Susceptibility trends of Bacteroides fragilis group isolates from Buenos Aires, Argentina
L. Fernández Canigia1*, L. Castello1, A. Di Martino1, G. Greco1, M. C. Legaria1, M. Litterio1, S. C. Predari1, R. Rollet1, A. Rossetti1, G. Carloni1, M. I. Sarchi2, H. Bianchini1
1 Subcomisión de Bacterias Anaerobias (SADEBAC), Asociación Argentina de Microbiología, Deán Funes 472/76
(C1214AAD) Ciudad Autónoma de Buenos Aires;
2 Cátedra de Matemática, Facultad de Farmacia y Bioquímica,
Universidad de Buenos Aires, Ciudad Autónoma de Buenos Aires. Argentina.
*Correspondencia. E-mail:lfcanigia@labdyl.com.ar
ABSTRACT
The aim of this study was to analyze the susceptibility trends to seven antibiotics of Bacteroides fragilis group isolates based on three survey studies performed by the Committee of Anaerobic Bacteria between 1989 and 2002. Fifty three, 82 and 65 B. fragilis group isolates were collected during each period. The antimicrobial agents included were: ampicillin, ampicillin-sulbactam (2:1), cefoxitin, piperacillin, imipenem, clindamycin, and metronidazole. Minimal inhibitory concentrations (MICs) were determined according to the reference agar dilution method described by the Clinical and Laboratory Standards Institute (CLSI, formerly NCCLS). The most active antibiotics for B. fragilis and non- B. fragilis species throughout the three periods were: imipenem with 99.1 and 100% of activity, respectively, and metronidazole with 100% of activity. The susceptibility to ampicillin-sulbactam showed a decrease, from 100% to 90.3% and to 82.4 % in the last period, for both B. fragilis and non-B. fragilis species, respectively. The overall susceptibility rates for cefoxitin, piperacillin, and clindamycin were significantly different between B. fragilis and non-B. fragilis species (84.2% vs. 56.5%; 85.9% vs. 66.7% and 88.8% vs. 64.7%, respectively, p< 0.05). Cefoxitin was the antibiotic that showed more variations as regards periods and species. The susceptibility rates for clindamycin were low, about 60%, for non-B. fragilis species during the last two periods. The variations observed in the susceptibility patterns of the B. fragilis group isolates emphasize the need to continue monitoring the emergence of resistance in order to guide the election of the most appropriate antibiotic therapy scheme for anaerobic infections.
Key words: Bacteroides fragilis group; Susceptibility; Antianaerobic drugs
RESUMEN
Tendencias en el perfil de sensibilidad de aislamientos del grupo Bacteroides fragilis obtenidos en Buenos Aires, Argentina. El objetivo de este estudio fue evaluar las variaciones en el perfil de sensibilidad frente a siete antimicrobianos de aislamientos del grupo Bacteroides fragilis, mediante el análisis de tres relevamientos realizados por la Subcomisión de Bacterias Anaerobias de la Asociación Argentina de Microbiología (años 1989-1991, 1996-1998 y 1999-2002). En los citados períodos se recolectaron 53, 82 y 65 aislamientos del grupo B. fragilis. Se evaluó la actividad de: ampicilina, ampicilina-sulbactama (2:1), cefoxitina, piperacilina, imipenem, clindamicina y metronidazol. La concentración inhibitoria mínima (CIM) se determinó utilizando el método de dilución en agar, según las normas del Clinical and Laboratory Standards Institute (CLSI, anteriormente NCCLS). En los tres períodos considerados, los antibióticos más activos frente a aislamientos de la especie B. fragilis como así también frente a aislamientos pertenecientes a otras especies del grupo B. fragilis fueron imipenem, con 99,1 y 100% de actividad, respectivamente, y metronidazol, con 100% de actividad. Con ampicilina-sulbactama se observó a lo largo del tiempo una disminución de la sensibilidad, desde el 100% en el primer período hasta un 90,3 y 82,4% en el último, para B. fragilis y para especies del grupo distintas de B. fragilis, respectivamente. Cuando se consideraron los tres períodos juntos, se observaron diferencias significativas entre la especie B. fragilis y los restantes aislamientos del grupo para cefoxitina, piperacilina y clindamicina (84,2% vs. 56,5%; 85,9% vs. 66,7% and 88,8% vs. 64,7%, respectivamente, p< 0.05). Cefoxitina fue el antibiótico que mostró mayores variaciones a través del tiempo y entre especies. Las tasas de sensibilidad a clindamicina fueron bajas (alrededor del 60%) entre los aislamientos no pertenecientes a la especie B. fragilis durante los últimos dos períodos. Las variaciones observadas en los perfiles de sensibilidad del grupo B. fragilis muestran la necesidad de vigilar periódicamente la emergencia de resistencia a los antimicrobianos, a fin de orientar el tratamiento de las infecciones por bacterias anaerobias.
Palabras clave: Grupo Bacteroides fragilis; Sensibilidad; Antianaeróbicos
Bacteroides fragilis group constitutes the dominant
anaerobic bacteria in the normal intestinal microflora, most
frequently isolated in clinical infections and having the
highest resistance rates to antimicrobial agents. However,
B. fragilis is more susceptible to most antimicrobial agents
than the other species within the group (5).
Most anaerobic infections are treated empirically
based on susceptibility patterns reported in the literature.
Studies carried out for the last ten years have demonstrated
an association between antibiotic resistant B.
fragilis and adverse outcome (11, 14). Thus, the increasing
antimicrobial resistance among B. fragilis group should be considered in the selection of empirical antimicrobial
therapy.
Currently available antibiotics with activity against
these organisms include clindamycin, metronidazole, blactams
and b-lactam/b-lactamase inhibitor combinations.
However, increased resistance to these agents has frequently
been reported in recent years, e.g. the emerging
resistance to metronidazole in several countries, the high
prevalence of clindamycin resistance, and the isolation
of B. fragilis group strains resistant to carbapenems and
to b-lactam/b-lactamase inhibitor combinations (5).
Therefore, there is an ongoing need for documentation
of the changing susceptibility patterns of these microorganisms.
In the present study, we analyze the susceptibility
trends of B. fragilis group strains isolated between
1989 and 2002 to seven antibiotics usually used in
Argentina.
The Committee of Anaerobic Bacteria of SADEBAC -
Asociación Argentina de Microbiología, coordinated three
survey studies on antimicrobial resistance in anaerobic
bacteria during the following periods: 1st period, 1989 to
1991 (1); 2nd period, 1996 to 1998 (7); and 3rd period,
1999 to 2002 (6). Fifty three, 82 and 65 non-duplicated
clinical isolates of B. fragilis group were collected during
each period. The isolates were obtained from seven medical
centers in Ciudad Autónoma de Buenos Aires: Centro
de Educación Médica e Investigaciones Clínicas Dr.
Norberto Quirno (CEMIC), Hospital de Agudos Enrique
Tornú, Hospital de Enfermedades Infecciosas Francisco
J. Muñiz, Hospital Italiano, Hospital Nacional de Pediatría
Juan P. Garrahan, Instituto de Investigaciones Médicas
Alfredo Lanari, and Sanatorio Mitre. The species distribution
in each period is shown in Table 1. Isolates were identified
according to conventional biochemical methods.
Table 1. Distribution of isolates for susceptibility studies during the three periods.
The following seven antimicrobial agents were included
in this study: ampicillin, ampicillin-sulbactam (2:1),
cefoxitin, piperacillin, imipenem, clindamycin, and metronidazole.
Minimal inhibitory concentrations (MICs) were
determined according to the reference agar dilution
method described by the Clinical and Laboratory Standards
Institute (CLSI, formerly NCCLS) (8-10). Briefly, agar
dilution test plates were inoculated with 105 CFU/spot using
a Steers multipoint replicator, and incubated at 37 °C
during 48 h in jars under anaerobic atmosphere generated
by commercially envelopes (GenBox, bioMérieux,
Marcy l´Étoile, France).
MIC was defined as the lowest concentration of antibacterial
agent that inhibited visible growth. The following
CLSI recommended susceptibility breakpoint values
(µg/ml) were used (11): ampicillin £1, ampicillin/sulbactam
£8/4, piperacillin £32, imipenem £4, cefoxitin £16, metronidazole
£8, and clindamycin £2.
Susceptibility rates were compared using the chisquare
test, Fisher exact test or the maximum probability
method.
Table 2 shows the susceptibility pattern analysis of B.
fragilis group against 7 antibiotics through the three mentioned
periods. Trends were analyzed comparing B.
fragilis and non-B. fragilis species susceptibility rates for
each period.
Table 2. Susceptibility patterns of Bacteroides fragilis group isolates through the three periods.
All isolates (n=200) were susceptible to metronidazole,
which turned out to be the most active antimicrobial agent
tested. Sporadic reports of metronidazole resistant B.
fragilis group have been published since 1978. However,
in recent years some cases of clinical failures were reported.
Fortunately, these resistances are still rare (13).
Although imipenem resistance due to metallo-b-lactamases
has first been reported in 1986 by Cuchural et al. (4), it does not seem to be increasing in the world
(3, 15). In this study, only in the first period we found one
strain with imipenem MIC = 8 µg/ml; while during the
followings, MICs values were 2 µg/ml or below. Interestingly,
the strain which showed imipenem MIC = 8 µg/ml
was considered susceptible because at that time, the
imipenem susceptible breakpoint was £8 µg/ml (8). Nowadays,
this susceptibility value would be considered intermediate.
Carbapenems resistance is encoded by the cfiA gene, also called ccrA gene, expressing a class B metallo b-
lactamase which confers resistance to all b-lactam antibiotics.
A small percentage of B. fragilis strains carrying
the cfiA gene, expresses the protein at a high enough
level to classify the strain as resistant (12). Strains with
intermediate level resistance to carbapenems, including
some susceptible strains, could reach high level expression
of this enzyme as a consequence of in vitro selection
with imipenem. Thus, patients infected with these strains,
would need antibiotic treatment other than b-lactams.
Imipenem is frequently used in Argentina for severely ill
patients. Although we have not found resistant strains to
imipenem during the last two periods, we suggest testing
imipenem activity on isolates from these patients.
Ampicillin-sulbactam was one of the most active blactams
studied. There has been a worrying increase in
the number of strains displaying decreased susceptibility to ampicillin-sulbactam, from 100% to 90.3% (p=0.08) and
to 82.4% (p > 0.05) in the last period, for both B. fragilis and non-B. fragilis species, respectively. This trend had
previously been observed by Bianchini H. (2). These rates
are somewhat higher than those shown by Snydman et
al. (15), who reported an overall resistance of 2.4%, and
observed a decrease from 3.6% to 1.7%, between 1997
and 2000. However, ampicillin-sulbactam and other blactams
with b-lactamase inhibitors as piperacillintazobactam
continue to be the most active b-lactams following
the carbapenems (3, 5, 15). In the last years, a
100% activity of piperacillin-tazobactam against B. fragilis group was reported, as it had been observed in our last
survey study (6).
The overall susceptibility rates for cefoxitin, piperacillin,
and clindamycin were significantly different between B.
fragilis and non-B. fragilis species (84.2% vs. 56.5%;
85.9% vs. 66.7% and 88.8% vs. 64.7%, respectively, p<
0.05).
Cefoxitin was the most variable antibiotic with respect
to periods and species. There was a significant susceptibility
decrease between 1989-1991, and 1996-1998 periods
for both groups, B. fragilis (95% vs. 75.6%, p < 0.05)
and non-B. fragilis species (92.9% vs. 40.5%, p < 0.05).
Although susceptibility rates showed to be lower in the
third period than in the first one, they were higher with
respect to the second period, even though they did not
demonstrate significant differences in the B. fragilis isolates
(Table 2). Furthermore, when comparing B. fragilis
with non-B. fragilis species, the susceptibility rates were
significantly higher for B. fragilis in the second and third
periods (p < 0.05) (Figure 1A).
Figure 1. B. fragilis;
non-B. fragilis species.
Variations in susceptibility percentage at specified breakpoints
for B. fragilis and non B. fragilis species over three periods for Acefoxitin
and B- clindamycin.
The overall rate of susceptibility to clindamycin (88.8%)
observed for B. fragilis was higher than that observed in
other studies (~ 67%) (3, 16) but similar to others (84%)
(15). Our data of susceptibilities were low, about 60%, for
non-B. fragilis species during the last two periods. The
reports in the literature are quite variable for the different
species of non-B. fragilis with values ranging from 25%
to 80% (3, 15, 16). Similar to cefoxitin, non-B. fragilis species showed susceptibility rates significantly lower than B. fragilis isolates comparing the second and third periods
(Figure 1B).
There was a trend to higher MIC values among the
antimicrobial agents tested as shown in Table 2. The exceptions
were imipenem and metronidazole. Nevertheless,
the difference on the susceptibility rates among periods
was not always significant. If this resistance increase
is due to overuse of antibiotics in our country, it is difficult
to prove.
The variations in susceptibility patterns observed during
these 14 years, emphasize the need to continue monitoring
the emergence of resistance against the most frequently
antianaerobic drugs used. This would provide information
to guide empirical treatment, although the results
in one region or medical center, might not be applicable
to another. Also, because susceptibility rates may
vary among members of the B. fragilis group, it would be
necessary to perform, routine identification to species level
in clinical laboratories.
REFERENCES
1. Bianchini H, Berenstein P, Notario R, Litterio M, Greco G,
Hardie N, et al. Estudio multicéntrico de la sensibilidad de
bacterias anaerobias en la Argentina. Infect Microbiol Clín
1994; 6: 7-16.
2. Bianchini H. Methods for susceptibility testing in anaerobes:
when and how they should be used. Anaerobe 1999; 5: 417-20.
3. Betriu C, Culebras E, Gómez M, Rodríguez-Avial I, Picazo
JJ. In vitro activity of tigecycline against Bacteroides species.
J Antimicrob Chemother 2005; 56: 349-52.
4. Cuchural GJ Jr, Malamy MH, Tally FP. Beta-lactamasemediated
imipenem resistance in Bacteroides fragilis.
Antimicrob Agents Chemother 1986; 30: 645-8.
5. Hecht DW. Anaerobes: antibiotic resistance. Clinical
significance, and the role of susceptibility testing. Anaerobe
2006; 12: 115-21.
6. Litterio M, Bianchini H, Carloni G, Di Martino A, Fernández
Canigia L, Greco G, et al. Actividad «in vitro» de 10
antimicrobianos frente a bacterias anaerobias. Estudio
multicéntrico 1999-2002. Rev Argent Microbiol 2004; 36:
130-5.
7. Litterio M, Castello L, Rollet R, Di Martino A, Greco G,
Fernández Canigia L, et al. Susceptibility of anaerobic gramnegative
rods to trovafloxacin. Drugs 1999; 58 Suppl 2:
193-4.
8. National Committee for Clinical Laboratory Standards.
Methods for antimicrobial susceptibility testing of anaerobic
bacteria; 3rd Approved standard, 1993; M11-A3. Villanova,
Pa, USA.
9. National Committee for Clinical Laboratory Standards.
Methods for antimicrobial susceptibility testing of anaerobic
bacteria; 4th Approved standard, 1997; M11-A4. Villanova,
Pa, USA.
10. National Committee for Clinical Laboratory Standards.
Methods for antimicrobial susceptibility testing of anaerobic
bacteria; 5th Approved standard, 2001; M11-A5. Wayne,
Pa, USA.
11. Nguyen MH, Yu VL, Morris AJ, McDermott L, Wagener MW,
Harrell L, et al. Antimicrobial resistance and clinical outcome
of Bacteroides bacteremia: findings of a multicenter
prospective observational trial. Clin Infect Dis 2000; 30: 870-
6.
12. Podglajen I, Breuil J, Casin I, Collatz E. Genotypic
identification of two groups within the species Bacteroides
fragilis by ribotyping and by analysis of PCR-generated
fragment patterns and insertion sequence content. J
Bacteriol 1995; 177: 5270-5.
13. Rotimi VO, Khoursheed M, Brazier JS, Jamal WY,
Khodakhast FB. Bacteroides species highly resistant to
metronidazole: an emerging clinical problem? Clin Microbiol
Infect 1999; 5: 166-9.
14. Salonen JH, Eerola E, Meurman O. Clinical significance
and outcome of anaerobic bacteremia. Clin Infect Dis 1997;
26: 1413-7.
15. Snydman DR, Jacobus NV, McDermott LA, Ruthazer R,
Goldstein EJ, Finegold SM, et al. National survey on the
susceptibility of Bacteroides fragilis group: report and
analysis of trends for 1997-2000. Clin Infect Dis 2002; 35:
S126-34.
16. Teng L-J, Hsueh P-R, Tsai J-C, Liaw S-J, Ho S-W, Luh KT.
High incidence of cefoxitin and clindamycin resistance
among anaerobes in Taiwan. Antimicrob Agents Chemother
2002; 46: 2908-13.
Recibido: 29/05/07
Aceptado: 07/08/07