Resistance of anaerobic bacteria Can we find any clinical impact? Szeged ESCMID School 2005 L. Dubreuil Faculté de Pharmacie Université de Lille II, France
Antibiotic resistance among anaerobes Intrinsic resistance all anaerobes aminoglycosides, quinolones, fosfomycin,trimethoprim,aztreonam species -dependant : metronidazole : Propionibacterium & Actinomyces rifampicin : F. necrophorum, F. mortiferum cephalosporins : C. difficile cefotetan & vancomycin : C. innocuum Bacteroides fragilis group : C1G, aminopenicillins Fusobacterium : macrolides Acquired resistance
Antibiotic resistance among anaerobes Acquired resistance : ß-lactams : ß -lactamase Clostridia (butyricum, + RIC group) Prevotella, Fusobacterium B. fragilis gene + promotor cfia carbapenemase, cfxa Hyperproduction of the chromosomal enzyme lack of porin & PBP modifications
Antibiotic resistance among anaerobes Acquired resistance : MLSb erm genes clindamycin Metronidazole nim genes Fluroquinolones gyrase, topoisomerase mutations resistance to Moxifloxacin Chloramphenicol cat genes
The situation in Europe Gram positive cocci susceptible penicillins > cephalosporins glycopeptides, linezolid, rifampicin acquired resistance clindamycin > 20% metronidazole < 5% variable fluoroquinolones
The situation in Europe Actinomyces susceptible penicillins > cephalosporins glycopeptides, linezolid, rifampicin acquired resistance variable tetracycline macrolides (rare) fluoroquinolones
The situation in Europe Propionibacterium susceptible penicillins > cephalosporins fluoroquinolones glycopeptides, linezolid, rifampicin acquired resistance tetracyclin macrolides & clindamycin increasing
The situation in Europe C. perfringens susceptible penicillins > cephalosporins, cefoxitin metronidazole, glycopeptides, linezolid, rifampicin most fluoroquinolones acquired resistance tetracyclins macrolides & clindamycin rare One strain resistant to metronidazole Faris & Poxton 1999
The situation in Europe C. difficile susceptible acquired resistance penicillins, metronidazole, glycopeptides, fusidic acid, bacitracin tetracyclins, chloramphenicol macrolides & clindamycin > 60% emerging? Vancomycin & metronidazole
The situation in Europe Other clostridia (with exception), susceptible penicillins > cephalosporins glycopeptides linezolid, metronidazole acquired resistance tetracyclins macrolides & clindamycin cefoxitin & cefotetan chloramphenicol glycopettides :C. boltae,c. hathewayi (Van B2)
The situation in Europe clostridia RIC group C. butyricum : penicillinase inhibited by clavulanic acid C. clostridioforme broad cephalosporinase low-level resistance to teicoplanin, daptomycin, ramoplanin C. ramosum broad cephalosporinase not inhibited by clavulanic acid low-level resistance to vancomycin, linezolid, daptomycin C. innocuum intrinsic resistance to cefoxitin and cefotetan low-level resistance to vancomycin & daptomycin
And now? Gram positive cocci Non-sporulated gram positive bacilli Clostridia No major clinical problems except for individual patients low rate of antibiotic diffusion (brain, bone,) be aware of RIC group glycopeptide resistance in clostridia.
What about gram negative anaerobes? Prevotella 2/3 ß-lactamase production + acquired resistance clindamycin 5% metronidazole rare variable fluoroquinolones Porphyromonas Fusobacterium penicillinase < 10%
Les mousquetaires anaérobies des infections pulmonaires et O.R.L QuickTime et un décompresseur TIFF (non compressé) sont requis pour visualiser cette image.
Prevotella : MIC 50/90% ß-lactamase - ß-lactamase + Amoxicillin 0,12 0,25 8 >64 Amox +clavu 0,06 0,06 0,06 2 Cefalotin 0,25 1 16 >64 Cefuroxime 0,12 1 8 >64 Cefixime 0,25 1 16 >64 Cefpodoxime 0,12 0,5 4 >64 Ceftriaxone 0,12 0,5 4 >64
Prevotella : MIC distribution Nb of strains 25 20 Amoxycillin 15 10 5 0 MIC in mg/l
Prevotella : distribution des CMI Nb of strains 12 10 Cephalotin 8 6 4 2 0 ß-lac + ß -lac - MIC in mg/l
Prevotella : distribution des CMI 14 Nb of strains 12 Cefuroxime 10 8 6 4 2 0 0.03 0.06 0.12 0.25 0.5 1 2 4 8 16 32 64 128 >128 ß-lac + ß lac - MIC in mg/l
Nb of strains 14 Prevotella : distribution des CMI 12 Cefpodoxime 10 8 6 4 2 0 ß -lac - ß-lac + MIC in mg/l
Antibiotic susceptibility in Porphyromonas Antibiotic MIC range MIC 50% MIC90% Amoxicillin 0,03-0,5 0,5 0,5 Co-amoxiclav 0,03-0,25 0,25 0,25 Cefotaxime 0,06-1 1 1 Clindamycin 0,03-0,25 0,06 0,25 Metronidazole 0,03-0,25 0,125 0,25 Roxithromycin 0,25 Pristinamycin 0,25
FQ susceptibility of Prevotella & Fusobacterium Antibiotic Prevotella Fusobacterium MIC 50% 90% 50% 90% Ciprofloxacin 1 4 2 4 Ofloxacin 2 8 1 4 Levofloxacin 2 4 1 2 Moxifloxacin 0,12 0,5 0,12 1 Gatifloxacin 0,25 2 0,5 4 Trovafloxacin 1 2 0,25 0.5
Anaerobic lung infections. High rate of penicillin failures associated with penicillin-resistant Prevotella melaninogenica 10/47 resistant strains 5 patients received benzyl-penicillin 2 M every 4 hours all failed Failure penicillin 8/18 vs 1/19 clindamycin Gudiol Arch inter Med 1990 150 :2525
Role of ß-lactamases from gram neg anaerobes Nord-Brook ß-lactamases excreted by Prevotella or Fusobacterium caused clinical failures when patients were treated by penicillins otitis - sinusitis-dental abcess-lung abcess-recurrent tonsillitis
Anaerobic infections Bacteroides fragilis group + Clostridium : intra-abdominal gyneco-obstetrics, biliary infections Perforated appendicitis Decubitus ulcers
Last but not least the B. fragilis group Chromosomal cep A ß-lactamase AMX + C1G+ C3G+C4G + overproduction TIC + PIP + lack of porin AMC, TTC, PTZ? carbapenemase All ß-lactams MLSb Clindamycin (20-40%) nim genes Nitro-imidazole
AMX TIC PIP CTX IMP MTZ CFT CFX CTT ERY CLN AMC TCC PTZ TE CM
Resistance to the combinations of ß- lactams and ß-lactamase inhibitors IS 1224 upstream the cep A gene increases the production of the chromosomal cephalosporinase Rogers et al.j Bacteriol 1994;176:4376-4384 High level production of the cephalosporinase alone or associated with modifications in the permeability barrier (defect of porin)
Phenotypes of resistance among imipenemsusceptible strains of the B. fragilis group Antibiotic wild hyperproduction of ß-lactamase or/and lack type of porin or silent carbapenemase % (72) (19) (6) (2) (1) Ticarcillin S R R R R Coamoxiclav S S S/I I/R R Ticar + clavu S S S S I/R Pipera+tazo S S S S I Cefotetan S/I I/R R R R
Figure 1 : Antibiotic resistance rates among the B. fragilis group (256 strains) in 2003 30 Antibiotic resistance rates (%) 27,5 28 25 20 15 10 5 0 0,8 1,9 1,9 4,3 6,2 0 Metronidazole Imipenem Ticar+ clavu Pipéra +tazo Co-amoxclav Cefoxitin Ticarcillin Clindamycin
Activity of fluoroquinolones against the B. fragilis group (130 strains) 60 50 40 30 20 10 0 0.06 0.25 1 4 16 64 GAR MOX LEV CIP CIP LEV MOX GAR
Antibiotic resistance rates (%) - B. fragilis group Are antibiotic resistance rates increasing? 35 30 25 20 15 10 5 0 MTR AMC CLN 1996 1998 2000 2003 1996 1998 2000 2003 MTR 0,4 1,9 4,7 4,3 AMC 4 3,8 5,6 4,3 CLN 20 29 32 28
Metronidazole resistance nim on mobilisable plasmids nim A B.vulgatus nim C B. thetaiotaomicron nim D B. fragilis nim B chromosomal B. fragilis, P. acnes, A. odontolyticus, C. bifermentans nim A in Prevotella bivia nim G B. fragilis
B. fragilis S or Intermediate to metronidazole
B. fragilis metronidazole-r
Metronidazole resistance Gal & Brazier, JAC 2004,54:109-116 15/ 206 isolates were found to possess nim genes and these had MICs of MTZ from 1.5 to >256 mg/l with 11.6% above the therapeutic breakpoint of 16 mg/l Disc diffusion 28-30 mm test on FAA with a 5 µg MTZ disc. 70% showed reduced zone of susceptibility or no zone 30% with no reduced zone MIC 1.5 to 4 mg/l No zone for 7 strains without nim gene.
Results (2) Bacteroides fragilis group (PCR-RFLP) Strain N 17: nimb 22: nime 23: nimd 29: nimd 31: nimb 39: nima Strain N 40: nima (sequencing) Veillonella sp. Strain N 18 : nime (sequencing) M 17 22 23 29 31 39 ND E c o R V F o k I H i n f I
Results (3) MTZ /R MTZ /R MTZ /I 2 copies of nima gene in the genome of: - MTZ Resistant strain (MIC = 64 mg/l) - MTZ intermediate strain (MIC = 16 mg/l) nima SfiI nima EcoRI nima EcoRI
Clinical significance of intermediate or resistant metronidazole clostridia? Elsaghier, Brazier et al. JAC 2003 Failure of treatment due to Bacteroides fragilis with reduced susceptibility to metronidazole. Pelvic collections grew mixed coliforms susceptible to cefuroxime Patient treated by Cefuroxime +metronidazole +gentamicin. Pyrexia after four days of Tt. WBC 16000/mm 3 B. fragilis MTR-R cultured from blood cultures (disk diffusion) MIC by E test 6 mg/l.
Metronidazole resistance in France French resistant strains MTZ-R Prevotella buccae et 3 souches de P. loescheii CMI = 16 mg/l B. fragilis MIC = 64 mg/l H. Marchandin et al. Anaerobes 2002;8:137 Veillonella nim E in a metronidazole-susceptible Veillonella sp. strain MIC = 4 mg/l H. Marchandin, Dubreuil, AAC 2004 48:3207
Bacteremia caused by Prevotella sp. with reduced susceptibility to metronidazole Mory et al.ecc Paris december 2004 78 old patient with sepsis Cefotaxime (3g)+ ofloxacin (400mg) E. coli S + Streptococcus anginosus S Patient pyrexial new blood cultures Day 7 Day 16 Teicoplanin 400mg and metronidazole 500mg tid added Prevotella AMX, CTX, OFL-R, MTR S but colonies appearing later in the inhibition zone MIC 64 mg/l. persistance of fever Piperacillin + tazobactam 12g + Cipro 800 mg. Apyrexia in 48h
Metronidazole resistance B. fragilis & B. ovatus from appendicetomy MIC >32 mg/l : failure in Koweit Rotimi et al. Clin Microbiol Infect 1999 ; 5:166-169 B. fragilis MIC =256 in New Delhi with clinical failure B. fragilis MIC >32 mg/l in Seattle Chauldry, Emerging Infect Dis 2001,7,485-486 Shapiro et al. J Clin Microbiol 2004 ;42:4127-4129 B. fragilis in Hungary (Nagy et al.), Poland (Wojcik) Prevotella loescheii MIC 12 mg/l subdural empyema in Cardiff Sandoe JAC 2001;47:366-367.
Metronidazole resistance C. difficile MIC =16 mg/l in Hong Kong, Madrid and Cardiff Wong et al.diagn Microbiol Infect Dis 1999 ;34:1-6 Pelaez AAC 2002,46:1647-1650 Brazier et al. JAC 2001;48:471-472 Equine isolates in Sacramento MIC = 32 mg/l Jang et al. Clin Infect Dis 1997, S266-267 Finegoldia magna MIC >128 mg/l Cape Town Theron et al. JAC 2004; 54 :240-242 C. perfringens in Edinburgh MIC >32 mg/l Faris et al. J Infect 1999,2:164-165.
In vivo veritas Many variables in addition to in vitro susceptibility testing Site of infection Surgical procedures Type of underlying disease Presence of other pathogens (mixed infections) Type of therapy used
Correlation of clinical outcomes in patients with B. fragilis group infections treated with cefoxitin Snydman et al AAC 1992 36 540 Patients in the treatment failure group were more likely to have - a longer cefoxitin dosing interval (p=0.019) -a shorter duration of cefoxitin (p = 0.051) -a longer duration of hospitalization -a higher observed maximum leucocyte count
Plot of clinical cure or treatment failure versus MIC and cefoxitin dosing interval MIC mg/l 64 Time above the MIC cured failed 32 16 8 4 6 8 Dosing interval (h)
PK/PD concentration T > MIC MIC T > MIC Time
Antibiotic Susceptibility Testing pro or con? Wilson CID 1995 suppl 2 S251 Chow 1974 appropriate antibiotic therapy of Bacteroides bacteremia is associated with a better outcome than inappropriate therapy but Surgeons advocated (MC Namara 1993) «routine practise of obtaining cultures in patients operated on for acute and complicated appendicitis should be abandoned» «No benefit from susceptibility data» (Mosdell 1991)
AST pro or con in peritonitis? Mosdell only 8.8% of patients who had inadequate empirical therapy was an appropriate change in antibiotic treatment Dougherty «culture results appeared to influence antibiotic therapy for only 7% of these patients» No benefit from susceptibility data (Mosdell 1991) Worrisome disjunction between surgeon and microbiologist
AST pro or con in peritonitis? Worrisome disjunction between surgeon and microbiologist Berne BUT In vitro susceptibility was significantly associated with infectious postoperativc outcome «89% of patients in whom primary therapy failed had resistant strains compared with 46% in whom therapy succeeded. So... The concept persists that in patients with polymicrobial infection adequate surgical treatment and antibiotic at least cover aerobic component of the flora will suffice
Hopkins 1993 AST pro or con in peritonitis? Study of 175 patients with intra-abdominal infections «both aerobic and anaerobic bacteria susceptibility was statistically correlated with outcome» 26% of resistant B. fragilis from those who had an infection vs 3% those who have not Patients with resolution of the IA infection 44% of resistant isolates vs 82% if complications
Retrospective study of anaerobic bacteremia Blood cultures positive for anaerobes 81 patients Clinically significant bacteremia 57 patients Clinically insignificant bacteremia 24 patients Initial tt ineffective Changed to effective 32% Initial tt effective 49% 5% died 17% died Initial tt ineffective Not changed 19% 55% died Salonen CID 1998 26:1413
Antimicrobial resistance and clinical outcome of Bacteroides bacteremia Well-designed prospective study with stratification of patients 81 patients treated with an antimicrobial agent to which the isolate was susceptible and 11 with a resistant isolate. Therapy Inactive active p= Mortality 45% 16% 0.03 Clinical failure 82% 22% 0.0002 Persistence 47% 12% 0.06
Antimicrobial resistance and clinical outcome of Bacteroides bacteremia Failure 4/6 clindamycin-resistant B. fragilis (MIC 16 mg/l) 1/13 clindamycin-susceptible B. fragilis Piperacillin 3 failures = all resistant isolates MIC 256 mg/l One cefotetan-resistant B. caccae = failure
Multivariate analysis of risk factors for death and clinical failure of treatment for patients with Bacteroides bacteremia Outcome, risk factor Multivariate analysis OR ( 95%CI) Death APACHE II score > 15.0002 21 (4-104) Inactive therapy.066 5 (0.9-28) Clinical failure APACHE II score > 15.0002 16 (4-71) Inactive therapy.001 38 (4-337) Liver disease.68 5 (1-29)
Prediction of patients response to therapy Specificity 97% Positive Predictive value 82% Sensitivity 63%
Principles of antibiotic treatment in community-acquired peritonitis To be started as soon as the diagnostic is done and surgery is decided Rapid treatment IV route Drugs may be different from those used in prophylaxis Antibiotics directed towards Enterobacteriaceae and anaerobes cause of mortality and abscesses formation Bartlett JG. Med Clin North Am. 1995;79:599-617
Pathophysiology of intra-abdominal infections Role of Enterobacteriaceae and anaerobes 100 75 Mortality Abscesses observed % 50 25 0 E. coli E. faecalis B. fragilis E. coli + E. faecalis E. coli + B. fragilis E. faecalis + B. fragilis Onderdonk. Infect Immun. 1976
Pathophysiology of intra-abdominal infections % 100 Role of Enterobacteriaceae and anaerobes Mortality 75 Abscesses 50 25 0 Control untreated Gentamicin Clindamycin Clindamycin + gentamicin Weinstein. JID. 1975
INVANZ (Ertapenem) in communauty acquired peritonitis Background relative influence of the outcome surgery 1 ICU 2 antibiotic therapy 3 Numerous identical protocols demonstrating the equivalence of antibiotics (cf. Consensus)
Impact of an initial adequate antibiotic treatment?
Community-acquired peritonitis : pronostic 100 90 80 70 Success 0 10 20 30 Mortality (%) 60 50 40 40 50 60 Success (%) 30 20 10 Death 70 80 90 0 100 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000
Mortality Appropriate surgical procedure + adequate ATB 6 % Appropriate surgical procedure + inadequate ATB 71 % Inappropriate surgical procedure + adequate ATB 90 % Inappropriate surgical procedure + inadequate ATB 100 % Carlet J. L infection en réanimation. Masson Paris 1986 pp 126-138
Inadequate antimicrobial therapy in peritonitis Effect on mortality 92 patients patients with secondary peritonitis 65 patients (71%) received an inappropriate antibiotic therapy - initially - or in the course of therapy 94 % ofothe patients who died received an inappropriate therapy Koperna. Arch Surg 1996;131:180
A1 Effect of inadequate initial antibiotic therapy in peritonitis (2) Financial implications Canada (1993-2001) 8 hospitals Initial therapy Adeqate n=383 Inadequate n=51 (14 %) Ssucces of treatment %s 100 80 60 40 20 n=3 (0.9%) n=46 n=282 (85.2%) n=2 (3.9%) (13.9%) n=21 41.2%) n=28 (54.9%) Ampi + Genta + Metro Cefoxitine Genta + Metro Cefazolin + Métro Cipro + Metro Other< 5% of cases Mean hospital cost 122 88 52 31 23-5 (4 %) 14 (16 %) 9 (17 %) 6 (19 %) 6 (26 %) - + 65 % 0 Adequatre n=331 Inadequate n=51 Success failure others Mean antibiotic cost Mean biochemistry cost Mean microbiology cost + 34 % + 158 % + 107 % Chabot I. et al. IDSA 2002. P 604
Folie 65 A1 Insert white background for graph. White fonts Admin; 08.05.2003
Infections including B. fragilis? YES If efficacy against B. fragilis Most anaerobes are cured NO Consider Prevotella & Fusobacterium and the ß-lactamase production
Conclusion Need for : antibiotic susceptibility testing ß-lactamase production Detection of a carbapenemase Detection of resistance and decreased susceptibility to metronidazole Clindamycin empiric treatment on anaerobes not warranted