International Journal of Antimicrobial Agents

International Journal of Antimicrobial Agents 35 (2010) 227 234 Contents lists available at ScienceDirect International Journal of Antimicrobial Agents journal homepage: http://www.elsevier.com/locate/ijantimicag Antimicrobial resistance among clinical isolates from the Chinese Meropenem Surveillance Study (CMSS), 2003 2008 Hui Wang a,, Minjun Chen a,, Yuxing Ni b, Yudong Liu a, Hongli Sun a, Yunsong Yu c, Xiujuan Yu d, Yaning Mei e, Min Liu f, Ziyong Sun g, Yunzhuo Chu h, Zhidong Hu i, Xinhong Huang j a Department of Clinical Laboratory, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences, No. 1 Shuaifuyuan, Wangfujing, Beijing 100370, China b Ruijin Hospital, School of Medicine, Shanghai Jiaotong University, Shanghai 200025, China c First Affiliated Hospital, College of Medicine, Zhejiang University, Hangzhou 310003, China d Qi Lu Hospital of Shandong University, Jinan 250012, China e The First Affiliated Hospital of Nanjing Medical University, Nanjing 210029, China f The First Affiliated Hospital of Sun Yat-Sen University, Guangzhou 510060, China g Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430030, China h The First Hospital of China Medical University, Shenyang 110001, China i General Hospital, Tianjin Medical University, Tianjin 300052, China j Fujian Medical University Union Hospital, Fuzhou 350001, China article info abstract Article history: Received 27 September 2009 Accepted 17 November 2009 Keywords: Antibiotic surveillance Gram-negative bacilli Resistance Meropenem The Chinese Meropenem Surveillance Study (CMSS) programme was initiated in 2003 with the aim of monitoring the antimicrobial activity of broad-spectrum agents against nosocomial Gram-negative bacilli in China. From 2003 to 2008, a total of 3892 isolates were collected from 10 teaching hospitals. The minimum inhibitory concentrations (MICs) of 11 antimicrobial agents were determined by the agar dilution method. During the study period, a marked decrease in the susceptibility of Acinetobacter spp. to meropenem and imipenem was noticed, from 94.6% to 60.7% and from 92.5% to 62.1%, respectively. However, for Pseudomonas aeruginosa the susceptibility was relatively stable, with susceptibility rates of 86.2% to 76.0% for meropenem and 74.8% to 70.5% for imipenem. Meropenem and imipenem exhibited the highest activities against enterobacterial organisms, with ranges of MIC 90 values (MIC for 90% of the organisms) from 0.064 mg/l to 0.25 mg/l and 0.25 to 4 mg/l, respectively. Except for Acinetobacter spp., the next most active agent against the majority of isolates was amikacin, with susceptibility ranging from 78.8% to 93.3%, followed by piperacillin/tazobactam (73.7% to 98.2%), cefoperazone/sulbactam (63.9% to 99.1%), cefepime (67.0% to 95.4%) and ceftazidime (54.5% to 93.3%). The percentage of isolates positive for extended-spectrum -lactamases among Escherichia coli, Klebsiella spp. and Proteus mirabilis ranged from 50.9% to 66.7%, 25.4% to 42.4% and 8.9% to 24.2%, respectively. These CMSS results have demonstrated increasing resistance of Acinetobacter spp. to carbapenems, resulting from the spread of highly resistant clones. Continued surveillance studies, including CMSS, as well as potent measures for controlling the spread of resistant clones are required. 2009 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved. 1. Introduction The increase in antimicrobial resistance among Gram-negative bacilli is a significant and increasing problem worldwide [1 4]. Antimicrobial surveillance programmes that offer extensive information on patterns of bacterial resistance in different geographical regions and different periods play a pivotal role in the fight against increasing bacterial resistance and contribute to the selection of the most appropriate antimicrobial agents in clinical practice [5]. Corresponding authors. Tel.: +86 10 6529 5415. E-mail addresses: wh bj@tom.com (H. Wang), chmj1933@hotmail.com (M. Chen). The Meropenem Yearly Susceptibility Test Information Collection (MYSTIC) programme, which was initiated in 1997, is a unique global multicentre longitudinal surveillance study that has provided significant longevity and value [6]. However, this programme has mainly focused on North and South America and Europe, with few data having been reported from other regions, particularly mainland China. To address this gap, the Chinese Meropenem Surveillance Study (CMSS) programme was initiated in 2003. The CMSS is a multicentre national surveillance programme of the activity of meropenem and comparator agents that includes 10 teaching hospitals from 10 major cities throughout China. In the present report from CMSS, minimum inhibitory concentrations (MICs) are presented, and susceptibility and resistance rates are interpreted among 0924-8579/$ see front matter 2009 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved. doi:10.1016/j.ijantimicag.2009.11.010

228 H. Wang et al. / International Journal of Antimicrobial Agents 35 (2010) 227 234 Table 1 Overall in vitro susceptibility to 11 antimicrobial agents of clinical Gram-negative isolates in China, 2003 2008. Organism Antimicrobial agent MIC Susceptibility a Range MIC 50 MIC 90 %S %R Escherichia coli (n = 590) Pseudomonas aeruginosa (n = 548) Klebsiella spp. (n = 531) Enterobacter spp. (n = 522) Acinetobacter spp. (n = 486) Meropenem 0.008 to 8 0.016 0.064 99.7 0 Imipenem 0.008 to 64 0.125 0.25 99.3 0.3 Cefoxitin 0.064 to 8 64 60.6 24.2 Cefotaxime 0.016 to 32 256 40.0 40.8 Ceftriaxone 0.016 to 32 36.1 49.5 Ceftazidime 0.016 to 1 32 81.7 13.6 Cefepime 0.008 to 4 32 69.4 18.7 CSL 0.016 to 8 64 72.2 10.8 TZP 0.25 to 2 8 95.3 2.5 Amikacin 0.064 to 2 87.6 10.5 Ciprofloxacin 0.008 to 32 64 22.2 76.6 Meropenem 0.016 to 1 16 80.5 13.9 Imipenem 0.032 to 256 2 32 71.5 23.7 Cefoxitin 0.064 to 256 0.9 99.1 Cefotaxime 0.032 to 32 12.8 49.5 Ceftriaxone 0.032 to 64 5.7 65.0 Ceftazidime 0.064 to 256 4 70.6 23.7 Cefepime 0.032 to 4 32 67.0 19.2 CSL 0.064 to 16 63.9 22.1 TZP 0.5 to 8 256 73.7 26.3 Amikacin 0.032 to 256 4 78.8 19.7 Ciprofloxacin 0.002 to 0.5 32 67.9 28.6 Meropenem 0.008 to 0.032 0.064 98.3 1.3 Imipenem 0.016 to 0.25 0.5 98.5 1.1 Cefoxitin 1 to 4 256 65.3 27.6 Cefotaxime 0.016 to 4 57.7 23.2 Ceftriaxone 0.016 to 4 256 53.4 35.7 Ceftazidime 0.032 to 1 64 76.8 19.4 Cefepime 0.016 to 1 32 80.2 10.9 CSL 0.032 to 4 64 76.8 12.1 TZP 0.25 to 4 64 87.0 9.0 Amikacin 0.064 to 1 80.7 19.1 Ciprofloxacin 0.004 to 0.5 64 57.9 37.7 Meropenem 0.008 to 0.032 0.25 99.0 0.4 Imipenem 0.032 to 0.25 1 99.8 0 Cefoxitin 0.125 to 256 4.9 93.0 Cefotaxime 0.016 to 16 256 48.0 34.0 Ceftriaxone 0.032 to 16 46.8 37.0 Ceftazidime 0.016 to 4 256 54.5 41.5 Cefepime 0.008 to 0.5 64 77.9 16.5 CSL 0.016 to 4 64 71.6 17.3 TZP 0.125 to 4 256 76.6 11.7 Amikacin 0.256 to 2 83.9 14.6 Ciprofloxacin 0.004 to 0.25 64 63.5 31.3 Meropenem 0.008 to 1 32 75.3 21.5 Imipenem 0.032 to 1 64 77.2 20.1 Cefoxitin 0.064 to 256 2.9 95.3 Cefotaxime 0.016 to 29.9 53.8 Ceftriaxone 0.016 to 256 18.4 54.4 Ceftazidime 0.016 to 32 42.9 52.5 Cefepime 0.008 to 16 40.0 46.9 CSL 0.016 to 16 64 55.4 15.5 TZP 0.032 to 32 47.5 39.7 Amikacin 0.064 to 64 48.5 50.2 Ciprofloxacin 0.004 to 32 64 40.4 58.8 Proteus spp. (n = 331) Meropenem 0.008 to 64 0.032 0.125 99.4 0.6 Imipenem 0.008 to 1 4 99.1 0.6 Cefoxitin 0.032 to 4 8 95.0 2.9 Cefotaxime 0.016 to 256 0.064 16 85.7 4.0 Ceftriaxone 0.016 to >512 0.064 32 83.5 7.6 Ceftazidime 0.016 to 0.064 2 93.3 6.4 Cefepime 0.008 to 256 0.064 4 95.4 2.4 CSL 0.032 to 64 0.5 4 99.1 0.3 TZP 0.125 to 0.5 1 98.2 0.6 Amikacin 0.032 to 2 8 93.3 6.4 Ciprofloxacin 0.002 to 64 0.5 32 59.5 35.1

H. Wang et al. / International Journal of Antimicrobial Agents 35 (2010) 227 234 229 Table 1 (Continued) Organism Antimicrobial agent MIC Susceptibility a Range MIC 50 MIC 90 %S %R Citrobacter spp. (n = 254) Meropenem 0.008 to 0.032 0.125 99.6 0.4 Imipenem 0.032 to 0.25 1 99.6 0.4 Cefoxitin 1 to 256 12.0 83.3 Cefotaxime 0.016 to 8 51.8 32.4 Ceftriaxone 0.032 to 16 256 49.0 38.7 Ceftazidime 0.032 to 4 256 59.7 34.4 Cefepime 0.008 to 0.5 32 80.6 10.7 CSL 0.016 to 8 64 69.8 11.9 TZP 0.25 to 4 77.1 10.3 Amikacin 0.064 to 2 87.0 11.5 Ciprofloxacin 0.004 to 0.5 32 55.3 36.4 Serratia spp. (n = 195) Meropenem 0.008 to 2 0.032 0.064 100 0 Imipenem 0.016 to 4 0.5 1 100 0 Cefoxitin 0.125 to 16 14.1 48.7 Cefotaxime 0.016 to 0.25 77.4 17.4 Ceftriaxone 0.016 to 0.5 76.4 15.9 Ceftazidime 0.032 to 0.25 8 90.8 9.2 Cefepime 0.016 to 0.125 32 85.1 11.8 CSL 0.032 to 2 32 83.6 5.6 TZP 0.125 to 256 2 8 96.4 2.1 Amikacin 0.064 to 2 84.1 15.9 Ciprofloxacin 0.002 to 64 0.125 2 83.1 9.7 MIC, minimum inhibitory concentration; MIC 50/90, MIC for 50% and 90% of the organisms, respectively; %S, percent susceptible; %R, percent resistant; CSL, cefoperazone/sulbactam; TZP, piperacillin/tazobactam. a According to Clinical and Laboratory Standards Institute guidelines [8]. Gram-negative bacteria collected in China between 2003 and 2008. 2. Materials and methods 2.1. Bacterial isolates In 2003, 2004, 2006 and 2008, 10 teaching hospitals from 10 central cities throughout China participated in the CMSS programme. The cities are distributed in southern China (including Shanghai, Hangzhou, Wuhan, Nanjing, Guangzhou and Fuzhou) and northern China (including Beijing, Tianjin, Shenyang and Jinan). From September to December of each surveillance year, each centre collected 100 consecutive, non-repeat clinical isolates of Gram-negative bacilli and sent them to the central laboratory (Clinical Microbiology, Peking Union Medical College Hospital, Beijing, China). Owing to its intrinsic resistance to carbapenems, Stenotrophomonas maltophilia was excluded. A total of 3892 non-repeat isolates was collected. Isolates were identified at the local laboratory, and again at the central laboratory for confirmation, using colonial morphology, routine biochemical tests and/or the Vitek system identification cards (biomérieux, Hazelwood, MO), as required. All isolates were stored at 80 C until MICs were measured. 2.2. Antimicrobial susceptibility testing The MICs of 11 antibiotics were determined for each isolate by agar dilution at the central laboratory according to Clinical and Laboratory Standards Institute (CLSI) guidelines [7]. The antimicrobials tested included meropenem (Sumitomo Pharmaceuticals Co., Osaka, Japan), imipenem (Sigma Chemical Co., St Louis, MO), ceftazidime (Sigma), cefotaxime (Sigma), ceftriaxone (Sigma), cefepime (Sigma), piperacillin/tazobactam (TZP) (Wyeth Pharmaceuticals, Collegeville, PA), cefoperazone/sulbactam (CSL) (2:1) (Sigma), cefotaxime/clavulanic acid (Sigma), cefoxitin (Sigma), amikacin (Sigma), ceftazidime/clavulanic acid (Sigma) and ciprofloxacin (Bayer AG, Leverkusen, Germany). The procedures for each set of tests were validated by determining the MICs for reference strains (Pseudomonas aeruginosa ATCC 27853, Escherichia coli ATCC 25922 and E. coli ATCC 35218), as recommended by CLSI standards. The results were interpreted according to CLSI breakpoints [8]. The cefoperazone MIC breakpoint was used for CSL as no breakpoint is available for this combination. The CLSI extended-spectrum -lactamase (ESBL) screening criterion (MIC 2 mg/l for either ceftazidime or cefotaxime) was applied to E. coli, Klebsiella spp. and Proteus mirabilis. ESBL production was confirmed using two pairs of drug combinations, cefotaxime/cefotaxime + clavulanic acid and ceftazidime/ceftazidime + clavulanic acid. An isolate was deemed to be ESBL-positive if the addition of clavulanic acid reduced the MIC of either of the -lactam agents by three-fold or more. Escherichia coli ATCC 25922 and Klebsiella pneumoniae ATCC 700603 were used as controls for the confirmatory ESBL test. 2.3. Data analysis For each isolate, the local laboratory recorded associated clinical information, including specimen type, collection date and medical specialty, on a standard case report form. All data, including MICs, were analysed by WHONET 5.4 software at the central laboratory. The prevalence of multidrug resistance was investigated among isolates of P. aeruginosa and Acinetobacter spp. Isolates of those species resistant to three or four antimicrobials among ceftazidime, meropenem or imipenem, amikacin and ciprofloxacin were considered to be multidrug-resistant (MDR). In the same species, pandrug resistance was defined as those isolates that were resistant to every antibacterial agent tested. 2.4. Repetitive extragenic palindromic sequence-based polymerase chain reaction (REP-PCR) typing REP-PCR using a method described previously [9] was employed to investigate epidemiological relationships among meropenemresistant Acinetobacter spp. isolated in 2008. The resulting patterns were examined visually. Identical profiles or highly similar profiles (up to two bands difference) were interpreted as indicating clonally identical or very closely related isolates [9].

Table 2 Cumulative percentage of each organism inhibited at different minimum inhibitory concentrations (MICs) by five antimicrobial agents. Organism Agent Cumulative percentage of isolates inhibited at MIC : Pseudomonas aeruginosa (n = 548) Acinetobacter baumannii (n = 454) 0.008 0.016 0.032 0.064 0.125 0.25 0.5 1 2 4 8 16 32 64 256 MEM 0.2 1.8 3.4 8.5 16.0 29.0 45.3 59.5 70.6 80.5 86.2 92.4 95.3 96.8 99.0 99.2 100 IPM 0 0 0.4 0.4 1.3 2.0 4.9 21.9 58.0 71.5 76.2 86.6 94.1 96.7 98.3 99.1 100 CAZ 0 0 0 0.2 0.7 0.9 2.4 22.5 47.5 62.3 70.6 74.6 83.9 87.9 91.2 93.8 100 TZP 0 0 0 0 0 0 0.9 3.8 20.4 43.8 52.7 62.6 69.0 73.7 84.8 97.0 100 CIP 0.2 0.7 4.9 13.7 36.5 48 61.7 67.9 71.2 78.5 84.3 88.1 94.3 99.6 100 100 100 MEM 0.4 0.6 4.4 6.9 12.3 32.4 48.5 61.7 72 75.3 78.4 83.0 90.1 95.5 99.7 99.7 100 IPM 0 0 0.2 0.4 3.7 31.1 44.3 53.5 72.1 77.2 79.8 83.4 89.5 97.2 99.7 100 100 CAZ 0 0.2 0.2 0.6 0.8 1.4 1.6 4.1 19.0 37.2 42.9 47.4 50.3 66.8 85.2 87.3 100 TZP 0 0 0.2 0.8 10.4 17.1 28.6 32.2 35.7 38.0 42.8 47.5 50.1 60.1 71.4 83.7 100 CIP 0.2 0.4 1 10.0 28.3 35.4 39.6 40.4 42.1 42.7 44.8 50.4 62.2 99.1 99.4 99.4 100 Escherichia coli (n = 590) MEM 4.1 57.1 84.6 93.3 96.2 98.2 98.9 99.4 99.6 99.7 100 100 100 100 100 100 100 IPM 0.2 0.4 0.7 1.0 5.4 59.5 90.0 96.8 98.3 99.3 99.3 99.6 99.7 100 100 100 100 CAZ 0 0.3 1.3 5.4 19.3 32.4 40.4 55.8 68.0 76.6 81.7 86.4 91.5 94.9 97.1 98.1 100 TZP 0 0 0 0 0 0.3 10.3 34.9 69.0 88.0 92.9 95.3 96.5 97.5 97.8 98.8 100 CIP 2 2.2 3.2 6.3 7.8 12.4 17.0 22.2 23.5 26.4 32.7 40.7 60.4 99.8 100 100 100 Klebsiella spp. (n = 531) MEM 0.8 32.8 80.9 91.7 94.2 95.3 96.4 97.1 98.0 98.3 98.8 99.6 99.6 99.6 99.8 99.8 100 IPM 0 0.4 0.4 3.2 40.6 80.9 82.4 96.4 97.2 98.5 98.9 99.1 99.6 99.7 99.8 99.8 100 CAZ 0 0 0.6 5.1 24.2 39.1 46.8 54.3 61.7 69.4 76.8 80.4 84.7 90.7 94.5 96.4 100 TZP 0 0 0 0 0 0.2 1 10.6 46.6 71.7 82.5 87.0 89.3 91.0 92.5 93.8 100 CIP 1.9 9.4 30.3 35.8 39.2 43.9 52.6 57.9 62.2 64.7 70.0 76.6 84.7 99.6 100 100 100 MEM, meropenem; IPM, imipenem; CAZ, ceftazidime; TZP, piperacillin/tazobactam; CIP, ciprofloxacin. 230 H. Wang et al. / International Journal of Antimicrobial Agents 35 (2010) 227 234

Table 3 Change in MIC 50 and MIC 90 values of each antimicrobial agent against Pseudomonas aeruginosa and Acinetobacter spp. Organism 2003 2004 2006 2008 %S MIC 50 MIC 90 MIC range %S MIC 50 MIC 90 MIC range %S MIC 50 MIC 90 MIC range %S MIC 50 MIC 90 MIC Range Pseudomonas aeruginosa MEM 86.2 0.5 8 0.008 to 81.8 0.5 16 0.016 to 79.1 1 16 0.032 to 76.0 1 32 0.032 to IPM 74.8 2 32 0.032 to 70.2 2 32 0.125 to 70.9 4 32 0.5 to 70.5 2 16 0.25 to CAZ 72.4 2 256 0.064 to 76.9 2 0.125 to 69.0 4 0.5 to 65.8 4 256 0.25 to FEP 71.5 4 32 0.032 to 64.5 4 64 0.5 to 62.7 8 64 0.5 to 69.9 4 32 0.125 to CSL 75.6 8 64 0.064 to 62.8 16 64 0.125 to 60.8 16 1 to 58.2 16 0.5 to TZP 74.0 4 256 0.5 to 76.9 4 256 0.5 to 74.1 16 256 1 to 70.5 8 256 1 to AMK 76.4 4 0.064 to 76.9 4 0.5 to 83.5 4 0.032 to 77.4 4 0.032 to CIP 72.4 0.25 16 0.032 to 64 63.6 0.5 32 0.004 to 64 65.8 0.25 32 0.002 to 64 69.9 0.25 32 0.016 to Acinetobacter spp. MEM 94.6 0.5 2 0.032 to 79.2 1 32 0.008 to 74.3 1 32 0.125 to 60.7 1 64 0.008 to IPM 92.5 0.5 2 0.064 to 80.2 1 32 0.125 to 79.9 1 32 0.25 to 256 62.1 1 64 0.032 to CAZ 46.2 16 1 to 43.6 16 0.064 to 41.7 64 1 to 41.4 64 0.016 to FEP 32.3 32 64 1 to 41.6 32 256 0.016 to 41.0 16 0.5 to 42.9 16 0.008 to 256 CSL 61.3 4 64 0.5 to 256 52.5 16 64 0.064 to 55.6 16 64 1 to 256 53.6 16 64 0.016 to TZP 53.8 16 256 0.032 to 52.5 16 0.032 to 43.8 64 0.125 to 43.6 64 0.125 to AMK 55.9 4 0.5 to 48.5 64 0.064 to 48.6 0.125 to 43.6 256 1 to CIP 44.1 8 64 0.032 to 40.6 16 64 0.016 to 64 39.6 32 64 0.004 to 64 38.6 32 64 0.032 to MIC 50/90, MIC for 50% and 90% of the organisms, respectively; MEM, meropenem; IPM, imipenem; CAZ, ceftazidime; FEP, cefepime; CSL, cefoperazone/sulbactam, TZP, piperacillin/tazobactam; AMK, amikacin CIP, ciprofloxacin. H. Wang et al. / International Journal of Antimicrobial Agents 35 (2010) 227 234 231

232 H. Wang et al. / International Journal of Antimicrobial Agents 35 (2010) 227 234 Table 4 Prevalence by year of multidrug resistance and pandrug resistance in Pseudomonas aeruginosa and Acinetobacter baumannii, and proportion of extended-spectrum -lactamase (ESBL)-producing Enterobacteriaceae. Organism % (no. of isolates) a 2003 2004 2006 2008 MDR P. aeruginosa b 11.3 (14/123) 14.9 (18/121) 13.3 (21/158) 17.1 (25/146) PDR P. aeruginosa c 2.4 (3/123) 2.5 (3/121) 5.1 (8/158) 7.5 (11/146) MDR A. baumannii 36.4 (32/88) 47.5 (47/99) 49.6 (69/139) 59.4 (76/) PDR A. baumannii 1.1 (1/88) 3.0 (3/99) 6.5 (9/139) 14.1 (18/) ESBL-producing Escherichia coli 55.8 (87/156) 50.9 (82/161) 66.7 (88/132) 61.0 (86/141) ESBL-producing Klebsiella spp. 42.4 (56/132) 25.4 (32/126) 35.9 (47/131) 36.6 (52/142) ESBL-producing Proteus mirabilis 13.0 (6/46) 8.9 (4/45) 24.0 (12/50) 24.2 (16/66) MDR, multidrug-resistant; PDR, pandrug-resistant. a Significant differences were observed between the years 2003 and 2008 for MDR A. baumannii ( 2 = 11.03, P < 0.01) and for PDR A. baumannii ( 2 = 10.98, P < 0.01), but not for other organisms. b Multidrug resistance for Gram-negative bacilli was defined as resistance to three or more of the following antimicrobials: ceftazidime; meropenem or imipenem; amikacin; and ciprofloxacin. c Pandrug-resistant isolates were defined as those that were resistant to every antibacterial agent tested. 2.5. Statistical analysis Fisher s exact test or 2 test was used to analyse qualitative variables, as appropriate. A P-value of <0.05 was considered statistically significant. 3. Results 3.1. Distribution of isolates The majority of isolates were recovered from respiratory tract specimens (1980/3892; 50.9%), followed by urinary tract (549/3892; 14.1%) and blood (349/3892; 9.0%); the remaining isolates were obtained from secretions (219/3892; 5.6%), pus (203/3892; 5.2%), drainage (148/3892; 3.8%), bile (115/3892; 3.0%), surgical wounds (81/3892; 2.1%) and other body sites. The most common organisms, in order of frequency, were E. coli (590/3892; 15.2%), P. aeruginosa (548/3892; 14.1%), Klebsiella spp. (531/3892; 13.6%), Enterobacter spp. (522/3892; 13.4%), Acinetobacter spp. (486/3892; 12.5%), Proteus spp. (331/3892; 8.5%), Citrobacter spp. (254/3892; 6.5%) and Serratia spp. (195/3892; 5.0%). 3.2. Antimicrobial activity against major organisms from 2003 to 2008 Antimicrobial susceptibility data for the major organisms are presented in Table 1. Meropenem and imipenem were consistently the most active agents against most of the isolates [>95% susceptible by CLSI criteria, except among P. aeruginosa and Acinetobacter spp., which exhibited lower frequencies of susceptible isolates (71.5 80.5% and 75.3 77.2%, respectively)]. Except for Acinetobacter spp., the next most active agent was amikacin, with susceptibility from 78.8% to 93.3%, followed by TZP (73.7 98.2%), CSL (63.9 99.1%), cefepime (67.0 95.4%) and ceftazidime (54.5 93.3%). Cefoxitin, in general, was observed to have the highest resistance rates among all of the agents tested. Proteus spp. exhibited high rates of susceptibility to most agents (>85%, except for ciprofloxacin and ceftriaxone), with a similar pattern observed in Serratia spp. (>75%, except for cefoxitin). Among Acinetobacter spp. isolates, only 40.4% were susceptible to ciprofloxacin, 47.5% to TZP and 48.5% to amikacin. In contrast, among P. aeruginosa isolates the susceptible percentages for the above three agents were 67.9%, 73.7% and 78.8%, respectively. TZP, amikacin and ceftazidime remained active against E. coli, with susceptible rates >80%. Table 2 lists the cumulative percent inhibition of four organism groups at different MICs by five antimicrobial agents. More than 95% of E. coli and Klebsiella spp. were inhibited by meropenem or imipenem at a concentration of 1 mg/l, and >80% of isolates were inhibited by TZP or ceftazidime at a concentration of 16 mg/l. However, only 22.2% of E. coli and 57.9% of Klebsiella spp. were inhibited by ciprofloxacin at 1 mg/l, which is the susceptible breakpoint. Nearly 70% of the E. coli isolates that were resistant to ciprofloxacin displayed a ciprofloxacin MIC of 64 mg/l. Among P. aeruginosa and Acinetobacter spp., >70% of isolates were inhibited by meropenem or imipenem concentrations of 4 mg/l. Meropenem exhibited a slightly higher susceptible rate than imipenem in P. aeruginosa (80.5% vs. 71.5%). Similar to the ciprofloxacin pattern in E. coli, nearly 70% of ciprofloxacin-resistant Acinetobacter spp. displayed a ciprofloxacin MIC of 64 mg/l; however, for P. aeruginosa, ca. 70% of isolates displayed an MIC of 1 mg/l and 99.6% isolates were inhibited by ciprofloxacin at 64 mg/l. 3.3. Changes in antimicrobial susceptibility of the five most common organisms 3.3.1. Pseudomonas aeruginosa and Acinetobacter spp. In general, susceptibility profiles of P. aeruginosa for most agents remained relatively stable over the course of the study. Meropenem, imipenem, TZP and amikacin maintained high activity against P. aeruginosa, with >70% of isolates being susceptible over the study period. Acinetobacter baumannii accounted for 93.4% of Acinetobacter spp. isolates (454/486). In contrast to the relatively stable susceptibility patterns for P. aeruginosa, a decline in susceptibility to meropenem and imipenem was observed in Acinetobacter spp. Specifically, susceptibility to meropenem decreased significantly from 94.6% in 2003 to 60.7% in 2008 ( 2 = 33.93, P < 0.01). Over the same period, the MIC 50 and MIC 90 values (MIC for 50% and 90% of the organisms, respectively) increased from 0.5 mg/l and 2 mg/l, respectively, to 1 mg/l and 64 mg/l. In the case of imipenem, the susceptibility rate declined from 92.5% to 62.1% ( 2 = 27.32, P < 0.01) and the MIC 50 and MIC 90 increased from 0.5 mg/l and 2 mg/l, respectively, to 1 mg/l and 64 mg/l (Table 3). Increases in rates of resistance were also revealed for other agents, such as CSL, TZP and amikacin. The prevalence of multidrug resistance in P. aeruginosa and A. baumannii from 2003 to 2008 is presented in Table 4. The prevalence of multidrug resistance increased in both species over the period of the study and in the case of A. baumannii this increase was statistically significant ( 2 = 11.03, P < 0.01). A similar trend was also observed in the prevalence of A. baumannii isolates resistant to all the tested agents (pandrug resistance), which increased from 1.1% to 14.1% ( 2 = 10.98, P < 0.01). Pan-resistance also occurred in

H. Wang et al. / International Journal of Antimicrobial Agents 35 (2010) 227 234 233 P. aeruginosa over the study time period, but this did not reach the level of significance. All 55 meropenem-resistant Acinetobacter spp. isolated in 2008 were investigated by REP-PCR and five REP-PCR types were identified. The most prevalent clone was REP-PCR type 1, which accounted for 61.5% of all isolates and was distributed across five hospitals. The next most common clone was type 3 (15.4%). 3.3.2. Escherichia coli, Klebsiella spp. and Enterobacter spp. During the 6-year study period, susceptibility profiles for E. coli and Klebsiella spp. remained stable for all agents (data not shown). In general, meropenem and imipenem were the most active agents against E. coli and Klebsiella spp. From 2003 to 2008, 98.7 100% of E. coli and 96.5 100% of Klebsiella spp. were susceptible to carbapenems. However, the MIC 50 and MIC 90 of imipenem against the combined E. coli isolates and the combined Klebsiella spp. isolates were four- to eight-fold higher than those of meropenem (Table 1). TZP was the second most active agent, against which 89.7 97.9% of E. coli and 82.5 90.8% of Klebsiella spp. isolates were susceptible. Amikacin, CSL, cefepime and ceftazidime also showed stable activity against these species (susceptibility remaining >70%). However, the percent susceptibility of E. coli to cefotaxime, ceftriaxone and ciprofloxacin was relatively low. Compared with E. coli, Klebsiella spp. were more susceptible to cefotaxime and ciprofloxacin. The percent susceptibilities to ceftazidime in E. coli and Klebsiella spp. were much higher than to cefotaxime and ceftriaxone. The frequencies of ESBL phenotypes in E. coli, Klebsiella spp. and P. mirabilis were 50.9 66.7%, 25.4 42.4% and 8.9 24.2%, respectively (Table 4). An ESBL phenotype was more frequently observed in E. coli than in Klebsiella spp. or P. mirabilis. The ESBL prevalence in these species varied from year to year but without any discernible trend to increase or decrease over time. Antimicrobial resistance in the combined Enterobacter spp. was higher than in E. coli and Klebsiella spp. (Table 1). Meropenem and imipenem retained excellent activity against Enterobacter spp. (>99% of isolates susceptible). Amikacin was the second most active agent (83.9% susceptible), followed by cefepime, TZP and CSL (Table 1). 4. Discussion The antimicrobial susceptibilities of Gram-negative bacteria in China from 2003 to 2008 remained relatively stable for many organisms. However, the overall rates of resistance in the majority of the organisms had increased compared with rates in a previous time period [10]. During the 6-year period, meropenem and imipenem continued to show potent and consistent activity against Gramnegative bacteria, including ESBL-producing organisms. With the exception of Acinetobacter spp. and P. aeruginosa, meropenem and imipenem MIC 50 and MIC 90 values remained stable, i.e. no MIC creep was observed. This finding is consistent with the results of other studies undertaken in Europe and the USA [11,12]. The most surprising finding in this study was the rather dramatic decrease in the percent of Acinetobacter spp. susceptible to meropenem or imipenem during the study period. This decrease was common across many of the cities surveyed, indicating that resistance of Acinetobacter spp. to meropenem and imipenem has become a widespread and serious problem in China. Clearly, prevention and reversal of this trend is a key public health need. Note that the trend was not confined to the carbapenems, as percent susceptibilities to CSL, TZP, amikacin and ciprofloxacin in Acinetobacter spp. decreased by 15 25% compared with a previous study [10]. The REP-PCR data for Acinetobacter spp. isolated in 2008 indicated that clonal spread was the main reason for the decrease in percent susceptible. In China, clones carrying the bla OXA-23 carbapenemase gene have previously been shown to account for high carbapenem resistance [13,14]. Thus, early recognition of the presence of carbapenem-resistant Acinetobacter spp. clones is necessary in order to prevent their spread within the hospital environment and thus eventually decrease the overall resistant rates. Compared with Acinetobacter spp., rates of resistance in P. aeruginosa were relatively stable. Amikacin, TZP and ceftazidime continued to be effective agents, but susceptibilities to TZP and CSL were ca. 10% lower than previously [10]. Unlike the decreasing susceptibility to ciprofloxacin in Acinetobacter spp., the percent susceptibility to ciprofloxacin in P. aeruginosa was stable. The rates of susceptibility in P. aeruginosa presented here are much lower than those in studies in other regions of the world. For example, in Europe the percent susceptibility to TZP in P. aeruginosa was recently reported to be 86.8% [11], in the USA it was 86.6% [12] and in Canada it was 90% [15]. Similar differences occurred with other antibacterial agents. An underlying serious problem that we have observed is a continuous increase in MDR isolates both of Acinetobacter spp. and P. aeruginosa. In this study, MDR isolates increased annually. Indeed, in 2008 the multidrug resistance rate among Acinetobacter spp. was 51% of isolates, much higher than the rate in other countries. Although carbapenems remain one of the choices for infections with Acinetobacter and P. aeruginosa, the continuous increase in MDR organisms will result in more complicated treatment challenges and increased healthcare burdens. Similar to a previous report [10], rates of resistance in E. coli and Klebsiella spp. remained stable, but with a high percentage of ESBL-producing isolates (Table 4). Increasing frequencies of ESBLproducing Enterobacteriaceae have become a challenging problem worldwide, occurring in Europe and the USA, but at lower percentages than observed in our study [11,12,15 17]. The high percentage of ESBL in China may be a consequence of the selective pressure of cefotaxime and ceftriaxone, which have been widely and predominantly used in Chinese hospitals in recent years. The results of this study would undermine the rationale for the use of extendedspectrum cephalosporins or penicillins as empirical therapy for severe infections caused by Gram-negative pathogens in China owing to the high prevalence of ESBL. Compared with a previous study [10], the consistent downward trend in ciprofloxacin activity against E. coli is also a severe concern in China. Now only ca. 20% of E. coli isolates are susceptible to ciprofloxacin, which indicates that ciprofloxacin would not be suitable for the treatment of E. coli infections, especially in the intensive care unit. Decreased activity of ciprofloxacin has also been reported in many other countries [11,12,15 17]. For example, a 12-year study in the USA indicated that the rate of ciprofloxacin resistance in E. coli increased from 0.9% in 1993 to 17.3% in 2004 [11]. The molecular mechanisms of fluoroquinolone resistance in Enterobacteriaceae in China include multiple substitutions in the gyra and parc genes as well as the plasmid-mediated quinolone-resistance genes qnr and aac(6 )-Ib-cr [18]. In conclusion, the susceptibility profiles for many organisms have remained constant over the past 6 years, but a sharp decrease in the activity of meropenem or imipenem against Acinetobacter spp. has been observed. The information gathered in this study can be used in preparing evidence-based guidelines for treatment, especially empirical treatment, of bacterial infections in China. Moreover, it is clear that clonal spread of MDR Acinetobacter spp. and P. aeruginosa must be controlled as a serious public health issue. Acknowledgment Editorial and style assistance from Dr Wright W. Nichols is acknowledged.

234 H. Wang et al. / International Journal of Antimicrobial Agents 35 (2010) 227 234 Funding: The CMSS project was funded by a research grant from Sumitomo Pharmaceuticals (Suzhou) Co., Ltd. It was partially supported by the National Natural Science Foundation of China (grant 30500023). Competing interests: None declared. Ethical approval: Not required. References [1] Paterson DL. Resistance in Gram-negative bacteria: Enterobacteriaceae. Am J Infect Control 2006;34(5 Suppl. 1):S20 8 [Discussion S64 73]. [2] Peleg AY, Paterson DL. Multidrug-resistant Acinetobacter: a threat to the antibiotic era. Intern Med J 2006;36:479 82. [3] Livermore DM. Multiple mechanisms of antimicrobial resistance in Pseudomonas aeruginosa: our worst nightmare? Clin Infect Dis 2002;34:634 40. [4] Livermore DM, Woodford N. The -lactamase threat in Enterobacteriaceae, Pseudomonas and Acinetobacter. Trends Microbiol 2006;14:413 20. [5] Felmingham D. The need for antimicrobial resistance surveillance. J Antimicrob Chemother 2002;50(Suppl. S 1):1 7. [6] Jones RN, Mendes C, Turner PJ, Masterton R. An overview of the Meropenem Yearly Susceptibility Test Information Collection (MYSTIC) Program: 1997 2004. Diagn Microbiol Infect Dis 2005;53:247 56. [7] National Committee for Clinical Laboratory Standards. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically; approved standard. 6th ed. Document M7-A6. Wayne, PA: NCCLS; 2003. [8] Clinical and Laboratory Standards Institute. Performance standards for antimicrobial susceptibility testing. 18th ed. Document M100-S18. Wayne, PA: CLSI; 2008. [9] Snelling AM, Gerner-Smidt P, Hawkey PM, Heritage J, Parnell P, Porter C, et al. Validation of use of whole-cell repetitive extragenic palindromic sequence-based PCR (REP-PCR) for typing strains belonging to the Acinetobacter calcoaceticus Acinetobacter baumannii complex and application of the method to the investigation of a hospital outbreak. J Clin Microbiol 1996;34:1193 202. [10] Wang H, Chen M; China Nosocomial Pathogens Resistance Surveillance Study Group. Surveillance for antimicrobial resistance among clinical isolates of Gram-negative bacteria from intensive care unit patients in China, 1996 to 2002. Diagn Microbiol Infect Dis 2005;51:201 8. [11] Nørskov-Lauritsen N, Marchandin H, Dowzicky MJ. Antimicrobial susceptibility of tigecycline and comparators against bacterial isolates collected as part of the TEST study in Europe (2004 2007). Int J Antimicrob Agents 2009;34:121 30. [12] Lockhart SR, Abramson MA, Beekmann SE, Gallagher G, Riedel S, Diekema DJ, et al. Antimicrobial resistance among Gram-negative bacilli causing infections in intensive care unit patients in the United States between 1993 and 2004. J Clin Microbiol 2007;45:3352 9. [13] Wang H, Guo P, Sun H, Wang H, Yang Q, Chen M, et al. Molecular epidemiology of clinical isolates of carbapenem-resistant Acinetobacter spp. from Chinese hospitals. Antimicrob Agents Chemother 2007;51:4022 8. [14] Zhou H, Yang Q, Yu YS, Wei ZQ, Li LJ. Clonal spread of imipenem-resistant Acinetobacter baumannii among different cities of China. J Clin Microbiol 2007;45:4054 7. [15] Zhanel GG, DeCorby M, Laing N, Weshnoweski B, Vashisht R, Tailor F, et al. Antimicrobial-resistant pathogens in intensive care units in Canada: results of the Canadian National Intensive Care Unit (CAN-ICU) study, 2005 2006. Antimicrob Agents Chemother 2008;52:1430 7. [16] Jones RN, Kirby JT, Rhomberg PR. Comparative activity of meropenem in US medical centers (2007): initiating the 2nd decade of MYSTIC program surveillance. Diagn Microbiol Infect Dis 2008;61:203 13. [17] Turner PJ. MYSTIC Europe 2007: activity of meropenem and other broadspectrum agents against nosocomial isolates. Diagn Microbiol Infect Dis 2009;63:217 22. [18] Yang H, Chen H, Yang Q, Chen M, Wang H. High prevalence of plasmidmediated quinolone resistance genes qnr and aac(6 )-Ib-cr in clinical isolates of Enterobacteriaceae from nine teaching hospitals in China. Antimicrob Agents Chemother 2008;52:4268 73.