JAC Advance Access originally published online on August 4, 2006
Journal of Antimicrobial Chemotherapy 2006 58(4):873-877; doi:10.1093/jac/dkl310
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Longitudinal European surveillance study of antibiotic resistance of Haemophilus influenzae
Eijkman-Winkler Center for Medical Microbiology and Infectious Diseases, University Medical Center Utrecht G 04.614, Heidelberglaan 100, 3584 CX Utrecht, The Netherlands
*Corresponding author. Tel: +31-30-2503566; Fax: +31-30-2541770; E-mail: W.T.M.Jansen{at}umcutrecht.nl
Received 28 April 2006; returned 25 May 2006; revised 7 July 2006; accepted 9 July 2006
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResults and discussionTransparency declarationsReferences |
|---|
Objectives: We assessed the current resistance rates of Haemophilus influenzae against ß-lactams and other agents in Europe and compared the results with those of our previously performed surveillance study.
Methods: MICs of the antibiotics were determined using broth microdilution. The penicillin-binding domain of PBP3 of ß-lactamase (BL)-negative, amoxicillin-resistant (BLNAR) isolates was sequenced.
Results: The percentage of BL-positive and BLNAR strains ranged from 0% to 17.6% and 0% to 33.9%, respectively. Compared with 1997/98 and 2002/03, the overall percentage of strains non-susceptible to amoxicillin decreased from 19.8% and 23.3%, respectively, to 16.4% in 2004/05. The percentage of BL-producing strains decreased from 11.0% and 13.7%, respectively, to 7.6%, whereas the number of BLNAR strains remained stable (8.8% and 9.6%, respectively, versus 8.8% in 2004/05). Comparison of penicillin binding protein (PBP) 3B gene sequences between BLNAR and susceptible strains revealed novel amino acid mutations.
Conclusions: In spite of large inter-regional differences, the overall resistance of H. influenzae to amoxicillin in Europe seems to decline due to a decreasing number of BL-producing strains, whereas the overall percentage of BLNAR strains seems relatively constant.
Keywords: ß-lactams , amoxicillin , penicillin binding proteins
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionTransparency declarationsReferences |
|---|
Haemophilus influenzae is a major respiratory tract pathogen, causing acute exacerbations of chronic bronchitis, acute sinusitis and community-acquired pneumonia. Resistance of H. influenzae to ß-lactam antibiotics is generally conferred by either the presence of ß-lactamase (BL) or mutations in the transpeptidase domain of the penicillin binding protein (PBP) resulting in a lowered affinity for ß-lactam antibiotics.1 To date, BL production is regarded as the most common resistance mechanism among this species. In the early 1980s the first BL-negative, ampicillin-resistant (BLNAR) isolates were reported. In Europe the emergence of BLNAR strains has been demonstrated in many countries ranging from 2% to 20%2 and a percentage up to 40% has been reported in Japan.3 Recently, a decreasing number of BL-producing H. influenzae isolates has been noted in the United States.4
We aimed to study the current resistance rates of European H. influenzae isolates against ß-lactams and other oral agents used for the treatment of respiratory tract infections and to assess the epidemiology of amoxicillin resistance in Europe over the past 7 years by comparing the results with data obtained from our previous surveillance study.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionTransparency declarationsReferences |
|---|
Isolates
A total of 578 H. influenzae isolates were tested, prospectively collected between October 2004 and April 2005 from patients with respiratory tract infections in various university hospitals in Austria, France, Germany, Ireland, Italy, The Netherlands, Poland, Portugal, Spain, Turkey and the UK. Only one isolate per patient was included. All strains were re-identified at the Eijkman-Winkler Institute by colony morphology on chocolate agar, X and V factor requirements and haemolysis on horse blood agar. BL production was tested by the nitrocefin test (Oxoid, Basingtoke, UK).
Susceptibility testing
The antimicrobial agents tested are listed in Table 1. Microtitre plates containing freeze-dried serial dilutions of the antibiotics were prepared by Trek Diagnostics (East Grinstead, West Sussex, UK). MICs were determined by a microdilution method according to CLSI guidelines. H. influenzae ATCC 49247 and ATCC 49766 were used for quality control. BL-negative strains that were non-susceptible to amoxicillin (MIC of 
2 mg/L) were categorized as BLNAR. BLNAR strains were further subdivided into high-BLNAR (amoxicillin MIC 4 mg/L) and intermediate-BLNAR (amoxicillin MIC 2 mg/L).2
|
PCR and DNA sequencing
Mutations relevant for ß-lactam antibiotic resistance in H. influenzae are present in the transpeptidase domain of PBP3B between amino acids 320 and 540.5,6 This domain was PCR amplified with the primers F1, 5'-TAA TGC GTA ACC GTG CAA TTA C, and F2, 5'-ACC ACT AAT GCA TTA CGA GGA TC.6
The purified amplification products were sequenced using four primers. Besides primers F1 and F2 described above, the primers PBP3.seqexf, 5'-CTG GGC AGA TAT TGA GCG TG, and PBP3.seqexr, 5'-CAC GCT CAA TAT CTG CCC AG, were used to sequence the products obtained with primers F1 and F2. Mutations were determined by comparing the sequence data with the PBP3B DNA sequence with GenBank accession no. L42023 [GenBank] .
|
Results and discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionTransparency declarationsReferences |
|---|
Of the 578 strains tested 95 (16.4%) were non-susceptible to amoxicillin (Table 1). In 44 strains (7.6%) resistance was due to BL production, whereas 51 (8.8%) showed the BLNAR phenotype, comprising 15 (2.6%) high-BLNAR and 36 (6.2%) intermediate-BLNAR strains. None of the BL-positive strains was resistant to co-amoxiclav. Levofloxacin, moxifloxacin (MIC90s
0.03 mg/L) and cefixime (MIC90 0.06 mg/L) were the most active agents tested, exhibiting comparably high activities against BL-producing and BLNAR isolates. All 578 strains were susceptible to these agents. Compared with the other cephalosporins tested, cefixime was 32 times more active than cefuroxime and 128 times more active than cefaclor. The in vitro activity of telithromycin was comparable to that of azithromycin (MIC50/90 1/2 mg/L) and these two agents were 8 times more active than clarithromycin (MIC50/90 8/16 mg/L). Apart from amoxicillin, reduced susceptibility rates were observed for cefaclor, cefuroxime and clarithromycin, being most pronounced with BLNAR strains (76.5%, 86.3% and 80.4%, respectively). However, according to the CLSI guidelines BLNAR strains should be considered resistant to agents such as co-amoxiclav, cefaclor and cefuroxime despite in vitro susceptibility. In addition to clarithromycin, a small percentage of the strains was non-susceptible to telithromycin and azithromycin. Resistance against these three macrolides may be due to active efflux of the drugs, in combination with mutations in the 23S rRNA and/or ribosomal proteins.7 Considerable variation in amoxicillin resistance was noticed among strains from different European countries. The percentage of BL-positive strains ranged from 0% to 17.6%, whereas the emergence of the BLNAR phenotype of H. influenzae was demonstrated in most of the countries with a percentage ranging from 1.5% to 33.9% (Table 2).
|
Compared with 1997/98 and 2002/03, the overall percentages of strains that were non-susceptible to amoxicillin decreased from 19.8% and 23.3%, respectively, to 16.4% in 2004/05 (Table 2, P = 0.005, Pearson
2, SPSS 12.1). The overall percentage of BL-producing strains decreased from 11.0% and 13.7% to 7.6%, respectively (P = 0.001), whereas the overall percentage of BLNARs remained stable (8.8% and 9.6% versus 8.8%; high-BLNAR 2.8% and 3.1% versus 2.6%). Taking only hospitals that participated both in the 2002/03 and 2004/05 surveillance into account, this trend remains the same: the percentage of BL+ strains decreased from 14.7% (n = 380) in 2002/03 to 9.4% in 2004/05 (n = 331, P = 0.03) whereas the proportion of BLNAR remained stable, being 9.6% in 2002/03 and 10.3% in 2004/05. Altered use of BL-unstable or less potent agents such as amoxicillin or cefaclor may have influenced the selection pressure for BL-producing and BLNAR strains.
All 51 BLNARs and 14 amoxicillin-susceptible isolates (MICs 0.12–1 mg/L) were selected for DNA sequencing of the penicillin binding protein PBP3B. The amino acid substitutions in PBP3B deduced for these isolates are listed in Table 3. Based on amino acid substitutions at position 526 (Asn
Lys) and 517 (ArgHis), BLNARs were subdivided into group I and group II, respectively, according to Ubukata et al.6 A total of 86.3% belonged to group I (compared with 87.5% in 2002/2003), whereas 2.0% belonged to group II (compared with 4.2% in 2002/2003). The predominance of group I is in agreement with a recent study in the USA.4 Except for the mutations at positions 499 (AsnSer) and 511 (ValAla), all mutations have been observed in our previous study2 and French BLNAR strains as well.5 However, only a subset of these mutations may be actually associated with ampicillin resistance: compared with the susceptible control isolates, BLNARs showed more than 2-fold higher mutation rates at position 377 (susceptible 7.1% versus BLNAR 21.6%, P = 0.3, Fisher’s Exact Test, SPSS 12.1), position 490 (7.1%, 23.5%, P = 0.3), 502 (28.6%, 70.6%, P < 0.01) and 526 (21.4%, 86.3%, P < 0.001). Thus, in addition to several previously identified mutations associated with amoxicillin resistance,5,6,8,9 the significant higher mutation rate at position 502 in BLNAR strains suggests that this mutation may also contribute to amoxicillin resistance. For the group I mutation at position 526, there was a positive correlation between amoxicillin resistance levels and percentage of isolates carrying this mutation (r = 0.77, n = 6, Pearson regression analysis). In contrast to Japanese BLNAR strains,8,10 we neither observed additional Ser385 and Leu389 mutations in European BLNARs nor an enhanced number of PBP3 mutations in addition to group I and II mutations in high-BLNARs compared with intermediate-BLNARs. Our results suggest that the group I mutation itself may already be associated with high-BLNARs, as the percentage of isolates carrying a mutation at this position increased with the resistance of these isolates against amoxicillin.
|
Although BLNAR strains were highly variable in PBP3 mutations, certain mutation profiles were country specific. The 350 (Asp
Asn), 502 (AlaThr) and 526 (AsnLys) mutation combination was unique and predominant among Spanish hospitals. This mutation profile was also uniquely observed in these hospitals in our previous study. This suggests that clonal spread of BLNAR strains may occur locally. This finding warrants further investigation by using other molecular typing methods. In conclusion, notwithstanding large inter-regional differences, the overall resistance of H. influenzae to amoxicillin seems to decline in Europe due to a decreasing number of BL-producing strains, whereas the overall number of BLNAR strains remained relatively constant. PBP sequencing suggests the presence of several mutations associated with resistance and clonal spread of BLNAR strains. Cefixime, levofloxacin and moxifloxacin retained their excellent activity against both BL-producing and BLNAR strains of H. influenzae and thus remain useful treatment options for respiratory tract infection also in areas with persisting resistance problems.
|
Transparency declarations |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionTransparency declarationsReferences |
|---|
None to declare.
|
Acknowledgements |
|---|
Isolates were kindly provided by: H. Mittermayer (Austria); J. Etienne and J. Ngyen (France); U. Frank and F. J. Schmitz (Germany); T. McGaley (Ireland); P. E. Varaldo, A. Cavallero, A. Debbia and S. Stefani (Italy); A. Scoczynska (Poland); J. M. Amorim and G. Ribeiro (Portugal); R. Martin and A. Pascual (Spain); A. B. Sumerkan, Z. Gulay and D. Gür (Turkey); and A. Jackson and J. Ashby (UK). This study was supported by Astellas Pharma Inc. (Fujisawa Pharmaceutical Co., Ltd).
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionTransparency declarationsReferences |
|---|
1 Matic V, Bozdogan B, Jacobs MR, et al. (2003) Contribution of ß-lactamase and PBP amino acid substitutions to amoxicillin/clavulanate resistance in ß-lactamase-positive, amoxicillin/clavulanate-resistant Haemophilus influenzae. J Antimicrob Chemother 52:1018–21.
2
Fluit AC, Florijn A, Verhoef J, et al. (2005) Susceptibility of European ß-lactamase-positive and -negative Haemophilus influenzae isolates from the periods 1997/1998 and 2002/2003. J Antimicrob Chemother 56:133–8.
3
Hasegawa K, Chiba N, Kobayashi R, et al. (2004) Rapidly increasing prevalence of ß-lactamase-nonproducing, ampicillin-resistant Haemophilus influenzae type b in patients with meningitis. Antimicrob Agents Chemother 48:1509–14.
4
Heilmann KP, Rice CL, Miller AL, et al. (2005) Decreasing prevalence of ß-lactamase production among respiratory tract isolates of Haemophilus influenzae in the United States. Antimicrob Agents Chemother 49:2561–4.
5
Dabernat H, Delmas C, Seguy M, et al. (2002) Diversity of ß-lactam resistance-conferring amino acid substitutions in penicillin-binding protein 3 of Haemophilus influenzae. Antimicrob Agents Chemother 46:2208–18.
6
Ubukata K, Shibasaki Y, Yamamoto K, et al. (2001) Association of amino acid substitutions in penicillin-binding protein 3 with ß-lactam resistance in ß-lactamase-negative ampicillin-resistant Haemophilus influenzae. Antimicrob Agents Chemother 45:1693–9.
7
Bogdanovich T, Bozdogan B, Appelbaum PC. (2006) Effect of efflux on telithromycin and macrolide susceptibility in Haemophilus influenzae. Antimicrob Agents Chemother 50:893–8.
8
Osaki Y, Sanbongi Y, Ishikawa M, et al. (2005) Genetic approach to study the relationship between penicillin-binding protein 3 mutations and Haemophilus influenzae ß-lactam resistance by using site-directed mutagenesis and gene recombinants. Antimicrob Agents Chemother 49:2834–9.
9 Hasegawa K, Yamamoto K, Chiba N, et al. (2003) Diversity of ampicillin-resistance genes in Haemophilus influenzae in Japan and the United States. Microb Drug Resist 9:39–46.[Web of Science][Medline]
10 Ubukata K. (2003) Problems associated with high prevalence of multidrug-resistant bacteria in patients with community-acquired infections. J Infect Chemother 9:285–91.[CrossRef][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  S. Ladhani, P. T. Heath, M. E. Ramsay, and M. P. E. Slack Changes in antibiotic resistance rates of invasive Haemophilus influenzae isolates in England and Wales over the last 20 years J. Antimicrob. Chemother., October 1, 2008; 62(4): 776 - 779. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Garcia-Cobos, J. Campos, E. Cercenado, F. Roman, E. Lazaro, M. Perez-Vazquez, F. de Abajo, and J. Oteo Antibiotic Resistance in Haemophilus influenzae Decreased, except for {beta}-Lactamase-Negative Amoxicillin-Resistant Isolates, in Parallel with Community Antibiotic Consumption in Spain from 1997 to 2007 Antimicrob. Agents Chemother., August 1, 2008; 52(8): 2760 - 2766. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Garcia-Cobos, J. Campos, F. Roman, C. Carrera, M. Perez-Vazquez, B. Aracil, and J. Oteo Low {beta}-Lactamase-Negative Ampicillin-Resistant Haemophilus influenzae Strains Are Best Detected by Testing Amoxicillin Susceptibility by the Broth Microdilution Method Antimicrob. Agents Chemother., July 1, 2008; 52(7): 2407 - 2414. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Takahata, Y. Kato, Y. Sanbongi, K. Maebashi, and T. Ida Comparison of the Efficacies of Oral {beta}-Lactams in Selection of Haemophilus influenzae Transformants with Mutated ftsI Genes Antimicrob. Agents Chemother., May 1, 2008; 52(5): 1880 - 1883. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Hotomi, K. Fujihara, D. S. Billal, K. Suzuki, T. Nishimura, S. Baba, and N. Yamanaka Genetic Characteristics and Clonal Dissemination of {beta}-Lactamase-Negative Ampicillin-Resistant Haemophilus influenzae Strains Isolated from the Upper Respiratory Tract of Patients in Japan Antimicrob. Agents Chemother., November 1, 2007; 51(11): 3969 - 3976. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Torrico, L. Aguilar, N. Gonzalez, M. J. Gimenez, O. Echeverria, F. Cafini, D. Sevillano, L. Alou, P. Coronel, and J. Prieto Influence of TEM-1 {beta}-Lactamase on the Pharmacodynamic Activity of Simulated Total versus Free-Drug Serum Concentrations of Cefditoren (400 Milligrams) versus Amoxicillin-Clavulanic Acid (2,000/125 Milligrams) against Haemophilus influenzae Strains Exhibiting an N526K Mutation in the ftsI Gene Antimicrob. Agents Chemother., October 1, 2007; 51(10): 3699 - 3706. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Fenoll, L. Aguilar, O. Robledo, M.-J. Gimenez, D. Tarrago, J.-J. Granizo, M. Gimeno, and P. Coronel Influence of the {beta}-lactam resistance phenotype on the cefuroxime versus cefditoren susceptibility of Streptococcus pneumoniae and Haemophilus influenzae recovered from children with acute otitis media J. Antimicrob. Chemother., August 1, 2007; 60(2): 323 - 327. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Garcia-Cobos, J. Campos, E. Lazaro, F. Roman, E. Cercenado, C. Garcia-Rey, M. Perez-Vazquez, J. Oteo, and F. de Abajo Ampicillin-Resistant Non-{beta}-Lactamase-Producing Haemophilus influenzae in Spain: Recent Emergence of Clonal Isolates with Increased Resistance to Cefotaxime and Cefixime Antimicrob. Agents Chemother., July 1, 2007; 51(7): 2564 - 2573. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Tristram, M. R. Jacobs, and P. C. Appelbaum Antimicrobial Resistance in Haemophilus influenzae Clin. Microbiol. Rev., April 1, 2007; 20(2): 368 - 389. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||