerts (when new articles
cite this article), more»
Information about commercial reprint orders: http://journals.asm.org/site/misc/reprints.xhtml
To subscribe to to another ASM Journal go to: http://journals.asm.org/site/subscriptions/
Downloaded from
http://aac.asm.org/
on January 19, 2013 by guest

ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Dec. 1997, p. 2612–2615 Vol. 41, No. 12
0066-4804/97/$04.00�0
Copyright © 1997, American Society for Microbiology
In Vitro Activities of �-Lactam–�-Lactamase Inhibitor Combinations
against Stenotrophomonas maltophilia: Correlation between
Methods for Testing Inhibitory Activity, Time-Kill
Curves, and Bactericidal Activity
J. L. MUÑOZ BELLIDO, S. MUÑOZ CRIADO, I. GARCÍA GARCÍA, M. A. ALONSO MANZANARES,
M. N. GUTIÉRREZ ZUFIAURRE, AND J. A. GARCÍA-RODRÍGUEZ*
Department of Microbiology, University Hospital, University of Salamanca, Salamanca, Spain
Received 30 December 1996/Returned for modification 2 April 1997/Accepted 1 September 1997
The activities of ampicillin, ampicillin-sulbactam, amoxicillin, amoxicillin-clavulanic acid, ticarcillin, ticarcillin-clavulanic
acid, piperacillin, piperacillin-tazobactam, aztreonam, and aztreonam-clavulanic against
Stenotrophomonas maltophilia strains for which the MICs of penicillins and commercially available �-lactam–
�-lactamase inhibitor combinations were higher than the breakpoints usually recommended for Pseudomonas
aeruginosa in commercially available broth microdilution methods were tested by the agar diffusion, agar
dilution, and broth microdilution methods. Time-kill curve studies were performed when discrepancies between
these methods were observed. The MICs obtained by the commercially available broth microdilution
method, the agar dilution method, and the broth microdilution method were almost identical. Twenty-five
percent of the strains tested showed inhibition diameters of >15 mm for ticarcillin-clavulanic acid, and 43.7%
of the strains tested showed inhibition diameters of >18 mm for piperacillin-tazobactam by the agar diffusion
method. The time-kill curves for these strains confirmed the results obtained by dilution methods. Aztreonamclavulanic
acid (2:1) at concentrations of 32-fold the MIC).
With 50 times the standard inoculum, MBCs were at least 32-fold the MICs for all the strains tested.
Stenotrophomonas maltophilia is an aerobic, nonfermenting,
gram-negative bacillus that causes significant morbidity and
mortality in hospitalized patients, mainly debilitated, neutropenic,
and immunosuppressed patients (4). The treatment of
infections caused by this microorganism is difficult, because S.
maltophilia is frequently resistant to most of the widely used
antibiotics (5, 7, 15). This resistance is usually attributed to low
cell wall permeability (3, 16) and, in the case of �-lactams,
mainly to two �-lactamases (L1, a metallo-�-lactamase included
in group 3 of the Bush classification, and L2, a �lactamase
included in group 2e among the cephalosporinases
inhibited by clavulanic acid [2]).
The activities of �-lactam–�-lactamase inhibitor combinations
against S. maltophilia vary notably in different studies (1,
5, 11, 13, 14). The combination aztreonam-clavulanic acid
(AZT-C), although not available commercially, almost systematically
has MICs for S. maltophilia lower than the breakpoint
usually considered for aztreonam alone against other gramnegative
bacteria, according to previous studies (1, 5, 6).
Otherwise, some studies have indicated that in vitro susceptibility
tests for S. maltophilia are problematic and that the
* Corresponding author. Mailing address: Department of Microbiology,
Hospital Universitario de Salamanca, Paseo de San Vicente 108,
37007 Salamanca, Spain. Phone: 34 23 264825. Fax: 34 23 262261.
E-mail: jagarrod@gugu.usal.es.
2612
dilution methods currently recommended by the National
Committee for Clinical Laboratory Standards (NCCLS) (9)
can lead to erroneous results (1, 11, 13, 17). We have tested the
in vitro inhibitory and bactericidal activities of several �-lactam–�-lactamase
inhibitor combinations against S. maltophilia
strains for which the MICs of several �-lactams and �-lactam–
�-lactamase inhibitor combinations have been shown by various
methods to be high to determine the correlation between
standard sensitivity methods when testing these combinations
against this microorganism. Moreover, we have tested the combinations
that showed the best behaviors against different inocula
to determine the influence of the inoculum on the efficacies
of these combinations.
MATERIALS AND METHODS
Microorganisms. Thirty-two epidemiologically unrelated clinical strains of
S. maltophilia from different sources were used. The MICs of penicillins and
commercially available �-lactam–�-lactamase inhibitor combinations for the S.
maltophilia strains were higher than the breakpoints usually recommended for
Pseudomonas aeruginosa (10) by the broth microdilution method (PASCO) Escherichia
coli ATCC 25922 and P. aeruginosa ATCC 27853 were used as control
strains.
Antibiotics. Ampicillin, amoxicillin, ticarcillin, piperacillin, aztreonam, clavulanic
acid, sulbactam, and tazobactam were kindly provided as standard powders
by their respective manufacturers. Amoxicillin plus clavulanic acid (A-C) and
ampicillin plus sulbactam (A-S) were combined at a penicillin:inhibitor ratio of
2:1, ticarcillin plus clavulanic acid (T-C) were combined by using a fixed clavulanate
concentration of 2 �g/ml, and piperacillin plus tazobactam (P-T) were
combined by using a fixed tazobactam concentration of 4 �g/ml. AZT and
Downloaded from
http://aac.asm.org/
on January 19, 2013 by guest

VOL. 41, 1997 �-LACTAM–INHIBITOR COMBINATIONS AGAINST S. MALTOPHILIA 2613
TABLE 1. Proportion of S. maltophilia isolates tested resistant
to �-lactams and �-lactam–�-lactamase inhibitor
combinations by three methods a
Antibiotics
% Resistant by the following method:
Agar diffusion Agar dilution Broth microdilution
Ampicillin 100 100 100
Amoxicillin 100 100 100
Ticarcillin 100 100 100
Piperacillin 100 100 100
AZT 100 100 100
A-C 100 100 100
A-S 100 100 100
T-C 75 100 100
P-T 56.3 100 100
AZT-C 0 0 0
a Resistance was defined according to the NCCLS breakpoints for P. aeruginosa.
For AZT-C, NCCLS breakpoints for AZT and P. aeruginosa were used.
clavulanic acid were combined at a �-lactam:inhibitor ratio of 2:1; this has
previously been shown to be the most active ratio against this species (5).
Combinations of penicillins (amoxicillin, ticarcillin, and piperacillin) plus aztreonam
and clavulanic acid were tested at penicillin:monobactam:inhibitor concentration
ratios of 2:2:1. Fixed concentrations of clavulanate were not used in these
triple combinations, because ratios of AZT-C of about 2:1 are necessary to
maintain the synergism of AZT-C according to previous results (5).
Susceptibility testing. Agar diffusion, agar dilution, and broth microdilution
studies were performed according to NCCLS standards (9, 10). The agar diffusion
studies with all antibiotics except AZT-C was performed with commercially
available disks (Oxoid, Basingstoke, United Kingdom). Since disks with AZT-C
are not commercially available, we added disks containing 10 �g of AZT with 2
�l of a solution of clavulanic acid at 2.5 g/liter (final amount, 5 �g per disk). Agar
dilution and broth microdilution studies were performed for all antibiotics individually
and the combinations A-C, A-S, T-C, P-T, AZT-C, amoxicillin-clavulanic
acid-AZT (A-C-AZT), ticarcillin-clavulanic acid-AZT (T-C-AZT), and
piperacillin-clavulanic acid-AZT (P-C-AZT). Time-kill curve studies were performed
as described by NCCLS (8) with a high inoculum (10 8 CFU/ml), peak
concentrations of antibiotics in serum (32 �g of ticarcillin per ml for T-C and
64 �g of piperacillin per ml for P-T), and fixed concentrations of inhibitors (2 mg
of clavulanate per liter and 4 mg of tazobactam per liter) for all strains for which
discrepant results were obtained between these methods. We considered the
results to be discrepant when the MICs obtained by the broth microdilution and
agar dilution methods were different by more than two twofold concentrations,
and when the MICs obtained by any of the dilution methods were higher than the
breakpoints usually recommended for P. aeruginosa, but when the diameters
obtained by the agar diffusion method were wider than those recommended as
being the minimum for consideration of the microorganisms as sensitive, or vice
versa. For AZT-C, the breakpoints recommended for aztreonam alone against
P. aeruginosa were used. Once our results indicated the methods that were the
most reliable for testing �-lactam–�-lactamase inhibitor combinations against
S. maltophilia and the most active combinations, we tested the inhibitory and
bactericidal activities of these combinations against S. maltophilia by the broth
microdilution method with different inocula (the standard inoculum recommended
by NCCLS and inocula 10 and 50 times the standard inoculum) to
determine the influence of the inoculum size on the activities of these combinations.
RESULTS
There was a correlation between the results obtained by the
three methods used to test ampicillin, amoxicillin, ticarcillin,
piperacillin, AZT, A-C, and A-S for their inhibitory activities
against all strains. The MICs for the 32 strains tested were
higher than the breakpoints previously established for all these
antibiotics and combinations by the three methods. By the agar
diffusion method, 8 strains (25%) had diameters of �15 mm
with T-C and 14 strains (43.7%) had diameters of �18 mm
with P-T (6 strains were included in both groups). The MICs
for all of these strains were higher than the breakpoints established
by the agar dilution and broth microdilution methods
(Table 1).
The combination AZT-C showed a correlation for all
strains, and for all of the strains the MICs were lower than the
breakpoint defined for AZT against P. aeruginosa (MIC at
which 50% of isolates are inhibited [MIC 50], 4 mg/liter; MIC 90,
8 mg/liter; MIC range, 1 to 16 mg/liter), and the diameters
were wider than the minimum recommended diameters required
to consider P. aeruginosa sensitive to AZT.
To assess the differences observed between the diffusion and
the dilution methods, we performed eight time-kill studies with
six strains (two strains with diameters of �15 mm with T-C by
the agar diffusion method, two strains with diameters of �18
mm with P-T, and two strains with both characteristics) (Fig.
1). Time-kill studies confirmed the resistance of these strains to
both combinations and then confirmed the results of the dilution
methods (Fig. 1).
The triple combination P-C-AZT had the same activity as
AZT-C alone (MIC 50, 4�g/ml; MIC 90, 8�g/ml; MIC range, 1
to 16 �g/ml) by both the broth microdilution and the agar
dilution methods. The activity of the triple combination T-C-
AZT was two- to fourfold higher than that of AZT-C (MIC 50s,
1 �g/ml by the agar dilution method and 2 �g/ml by the broth
microdilution method; MIC 90s, 4 �g/ml by both methods; MIC
range, 0.5 to 4 �g/ml by the agar dilution method and 1 to
4 �g/ml by the broth microdilution method).
Since the MICs of AZT-C for S. maltophilia strains were low
but the MICs of all other �-lactam–�-lactamase inhibitor com-
FIG. 1. (A) Time-kill curves for T-C against four strains with diameters of
�15 mm by the agar diffusion method and resistant by the broth microdilution
method. ■, strain 1; }, strain 2; å, strain 3; F, strain 4. (B) Time-kill curves for
P-T against four strains with diameters of �18 mm by the agar diffusion method
and resistant by the broth microdilution method. ■, strain 3; }, strain 4; å, strain
5; F, strain 6.
Downloaded from
http://aac.asm.org/
on January 19, 2013 by guest

2614 MUÑOZ BELLIDO ET AL. ANTIMICROB. AGENTS CHEMOTHER.
TABLE 2. Inhibitory and bactericidal concentrations of AZT-C and
AZT-C-T against S. maltophilia at the standard inoculum
and 10 and 50 times the standard inoculum
Antibiotics and
inoculum a
binations were high and since ticarcillin is the only penicillin
that increased the activity of this combination, we tested the
MICs and minimum bactericidal concentrations (MBCs) by
the agar microdilution method with the standard inoculum and
with inocula 10 and 50 times the standard inoculum. MICs
were similar for all three inocula for all strains (Table 2). The
MIC for the highest inoculum was not higher than twofold the
MIC for the standard inoculum of any strain. The behaviors of
the MBCs were different. With the standard inoculum, MBCs
were between the MIC and fourfold the MIC for both combinations
for all strains. With an inoculum that was 10 times the
standard inoculum, the behavior was heterogeneous. For 10%
of the strains the MBC was the same as the MIC, for 45% of
the strains the MBC was between two- and fourfold the MIC,
and for 45% of the strains the MBC was at least eightfold the
MIC. For 9% of the strains the MBC was �32-fold the MIC
(Table 2). With an inoculum that was 50 times the standard
inoculum, bacterial killing required antibiotic concentrations
much higher than the MICs for all strains. The MBCs were
32-fold the MICs for 30% of strains and �32-fold the MICs for
70% of them.
DISCUSSION
MIC (�g/ml)/MBC (�g/ml)
50% 90% Range
AZT-C
Standard inoculum (1 � 10 6 ) 4/8 16/32 1–16/2–32
10� (1 � 10 7 ) 4/16 16/64 1–16/2–64
50� (5 � 10 7 ) 8/�128 16/�128 2–32/128–�128
AZT-C-T
Standard inoculum (1 � 10 6 ) 2/4 4/16 1–4/2–16
10� (1 � 10 7 ) 2/8 4/16 1–4/2–32
50� (5 � 10 7 ) 4/�128 4/�128 2–8/128–�128
a 10� and 50�, 10 and 50 times the standard inoculum, respectively.
The results obtained in the present study confirm previously
published data with respect to the usefulness of some methods
of testing the susceptibility of this microorganism (1, 11, 13,
17). Agar diffusion seems to lead to false results of sensitivity
to �-lactam–�-lactamase inhibitor combinations with some frequency.
The results of the agar dilution and broth dilution
methods correlate perfectly with those of time-kill studies.
Other interesting data are those related to the activities of
AZT-C and T-C-AZT against S. maltophilia. This microorganism
usually harbors two types of �-lactamases: L1, a metallo-
�-lactamase that hydrolyzes all �-lactams except aztreonam
and that is not inhibited by clavulanic acid, and L2, a Bush
group 2e �-lactamase that hydrolyzes aztreonam but that is
inhibited by clavulanic acid (2). More recently, another group
of non-metallo-�-lactamases has been described in S. maltophilia;
the pIs of these �-lactamases range from 5.2 to 6.6, and
in general they hydrolyze all �-lactams except imipenem and
aztreonam and are not inhibited by clavulanic acid (12). This
profile of �-lactamase production can explain the behaviors of
the S. maltophilia strains tested. S. maltophilia strains harboring
these �-lactamases can hydrolyze almost all �-lactams and
�-lactam–�-lactamase inhibitor combinations with the exception
of AZT-C, since the only �-lactamase from S. maltophilia
that hydrolyzes AZT (L2) is inhibited by clavulanic acid. The
fact that the high level of activity of the combination AZT-C is,
in vitro, much pronounced at ratios of about 2:1, as has been
reported previously (3), remains unexplained. The activity of
T-C-AZT is higher than that of AZT-C in strains for which
ticarcillin and T-C MICs are high, and this is also difficult to
explain. Nevertheless, synergy between ticarcillin and T-C and
other antibiotics (co-trimoxazole and ciprofloxacin) against
S. maltophilia has already been described (13).
According to our results, the in vitro sensitivity of S. maltophilia
to �-lactam–�-lactamase inhibitor combinations by the
agar diffusion method should be considered with caution and
the results should be confirmed by other methods. As we reported
previously (5), the combination AZT-C (2:1) is, in our
experience, active against all S. maltophilia strains tested, including
strains for which the MICs of all other �-lactam–�lactamase
inhibitor combinations tested are high. Other investigators
(1) have reported that 85% of S. maltophilia strains are
sensitive to the combination AZT-C (1:1). This activity is even
increased by the addition of ticarcillin, although not by the
addition of amoxicillin, ampicillin, or piperacillin. The clinical
usefulness may be impaired by the fact that a proportion of
AZT-C of about 2:1 is important for the activity of the combination,
and the pharmacokinetics of AZT and clavulanic acid
are quite different.
Both AZT-C and T-C-AZT had good bactericidal activities
against all strains tested except when very large inocula were
used. Then, if pharmacokinetic profiles that allow for the proportions
necessary for synergy are reached, monobactam–�lactamase
inhibitor combinations and, eventually, T-C-monobactam
combinations might be good alternatives for the
treatment of infections caused by multidrug-resistant strains of
S. maltophilia.
REFERENCES
1. Arpi, M., M. A. Victor, Y. Mortensen, A. Gottschau, and B. Bruun. 1996. In
vitro susceptibility of 124 Xanthomonas maltophilia (Stenotrophomonas maltophilia)
isolates: comparison of the agar dilution method with the E-test and
two agar diffusion methods. APMIS 104:108–114.
2. Bush, K., G. A. Jacoby, and A. A. Medeiros. 1995. A functional classification
scheme for �-lactamases and its correlation with molecular structure. Antimicrob.
Agents Chemother. 39:1211–1233.
3. Cullmann, W. 1991. Antibiotic susceptibility and outer membrane proteins
of clinical Xanthomonas maltophilia isolates. Chemotherapy (Basel) 37:246–
250.
4. Elting, L. S., N. Khardori, G. P. Bodey, and F. Fainstein. 1990. Nosocomial
infection caused by Xanthomonas maltophilia: a case-control study of predisposing
factors. Infect. Control Hosp. Epidemiol. 11:134–148.
5. García-Rodríguez, J. A., J. E. García Sánchez, M. I. García García, E.
García Sánchez, and J. L. Muñoz Bellido. 1991. Antibiotic susceptibility
profile of Xanthomonas maltophilia: in vitro activity of �-lactam/�-lactamase
inhibitor combinations. Diagn. Microbiol. Infect. Dis. 14:239–243.
6. García-Rodríguez, J. A., J. E. García Sánchez, J. L. Muñoz Bellido, M. I.
García-García, and E. García Sánchez. 1991. Kinetics of antimicrobial activity
of aztreonam/clavulanic acid (2:1) against Xanthomonas maltophilia. J.
Antimicrob. Chemother. 27:552–554.
7. Liu, P. I., Y. J. Lau, B. S. Hu, J. M. Shyr, Z. Y. Shi, W. S. Tsai, Y. H. Lin, and
C. Y. Tsen. 1995. Comparison of susceptibility to extended-spectrum �lactam
antibiotics and ciprofloxacin among gram-negative bacilli isolated
from intensive care units. Diagn. Microbiol. Infect. Dis. 22:285–291.
8. National Committee for Clinical Laboratory Standards. 1992. Document
M26-T. Methods for determining bactericidal activity of antimicrobial
agents. Tentative guideline, vol. 12, no. 19. National Committee for Clinical
Laboratory Standards, Villanova, Pa.
9. National Committee for Clinical Laboratory Standards. 1993. Document
M2-A5. Performance standards for antimicrobial disk susceptibility tests, 5th
ed. Approved standard, vol. 13, no. 24. National Committee for Clinical
Laboratory Standards, Villanova, Pa.
10. National Committee for Clinical Laboratory Standards. 1993. Document
M7-A3. Methods for dilution antimicrobial susceptibility tests for bacteria
that grow aerobically. 3rd ed. Approved standard, vol. 13, no. 25. National
Committee for Clinical Laboratory Standards, Villanova, Pa.
11. Pankuch, G. A., M. R. Jacobs, and P. C. Appelbaum. 1994. Susceptibilities of
123 Xanthomonas maltophilia strains to clinafloxacin, PD 131628, PD
Downloaded from
http://aac.asm.org/
on January 19, 2013 by guest

VOL. 41, 1997 �-LACTAM–INHIBITOR COMBINATIONS AGAINST S. MALTOPHILIA 2615
138612, PD 140248, ciprofloxacin, and ofloxacin. Antimicrob Agents Chemother.
38:369–370.
12. Paton, R., R. S. Miles, and S. G. Amyes. 1994. Biochemical properties of
inducible �-lactamases produced from Xanthomonas maltophilia. Antimicrob.
Agents Chemother. 38:2143–2149.
13. Poulos, C. D., S. O. Matsumura, B. M. Willey, D. E. Low, and A. McGeer.
1995. In vitro activities of antimicrobial combinations against Stenotrophomonas
(Xanthomonas) maltophilia. Antimicrob. Agents Chemother. 39:2220–
2223.
14. Smit, W. J., A. L. Boquest, J. E. Gedds, and F. A. Tosolini. 1994. The
antibiotic susceptibilities of Xanthomonas maltophilia and their relation to
clinical management. Pathology 26:321–324.
15. Vartivarian, S., E. Anaissie, G. Bodey, H. Sprigg, and K. Rolston. 1994. A changing
pattern of susceptibility of Xanthomonas maltophilia to antimicrobial
agents: implications for therapy. Antimicrob. Agents Chemother. 38:624–627.
16. Yamazaki, E., J. Ishii, K. Sato, and T. Nakae. 1989. The barrier function of
the outer membrane of Pseudomonas maltophilia in the diffusion of saccharides
and �-lactam antibiotics. FEMS Microbiol. Lett. 60:85–88.
17. Yao, J. D., M. Louie, L. Louie, J. Goodfellow, and A. E. Simor. 1995. Comparison
of E-test and agar dilution for antimicrobial susceptibility testing of
Stenotrophomonas (Xanthomonas) maltophilia. J. Clin. Microbiol. 33:1428–
1430.
Downloaded from
http://aac.asm.org/
on January 19, 2013 by guest