Microbiology, Southmead Hospital, Southmead Road, Bristol BS10 5NB, UK

Transcription

1 Journal of Antimicrobial Chemotherapy (2008) 62, Suppl. 2, ii97 ii103 doi: /jac/dkn356 Non-susceptibility trends among Haemophilus influenzae and Moraxella catarrhalis from community-acquired respiratory tract infections in the UK and Ireland, Ian Morrissey 1 *, Kirsty Maher 1, Laura Williams 1, Jemma Shackcloth 1, David Felmingham 1 and Rosy Reynolds 2 on behalf of the BSAC Working Parties on Resistance Surveillance 1 Quotient Bioresearch Limited, Microbiology, 7-9 William Road, London NW1 3ER, UK; 2 Department of Medical Microbiology, Southmead Hospital, Southmead Road, Bristol BS10 5NB, UK Introduction Objectives: To determine the antimicrobial susceptibility of Haemophilus influenzae and Moraxella catarrhalis causing community-acquired respiratory tract infections in the UK and Ireland from 1999/ 2000 to 2006/07. Methods: Sentinel laboratories across the UK and Ireland contributed up to a fixed quota of isolates of defined organisms per annum. A central laboratory confirmed the isolates identities, measured MICs by the BSAC agar dilution method and undertook further testing by standard methods. The variability of the MIC method was assessed by repeated annual testing of control isolates. BSAC and EUCAST breakpoints were used. Statistical analysis adjusted for inter-centre variation by random effects logistic regression. Results: A total of 7371 H. influenzae and 2529 M. catarrhalis isolates were investigated. Over 90% of the H. influenzae isolates were susceptible to most of the antimicrobials tested, the exceptions being ampicillin (84.6% susceptible), trimethoprim (84.0%), cefuroxime (82.9%), amoxicillin (77.2%) and cefaclor (11.7%). For M. catarrhalis, resistance was solely due to b-lactamase ( prevalence over 91%) reducing susceptibility to penicillins only. There was little evidence of decreased antimicrobial susceptibility between 1999 and 2007 in either pathogen, except for a reduction in susceptibility to trimethoprim in H. influenzae (90.3% to 82.6%, P < ). On the other hand, tetracycline susceptibility in H. influenzae increased over this period in the UK and Ireland (96.5 to 98.8%, P ). Conclusions: Despite increased resistance in respiratory pathogens from other parts of the world, the susceptibility of H. influenzae and M. catarrhalis to all agents, except tetracycline and trimethoprim in the case of H. influenzae, has remained constant during this longitudinal study. Keywords: resistance, surveillance, MIC, breakpoint Haemophilus influenzae and Moraxella catarrhalis (formerly Branhamella catarrhalis or Neisseria catarrhalis) are important respiratory pathogens, with the former being the more prevalent bacterium associated with community-acquired pneumonia and acute exacerbations of chronic obstructive airways disease, whereas the latter is found most commonly in patients with sinusitis. 1 Tetracycline resistance has been documented in H. influenzae since the early 1970s, 2 with a small collection of USA isolates (n ¼ 35) from 1972 showing complete resistance to this agent. 3 In that study, no isolate was found to be ampicillin-resistant, although the authors state that some ampicillin-resistant H. influenzae had been found in the same hospital in Other reports from the same period acknowledge that ampicillin resistance was rare in the 1970s. 4 Interestingly, another study found ampicillin resistance but not tetracycline resistance in a collection of 40 H. influenzae isolates from the USA around the same time. 5 During the 1970s there was no specific surveillance programme like those that exist today, and it is unclear what sampling and testing differences there may have been between the two studies discussed above. However, at that time, efforts were beginning to be made to standardize susceptibility methods for use when testing H. influenzae. 6 Between the mid-1970s and early 1990s, the susceptibility of UK isolates of H. influenzae was evaluated in a more formal... *Corresponding author. Tel: þ ; Fax: þ ; ian.morrissey@quotientbioresearch.com... ii97 # The Author Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please journals.permissions@oxfordjournals.org

2 Morrissey et al. way at the London Hospital, led by the late Professor J. D. Williams. A collection of 952 H. influenzae isolates from 1977 included 1.6% that were resistant to ampicillin (all of these being b-lactamase-positive) and 2.7% resistant to tetracycline. 7 Repeat surveillance in 1981 (with a larger collection of 1841 isolates) indicated that resistance to ampicillin in the UK had risen almost 4-fold to 6.2%, whereas tetracycline resistance remained fairly static at 3.1%. 8 Unlike in 1977, some b-lactamase-negative ampicillin-resistant (BLNAR) isolates were found (0.9% of all isolates), BLNAR being defined as b-lactamase non-producing strains with ampicillin MIC 4 mg/ L In 1986, further increases in ampicillin resistance were observed (then at 7.8%), but the rise was not as dramatic as that seen between 1977 and BLNAR accounted for 1.6% of all H. influenzae in 1986, almost double that for Tetracycline resistance was 2.7% in 1986, thus with little change from 1977, despite complete resistance reported from a collection of 35 isolates from a USA hospital (tetracycline, MIC range mg/l) in The final study of the London Hospital series, in 1991, showed a further increase in ampicillin resistance, up to 14.4%. The increase in BLNAR was even greater with these isolates and accounted for 5.8% of all H. influenzae in By this time, tetracycline resistance dropped 4-fold to 1.4%. 10 Two other studies of susceptibility in H. influenzae in the UK were carried out in the mid-1990s, 11,12 where amoxicillin resistance was around 16%, with b-lactamase prevalence hardly changed since the early 1990s. Cefaclor resistance was 14% and clarithromycin resistance 4% in the mid-1990s. 11,12 Tetracycline susceptibility was not assessed in either of these studies. Unlike H. influenzae, M. catarrhalis was only recognized as a bacterial pathogen, rather than a commensal, in the mid-1970s and even as recently as the late 1980s was often still referred to as N. catarrhalis and dismissed as clinically irrelevant. 13 Studies on penicillins in the 1960s suggest that this pathogen was susceptible to ampicillin at that time. 14 The first reported b-lactamase-producing M. catarrhalis were found in Sweden in 1976, at a prevalence rate of 3.8%. 15 A study of 121 M. catarrhalis isolated in late 1979, also from Sweden, indicated that b-lactamase prevalence had risen to 17%. 16 During 1981 and 1982, the prevalence of b-lactamase in M. catarrhalis was around 45% in an Edinburgh hospital 17 and by 1984 was at nearer 70%. 18 Elsewhere in the UK, b-lactamase prevalence was around 50% in the mid-1980s. 19 In the USA, higher rates (80 to 90%) were reported over this period, and similar extremely high prevalence was also recorded in the UK in the early 1990s 22,23 and globally by the mid- to late-1990s. 12,24 Although b-lactamase was not formally monitored in M. catarrhalis, it is clear that b-lactamase levels have increased from next to nothing in the 1960s to almost complete carriage worldwide in the 1990s. This paper describes susceptibility data for both pathogens from the BSAC Respiratory Surveillance Programme from 1999/ 2000 to 2006/07. Materials and methods The methodology used is described in the survey, laboratory and statistical methods for the BSAC Resistance Surveillance Programmes paper in this Supplement. 25,26 Briefly, approximately 20 laboratories across the UK and Ireland submitted up to 50 S. pneumoniae and H. influenzae and 25 M. catarrhalis each winter (October April) from 1999 to 2007, excluding samples taken.48 h after hospitalization. These isolates were re-identified using standard methods, and MICs and susceptibility were determined using BSAC standard methods. Results Isolate numbers and patient demographics The numbers of H. influenzae and M. catarrhalis isolates tested and source data are shown in Table 1. Each pathogen had a group of core antibiotics that were tested each year, whereas other antibiotics were either excluded or introduced in more recent years. Both pathogens were investigated for the presence of b-lactamase. 25 The total collection of isolates included 7371 H. influenzae and 2529 M. catarrhalis. Almost the entire collection of isolates originated from sputum samples (.93%). Throughout, patient characteristics were very similar for either pathogen year on year. Most isolates were from the year and year old age groups (Table 1), and patients with either pathogen were approximately evenly split between female and male (Table 1). Slightly more than 50% of the isolates originated from hospitalized patients (collected within 48 h of admission), with most of the remainder from General Practice (Table 1). Country of origin was also very similar for both pathogens: 61% from England, 13% from Scotland, 11% from Wales, 10% from the Republic of Ireland and 5% from Northern Ireland (data not shown). This is not surprising because collection targets were based on a set number of each isolate per country. All these characteristics did not alter significantly from year to year (statistical analysis not shown), with the exception of care setting where in the last two seasons (2005/06 and 2006/ 07) a significantly higher proportion came from General Practice (56% of both pathogens compared with 40% to 44% for the first six seasons). H. influenzae susceptibility Care setting, specimen type, age and sex had no significant relationship with susceptibility or b-lactamase production (data and statistical analysis not shown). Antibiotic non-susceptibility (intermediate and resistant categories combined) and b-lactamase prevalence for H. influenzae from 1999/2000 to 2006/07 are shown in Figure 1. EUCAST/ BSAC breakpoints aim to categorize H. influenzae as intermediate to macrolides so data for erythromycin and clarithromycin are omitted from Figure 1. Cefaclor is also omitted because percentage non-susceptibility was high (at least 70% each year). Data for the fluoroquinolones (ciprofloxacin, levofloxacin, moxifloxacin and gemifloxacin) and for cefotaxime and ertapenem are omitted because the percentage non-susceptibility was low in each year (0.6% for fluoroquinolones, 1.7% for cefotaxime and 0.3% for ertapenem). No susceptibility breakpoints are published for faropenem, minocycline or tigecycline; therefore, data cannot be calculated to be included in Figure 1. Summary MIC and susceptibility data for all H. influenzae isolates combined are shown in Table 2. Tigecycline was approximately one doubling dilution more active than minocycline against isolates with tetracycline MIC. 1 mg/l (data not ii98

3 H. influenzae and M. catarrhalis susceptibility in the UK Table 1. Sources of H. influenzae and M. catarrhalis isolates collected between October 1999 and April 2007 Patient location (%) a Sex (%) GP hospital,48 h institutional c male female Mode of age distribution b H. influenzae (7371) M. catarrhalis (2529) a Percentages do not total 100 because of isolates with missing data. b Age bands were.5, 5 19, 20 29, 30 39, 40 49, 50 59, 60 69, and.80 years. c Institutional refers to prison or nursing home. Percent non-susceptible Amoxicillin/clavulanate Ampicillin Amoxicillin Beta-lactamase Cefuroxime Tetracycline Trimethoprim Figure 1. Change in percentage non-susceptibility for antibiotics against H. influenzae collected between 1999/2000 and 2006/07. shown). Faropenem activity was very similar to that seen with amoxicillin/clavulanate (Table 2). As would be expected, b-lactamase prevalence over time mirrored ampicillin non-susceptibility, which wavered around 15% year on year (Figure 1). Over the whole collection, 0.9% or 69 isolates were BLNAR according to the current BSAC definition of ampicillin resistance (MIC. 1 mg/l). 27 Ampicillin MIC against all but four BLNAR isolates was 2 or 4 mg/l (i.e. just above the breakpoint). The four isolates with higher ampicillin MIC (three at 8 mg/l and one at 32 mg/l) were collected from three hospitals and spread over the whole study period (data not shown). It can be seen that amoxicillin non-susceptibility remained steady at around 20 to 22% until 2003/04 before dropping to around 16% in 2004/05 and then increasing to around 30% in 2005/06 and 2006/07 (Figure 1). A similar pattern was also observed with amoxicillin/clavulanate (albeit at a lower nonsusceptibility rate). Interestingly, this trend did not occur with ampicillin (Figure 1). Figure 2 shows the MIC distribution for both amoxicillin and ampicillin (data pre-2003 omitted for clarity). It can be seen that the amoxicillin MIC distribution was generally one dilution higher than that for ampicillin, producing mode MICs of 0.5 and 0.25 mg/l, respectively. It is clear that an increase in the prevalence of isolates with amoxicillin MIC of 2 mg/l is the cause of higher amoxicillin non-susceptibility in the latter 2 years of the study (Figure 2b). A similar MIC shift also occurred with ampicillin but at 1 mg/l (Figure 2a) and below the breakpoint. Therefore, amoxicillin susceptibility, but not ampicillin susceptibility, was affected by the MIC shift. In these cases, where MIC breakpoints divide the wild-type MIC distribution, it is difficult to be confident that small MIC shifts truly represent a change in susceptibility, so it is unlikely that this change truly represents the development of resistance to amoxicillin. The same argument is also true for amoxicillin/ clavulanate. Cefuroxime non-susceptibility fluctuated between 13% and 24% over the eight seasons (Figure 1), but there was no significant trend. Trimethoprim non-susceptibility, on the other hand, rose significantly from around 10% in 1999/2000 to around 17% in 2006/07 (P, ). Conversely, there was a significant though slow downward trend (P ¼ ) in tetracycline nonsusceptibility which reduced from 3.5% in 1999/2000 to 1.2% in 2006/07 and dipped as low as 0.9% in 2004/05. ii99

4 Morrissey et al. Table 2. Combined summary MIC and susceptibility data for all H. influenzae collected between October 1999 and April 2007 MIC (mg/l) Percentage of isolates Antibiotic Total no. of isolates tested a min b 50% c 90% d max e susceptible intermediate resistant Amoxicillin Amoxicillin/clavulanate Ampicillin Cefaclor Cefotaxime Cefuroxime Ciprofloxacin Clarithromycin Ertapenem Erythromycin Faropenem f Gemifloxacin Levofloxacin Minocycline Moxifloxacin Tetracycline Tigecycline Trimethoprim a Not all antibiotics were tested each year. 25 b Min, minimum MIC. c 50%, MIC where 50% of the isolates are inhibited. d 90%, MIC where 90% of the isolates are inhibited. e Max, maximum MIC. f, breakpoint not given. M. catarrhalis susceptibility The susceptibility of isolates collected in the 2001/02 and 2003/ 04 seasons was not determined, so the data presented here do not include isolates collected during those seasons. The most distinctive, but unsurprising, feature of M. catarrhalis is b-lactamase prevalence, which was in excess of 91% from 1999/2000 to 2006/07. No significant trend of changing susceptibility was observed for any antibiotic over time (data not shown), so only summary MIC and susceptibility data for the isolate collection as a whole are given (Table 3). It can be seen that, as expected, the high b-lactamase prevalence produced almost uniform resistance to ampicillin (no breakpoints are given for amoxicillin) and almost 80% resistance to cefaclor (Table 3). Full resistance was less common to cefuroxime (5.6%) but intermediate susceptibility was observed in almost 30% of the isolates (Table 3). More than 99% susceptibility was observed for all other antibiotics (Table 3). Discussion Susceptibility data presented here show that since 1999 ampicillin non-susceptibility (including BLNAR) and b-lactamase prevalence in H. influenzae from the UK have remained constant at around 15%, despite a steady increase in the 1970s and 1980s. Levels observed in 1999 were no different to those found in 1994/95 and 1995/96. 11,12 Tetracycline resistance, on the other hand, has significantly reduced over time. This may reflect a tendency in recent years not to use older tetracyclines for the treatment of community-acquired respiratory tract infection (RTI) following local or national guidelines. 28 It might appear, at first glance, that cefaclor resistance has increased dramatically in the UK from around 14% during pre-bsac surveillance in the mid-1990s 11,12 to over 80% in Similarly, the pattern of clarithromycin resistance would appear to have altered considerably over the same period. However, these effects are artefacts due to the use of different breakpoints. For example, cefaclor data from 1994/95 and 1995/ 96 were based on a resistance breakpoint of 32 mg/l. 11,12 For this current study, the BSAC breakpoint of.1 mg/l was used, leading to a resistance rate of 88.3%. 27 Re-analysis of the current data using a 32 mg/l breakpoint would estimate cefaclor resistance as 3.1% (data not shown), showing that, if anything, cefaclor resistance has fallen since This illustrates that care should be taken when comparing historical data with recent studies, where breakpoints may be quite different to those used now. Similarly, current breakpoints used by other testing methods may vary compared with those used by the BSAC. Direct comparisons can, however, be made between isolates collected over the eight winter seasons during the current BSAC UK surveillance study. It would appear that the susceptibility of H. influenzae to antibacterial agents has hardly changed over the study period. High levels of susceptibility were seen with fluoroquinolones, cefotaxime and ertapenem. No breakpoints were available for faropenem, minocycline or tigecycline, but on MIC values alone these agents would appear to be at least as active as most other agents tested. However, amoxicillin ii100

5 H. influenzae and M. catarrhalis susceptibility in the UK (a) 60% 50% % Prevalence 30% 20% 10% 0% (b) 60% 50% 40% 30% 20% 10% 0% Prevalence Ampicillin MIC (mg/l) Amoxicillin MIC (mg/l) Figure 2. MIC distribution of (a) ampicillin and (b) amoxicillin against H. influenzae from 2003/04 to 2006/07. Vertical line represents breakpoint between susceptible and non-susceptible. non-susceptibility appears to have increased between 2005 and 2007 despite ampicillin non-susceptibility remaining fairly constant. This is due to a slight MIC shift that pushes amoxicillin MICs, but not ampicillin MICs, over the breakpoint. Although this shift may be very small and may simply be due to minor experimental variability, it does beg the question whether susceptibility breakpoints should be the same for these agents that are similarly affected by b-lactamases but have differing MIC distributions. 12 Furthermore, susceptibility reporting will vary from laboratory to laboratory depending on whether ampicillin or amoxicillin is used as the indicator. It would appear more appropriate to set amoxicillin breakpoints that match b-lactamase prevalence and ampicillin breakpoints. The susceptibility of M. catarrhalis has also changed little since It is interesting to note that, despite almost universal b-lactamase prevalence, resistance to other antibacterial agents has not developed in M. catarrhalis. Antibiotic resistance in community-acquired RTI pathogens in the UK and Ireland is generally perceived as being low compared with many other parts of the world, but this is mainly in relation to Streptococcus pneumoniae. 29 Within the European countries, b-lactamase prevalence in H. influenzae has been ii101

6 Morrissey et al. Table 3. Combined summary MIC and susceptibility data for all M. catarrhalis collected between October 1999 and April 2007, excluding isolates collected in 2001/02 and 2003/04 MIC (mg/l) Percentage of isolates Antibiotic Total no. of isolates tested a min b 50% c 90% d max e susceptible intermediate resistant Amoxicillin f Amoxicillin/clavulanate Ampicillin Cefaclor Cefotaxime Cefuroxime Ciprofloxacin Clarithromycin Ertapenem Erythromycin Faropenem Gemifloxacin Levofloxacin Minocycline Moxifloxacin Tetracycline Tigecycline Trimethoprim a Not all antibiotics were tested each year. 25 b Min, minimum MIC. c 50%, MIC where 50% of the isolates are inhibited. d 90%, MIC where 90% of the isolates are inhibited. e Max, maximum MIC. f, breakpoint not given. shown to be lower in Germany (2.0%), Italy (3.3%), The Netherlands (5.9%), Poland (5.7%), Turkey (5.0%) and Spain (13.9%), but higher in France (29.1%) and Portugal (38.2%), than in the UK. 30 A recent study of eastern Europe has also shown high b-lactamase prevalence (as approximated from ampicillin susceptibility) in Romania (31.9%). 31 High b-lactamase prevalence has been reported in Kuwait (26.7%), Lebanon (20.4%), Tunisia (21.1%), 32 USA (27.4%), 33 Thailand (48.4%) 34 and South Korea (64.7%). 35 Data presented here confirm that the UK and Ireland have relatively high b-lactamase prevalence in H. influenzae compared with most other European countries, but higher prevalence occurs elsewhere in the world. The reason for country variation is unknown, but what is clear is that even higher resistance rates could, at least theoretically, occur in the UK. Interestingly, it has been suggested that b-lactamase prevalence in H. influenzae has decreased in recent years in some countries. 36 Future years of BSAC respiratory surveillance activity are planned to track any such changes in antibacterial susceptibility in the UK and Ireland. Additional information on the isolates collected in the Project is available on the BSAC surveillance website ( or through a link on the BSAC homepage See page ii12 (Publications) for a full list of previous publications from the Project, which may include parts of the information presented here. Funding The BSAC Respiratory Resistance Surveillance Programme 1999/ /07 has received financial support from Abbott, Aventis, Bayer, GeneSoft, GlaxoSmithKline, Merck Sharp & Dohme, Wyeth or their predecessors. The BSAC funds the work of the Resistance Surveillance Co-ordinator (R. R.) and Resistance Surveillance Working Party. Transparency declarations Acknowledgements We are grateful to all who have contributed to the success of the BSAC Resistance Surveillance Project, in particular the many laboratories that have collected isolates and all who have played a part in testing them [see page ii10 (Acknowledgements)]. This article is part of a Supplement sponsored by the British Society for Antimicrobial Chemotherapy. I. M. and D. F. have accepted grants, speaking invitations and conference invitations from most major pharmaceutical companies in recent years. All other authors have none to declare. ii102

7 H. influenzae and M. catarrhalis susceptibility in the UK References 1. Pfaller MA, Ehrhardt AF, Jones RN. Frequency of pathogen occurrence and antimicrobial susceptibility among community-acquired respiratory tract infections in the respiratory surveillance program study. Am J Med 2001; 111 Suppl 9A: 4S 12S. 2. Anon. Chemotherapy of bronchitis. BMJ 1970; 1: Finland M, Garner C, Wilcox C et al. Susceptibility of pneumococci and Haemophilus influenzae to antibacterial agents. Antimicrob Agents Chemother 1976; 9: Gordon RC, Thompson TR, Stevens LI et al. In vitro susceptibility of Haemophilus influenzae to eight antibiotics. Antimicrob Agents Chemother 1974; 6: Thornsberry C, Kirven LA. Antimicrobial susceptibility of Haemophilus influenzae. Antimicrob Agents Chemother 1974; 6: Jorgensen JH, Jones PM. Simplified medium for ampicillin susceptibility testing of Haemophilus influenzae. Antimicrob Agents Chemother 1975; 7: Howard AJ, Hince CJ, Williams JD. Antibiotic resistance in Streptococcus pneumoniae and Haemophilus influenzae. BMJ 1978; 1: Philpott-Howard J, Williams JD. Increase in antibiotic resistance in Haemophilus influenzae in the United Kingdom since 1977: report of study group. BMJ 1982; 284: Powell M, Koutsia-Carouzou C, Voutsinas D et al. Resistance of clinical isolates of Haemophilus influenzae in United Kingdom BMJ 1987; 295: Powell M, Fan YS, Seymour A et al. Antimicrobial resistance in Haemophilus influenzae from England and Scotland in J Antimicrob Chemother 1992; 29: Felmingham D, Robbins MJ, Dencer C et al. Antimicrobial susceptibility of community-acquired bacterial lower respiratory tract pathogens. J Antimicrob Chemother 1996; 38: Felmingham D, Robbins MJ, Tesfaslasie Y et al. Antimicrobial susceptibility of community-acquired lower respiratory tract bacterial pathogens isolated in the UK during the cold season. J Antimicrob Chemother 1998; 41: Catlin BW. Branhamella catarrhalis: an organism gaining respect as a pathogen. Clin Microbiol Rev 1990; 3: Barber M, Waterworth PM. Antibacterial activity of the penicillins. BMJ 1962; 1: Malmvall B-E, Brorsson J-E, Johnsson J. In vitro sensitivity to penicillin V and b-lactamase production of Branhamella catarrhalis. J Antimicrob Chemother 1977; 3: Brorson JE, Martinell J, Wilske H. Branhamella catarrhalis: antibiotic susceptibility and b-lactamase production. J Antimicrob Chemother 1981; 7: McLeod DT, Ahmad F, Power JT et al. Bronchopulmonary infection due to Branhamella catarrhalis. BMJ 1983; 287: McLeod DT, Ahmad F, Capewell S et al. Increase in bronchopulmonary infection due to Branhamella catarrhalis. BMJ 1986; 292: Winstanley TG, Spencer RC. Moraxella catarrhalis antibiotic susceptibility with special reference to trimethoprim. J Antimicrob Chemother 1986; 18: Sweeney KG, Verghese A, Needham CA. In vitro susceptibilities of isolates from patients with Branhamella catarrhalis pneumonia compared with those of colonizing strains. Antimicrob Agents Chemother 1985; 27: Alvarez S, Jones M, Holtsclaw-Berk S et al. In vitro susceptibilities and beta-lactamase production of 53 clinical isolates of Branhamella catarrhalis. Antimicrob Agents Chemother 1985; 27: Powell M, McVey D, Kassim MH et al. Antimicrobial susceptibility of Streptococcus pneumoniae, Haemophilus influenzae and Moraxella (Branhamella) catarrhalis isolated in the UK from sputa. J Antimicrob Chemother 1991; 28: Fung CP, Powell M, Seymour A et al. The antimicrobial susceptibility of Moraxella catarrhalis isolated in England and Scotland in J Antimicrob Chemother 1992; 30: Doern GV, Jones RN, Pfaller MA et al. Haemophilus influenzae and Moraxella catarrhalis from patients with community-acquired respiratory tract infections: antimicrobial susceptibility patterns from the SENTRY antimicrobial Surveillance Program (United States and Canada, 1997). Antimicrob Agents Chemother 1999; 43: Reynolds R, Hope R, Williams L et al. Survey, laboratory, and statistical methods for the BSAC Resistance Surveillance Programmes. J Antimicrob Chemother 2008; 62 Suppl 2: ii Reynolds R, Lambert P, Burton P et al. Analysis, power and design of antibiotic resistance surveillance studies taking account of inter-centre variation and turnover. J Antimicrob Chemother 2008; 62 Suppl 2: ii disc_susceptibility_method.cfm (24 January 2008, date last accessed). 28. Woodhead M, Macfarlane J, BTS CAP Guidelines Committee. Local antibiotic guidelines for adult community-acquired pneumonia (CAP): a survey of UK hospital practice in J Antimicrob Chemother 2000; 46: Marchese A, Schito GC. Recent findings from multinational resistance surveys: are we PROTEKTed from resistance? Int J Antimicrob Agents 2007; 29 Suppl 1: S2 S Fluit AC, Florijn A, Verhoef J et al. Susceptibility of European b-lactamase-positive and -negative Haemophilus influenzae isolates from the periods 1997/1998 and 2002/3. J Antimicrob Chemother 2000; 45: Gracia M, Díaz C, Coronel P et al. Antimicrobial susceptibility of Haemophilus influenzae and Moraxella catarrhalis isolates in eight Central, East and Baltic European countries in : results of the cefditoren Surveillance Study. J Antimicrob Chemother 2008; 61: Alpuche C, Garau J, Lim V. Global and local variations in antimicrobial susceptibilities and resistance development in the major respiratory pathogens. Int J Antimicrob Agents 2007; 30S: S Critchley IA, Brown SD, Traczewski MM et al. National and regional assessment of antimicrobial resistance among communityacquired respiratory tract pathogens identified in a U.S. Faropenem Surveillance Study. Antimicrob Agents Chemother 2007; 51: Srifuengfung S, Chayakulkeeree M, Chokephaibulkit K et al. Five-year study of antimicrobial susceptibility and beta-lactamase production in Haemophilus influenzae. Southeast Asian J Trop Med Public Health 2007; 38: Hoban D, Felmingham D. The PROTEKT surveillance study: antimicrobial susceptibility of Haemophilus influenzae and Moraxella catarrhalis from community-acquired respiratory tract infections. J Antimicrob Chemother 2002; 50 Suppl 2: Tristram S, Jacobs MR, Appelbaum PC. Antimicrobial resistance in Haemophilus influenzae. Clin Microbial Rev 2007; 20: ii103