Antimicrobial susceptibility of clinical isolates from earthquake victims in Wenchuan

ORIGINAL ARTICLE 10.1111/j.1469-0691.2008.02129.x Antimicrobial susceptibility of clinical isolates from earthquake victims in Wenchuan M. Kang 1,2, Y. Xie 1, C. Mintao 1, Z. Chen 1, H. Chen 1, H. Fan 1, W. Chen 1 and X. Guo 1 1) Division of Clinical Microbiology, Department of Laboratory Medicine and 2) Department of Microbiology, College of Basic Medicine and Forensic Medicine, Sichuan University, Chengdu, China Abstract On 12 May 2008, an earthquake measuring 8.0 on the Richter scale struck Wenchuan County, Sichuan, China. Between 12 May and 11 June, 1823 victims were hospitalized in West China Hospital. These patients were severely injured, and most of their wounds were contaminated. Here, the results of bacteriological identification and antibiotic susceptibility testing of 725 non-duplicate isolates from earthquake victims are presented. Gram-negative bacilli were most frequently isolated (71.3%). Only 18.9% of isolates were Gram-positive bacteria; Candida spp. accounted for 9.7%, and Gram-negative cocci for 0.1%. After anaerobic culture, four Clostridium sordellii strains and one Clostridium bifermentans strain were isolated from deep wounds. Specimen culture from earthquake victims revealed a spectrum of pathogens and antibiotic susceptibilities that was different from that usually encountered in West China Hospital, especially concerning methicillin-resistant Staphylococcus aureus, extended-spectrum b-lactamase producers, and multidrug-resistant (MDR) non-fermenting Gram-negative bacilli. The pathophysiology of the injuries in earthquake victims was different from that in the patients who were not earthquake victims. A combination of environmental bacteria with a high proportion of Gram-negative bacteria was often observed in the earthquake victims. Approximately 26% of all earthquake victims were shown to be carriers of MDR microorganisms. Therefore, appropriate microbiological assessment upon admission, and identification of patients to be put in quarantine, is of paramount importance. Keywords: China, drug resistance, earthquake, nosocomial infection, quarantine, susceptibility testing Original Submission: 30 June 2008; Revised Submission: 18 September 2008; Accepted: 26 September 2008 Editor: D. Raoult Clin Microbiol Infect 2009; 15: 87 92 Corresponding author and reprint requests: Y. Xie, Division of Clinical Microbiology, Department of Laboratory Medicine, West China Hospital, Sichuan University, Chengdu, Sichuan Province 610041, China E-mail: xie_yi_77@sina.com Introduction On 12 May, at 14:28 Beijing time, an earthquake measuring 8.0 on the Richter scale struck Wenchuan County, Sichuan Province, China. The earthquake disaster caused more than 374 171 injuries and 69 185 deaths, and 18 467 people were missing in southwestern China (http://news.xinhuanet.com/ newscenter/). Victims who survived the earthquake were often severely injured, and because of the insufficient supply of medical workers, surgical equipment and antibiotics in the earthquake-stricken areas, the incidence of wound infections was high [1,2]. From 12 May to 11 June 2008, c. 2695 earthquake victims were transferred to West China Hospital, and 1823 victims were hospitalized. The most common problems among these patients were crush syndrome and traumatic wounds. Under such circumstances, information about infection-causing bacteria and the antimicrobial susceptibilities of isolates from earthquake victims becomes very important. This information will be useful to clarify whether the isolation rate of multidrug-resistant (MDR) microorganisms will change after the earthquake. In this study, the results of bacteriological examination of Wenchuan earthquake victims were reviewed, and the pattern of antibiotic susceptibility of isolates from earthquake victims was compared with that of isolates from other patients. Materials and Methods Patients and methods Upon admission, the clothes of all victims were removed; the patients received hospital gowns in an isolated area. All wounds were systematically examined and documented. The Journal Compilation ª2009 European Society of Clinical Microbiology and Infectious Diseases

88 Clinical Microbiology and Infection, Volume 15 Number 1, January 2009 CMI diagnosis of wound infection was made by a physician, based on the presence of clinical signs. Swabs for culture were taken from all wounds; urine, blood or sputum samples were taken only when indicated. Normal cultures, blood cultures, anaerobic cultures and vascular catheter cultures were performed following the CLSI guidelines [3]. Bacterial identification and antimicrobial susceptibility testing The MicroScan Walkaway 96SI system (Dade Behring Diagnostics Co., Ltd, Sacramento, CA, USA) and the PHOENIX Automated Microbiology System (BD Diagnostics, Franklin Lakes, NJ, USA) were used for the identification of microorganisms and susceptibility testing following the performance standards for antimicrobial susceptibility testing as defined by the CLSI [4]. If necessary, biochemical methods were used to identify strains isolated from clinical samples. Phenotypic confirmatory tests for extended-spectrum b-lactamase (ESBL) production in Klebsiella pneumoniae, Klebsiella oxytoca and Escherichia coli were carried out following the CLSI guidelines [4]. For the isolation of anaerobic strains, GENbag anaer (Biomerieux, Marcy l Etoile, France) and API 20A strips (Biomerieux) were used to culture and identify anaerobic strains. If necessary, the anaerobic identification test was carried out more than once. Antibiotic susceptibility testing data analysis and statistical methods Data from antimicrobial susceptibility testing were analysed with Whonet 5.4 software from the clinical microbiology laboratory information system (http://www.who.int/ drugresistance/whonetsoftware). Duplicate isolates of the same organism from the same patient were eliminated using the same software. Antibiotic resistance patterns of each organism from Wenchuan earthquake victims were analysed and compared with those of organisms from other patients admitted before the earthquake. A two-sided chi-square test was used to test differences in resistance rates for isolates from earthquake victims and other patients. A p-value 0.05 was taken as being statistically significant. Results The mean age of the patients (939 males, 844 females) was 45.2 years (80 days to 105 years). Between 12 May and 11 June 2008, 2002 specimens from earthquake victims were collected, cultured and tested. The most common specimens were wound swabs (1243, 62%), blood (600, 30%), sputum (102, 5%), vascular catheters (45, 2.2%) and urine (12, 0.5%). In total, 74 anaerobic cultures were taken from deep wounds. Altogether, 725 non-duplicate clinical isolates from earthquake victims were identified and analysed, and compared with 690 isolates from pre-earthquake patients. Gram-negative bacilli were the most commonly isolated pathogens (517 isolates (71.3%)), including Acinetobacter baumannii, E. coli, Pseudomonas aeruginosa, Enterobacter cloacae, K. pneumoniae, and Aeromonas hydrophila (Table 1). Only 18.9% of the isolates were Gram-positive bacteria. The most frequently isolated Gram-positive bacteria were Staphylococcus aureus (90 isolates (12.4%)) (Table 1). Other Gram-positive cocci were enterococci (29 isolates) and coagulase-negative staphylococci (16 isolates). The remaining species were Candida spp. (9.7%) and Gram-negative cocci (0.1%). Regarding anaerobic bacteria, four Clostridium sordellii isolates and one Clostridium bifermentans isolate were recovered after culture from deep wounds of five earthquake victims. The culture of specimens from earthquake victims revealed a spectrum of pathogens that was different from that usually found in the same hospital. In the 2 weeks after the Wenchuan earthquake, the number and relative rate of isolation of Aeromonas hydrophila isolates increased most notably, but they also increased with respect to Candida albicans and, to a lesser degree, S. aureus, whereas they decreased somewhat with respect to K. pneumoniae (Table 1). Antibiotic susceptibility results concerning the six organisms most frequently isolated from the earthquake victims are summarized in Fig. 1, which gives the organisms, the antimicrobial agents tested, and an indication of the trends observed over the study period. Methicillin-resistant TABLE 1. Microorganisms isolated from earthquake victims and patients who were not earthquake victims at West China Hospital, Chengdu Organism No. of isolates from earthquake victims (ranking no.) No. of isolates from patients who were not earthquake victims (ranking no.) 12 May to 12 May to 12 April to 31 May 2008 11 June 2008 11 May 2008 Acinetobacter baumannii 95 (1) 130 (1) 107 (1) Escherichia coli 88 (2) 119 (2) 94 (2) Staphylococcus aureus 81 (3) 90 (3) 66 (5) Pseudomonas aeruginosa 43 (4) 67 (4) 86 (3) Enterobacter cloacae 43 (5) 64 (5) 36 (6) Klebsiella pneumoniae 23 (6) 47 (6) 68 (4) Aeromonas hydrophila 14 (7) 14 (11) 1 (36) Candida albicans 13 (8) 43 (7) 6 (20) Stenotrophomonas (Xanthomonas) maltophilia 12 (9) 22 (8) 23 (8)

CMI Kang et al. Resistance of pathogens from earthquake victims 89 FIG. 1. Antimicrobial resistance patterns of the six most frequently isolated organisms during the 1-month periods before (bright green columns) and after (purple columns) the earthquake in West China Hospital. (a) Acinetobacter baumannii; (b) Pseudomonas aeruginosa; (c) Escherichia coli; (d) Enterobactercloacae; (e) Klebsiella pneumoniae; (f) Staphylococcus aureus. S. aureus (MRSA), ESBL-producing bacteria and MDR non-fermenting Gram-negative bacilli were common causes of severe nosocomial infections. The decrease in isolation rates of MRSA was obvious (Fig. 2). Fig. 3 shows the prevalence of ESBL production in E. coli, K. pneumoniae and K. oxytoca. The rate of isolation of MDR non-fermenting Gram-negative bacilli showed a discrete rise (Fig. 4). Discussion The characteristics of the injuries of earthquake victims are different from those of other, average, patients. Most earthquake victims are physically injured, and have wound contamination and progressing tissue damage. Aside from the

90 Clinical Microbiology and Infection, Volume 15 Number 1, January 2009 CMI FIG. 1. Continued environmental factors present at the time of injury, prior surgical and antibiotic treatment influence, to a great extent, the development of wound infections [5,6]. In earthquake-hit areas, owing to the large number of patients, insufficient numbers of medical workers and the scarcity of surgical equipment and antibiotics, wounds are usually heavily contaminated. Many patients rapidly develop foul-smelling wounds with abundant pus; meanwhile, empirical antibiotic treatment is given without knowledge of the infecting species. Under such circumstances, a combination of environ-

CMI Kang et al. Resistance of pathogens from earthquake victims 91 FIG. 2. Isolation rates of methicillin-resistant Staphylococcus aureus (MRSA) before and after the earthquake. FIG. 3. Rates of isolation of extended-spectrum b-lactamase (ESBL)- producing Escherichiacoli, Klebsiella pneumoniae and Klebsiella oxytoca before (12 4 11 5, bright green columns) and after (12 5 11 6, purple columns) the earthquake. FIG. 4. Rates of isolation of multidrug-resistant (MDR) Pseudomonas aeruginosa and Acinetobacter baumannii before (12 4 11 5, bright green columns) and after (12 5 11 6, purple columns) the earthquake. mental bacteria (including Aeromonas hydrophila) and a high prevalence of Gram-negative bacteria (such as A. baumannii) is often found in cultures from wounds and other specimens. Aeromonas hydrophila, present in earth and fresh water, has previously been reported to cause rapidly progressing wound infections [7,8]. In West China Hospital, only one Aeromonas sp. isolate was recovered from patients who were not earthquake victims between 12 April and 11 May 2008, and the isolation rate was only c. 0.14%. However, 14 Aeromonas sp. isolates were recovered from earthquake victims in the first 2 weeks after the earthquake, which represents a significant increase (p 0.001) in the rate of isolation. After 48 h of hospitalization, most wound-infecting pathogens are MDR [9]. This observation indicates that wound infection may be a significant factor associated with nosocomial infections. Acinetobactet baumannii (A. baumannii) is an increasingly important opportunistic pathogen and a relatively common cause of nosocomial infections in areas of natural disaster or war [1,10 12]. The treatment of A. baumannii infections has become increasingly difficult, because of the development of multiple antibiotic resistance [13]. In the present study, A. baumannii isolates were resistant to cefepime, cefotaxime, ceftazidime, ciprofloxacin, gentamicin and amikacin, but they were mainly susceptible to imipenem, with a susceptibility rate of 70% (Fig. 1a). The imipenem resistance rate in P. aeruginosa isolates from patients who were not earthquake victims was higher than in those from earthquake victims. Approximately half of the P. aeruginosa isolates from earthquake victims were resistant to cefepime and ciprofloxacin. The rates of resistance to gentamicin, amikacin, ceftazidime, piperacillin, piperacillin tazobactam and imipenem were relatively low (Fig. 1b). All E. coli, Enterobacter cloacae and K. pneumoniae isolates were usually susceptible to piperacillin tazobactam, imipenem and amikacin (Fig. 1c e). However, the Enterobacter cloacae isolation rate increased from 5.2% in patients who were not earthquake victims (12 April to 11 May) to 8.8% in earthquake victims, probably mostly due to the bacterium s association with environmental matter, including dust. Among all organisms analysed, the antibiotic resistance rate increased most notably in Enterobacter cloacae, with an average increase of 9.9% (Fig. 1d). The trimethoprim sulphamethoxazole resistance rate increased by 17% in K. pneumoniae isolates from earthquake victims, and to 59.6%, up from 42.6%, in isolates from other patients. The gentamicin and cefotaxime resistance rates of K. pneumoniae were higher in isolates from earthquake victims than in isolates from other patients (Fig. 1e). Although S. aureus is the most common microorganism causing wound infections [14], the isolation rates of A. baumannii and E. coli ranked first and second in earthquake victims. Isolates resistant to oxacillin were significantly less common (50%) in earthquake victims than in other patients (p 0.001) (Figs 1f and 2).

92 Clinical Microbiology and Infection, Volume 15 Number 1, January 2009 CMI Imipenem, piperacillin tazobactam and amikacin appeared to be effective against most causative pathogens, except A. baumannii. On the basis of the drug susceptibility results, treatment with only second-generation cephalosporins, fluoroquinolones and trimethoprim sulphamethoxazole appeared to be inadequate. Fourth-generation cephalosporins, e.g. cefepime, were not particularly effective against A. baumannii and E. coli, confirming observations made by Geertruid et al.[15] after the tsunami disaster. In the present study, c. 26% of all earthquake victims were shown to be carriers of MDR microorganisms, including MRSA and ESBL producers. Additionally, in several cases, MDR Gram-negative bacteria were isolated from the same patient or ward. This might reflect the risk of nosocomial infection. Further studies are needed to address the relationship between isolates from earthquake victims and nosocomial infection. The results of this study show that the rates of isolation of the environmental organism Aeromonas hydrophila and S. aureus clearly increased in earthquake victims, and that most of the isolates were susceptible to antibiotics. On the other hand, strains of the species A. baumnnii, E. coli, P. aeruginosa and K. pneumoniae showed a high level of resistance to frequently used antibiotics, and the numbers of ESBL producers and MDR bacteria increased slightly (Figs 3 and 4). In conclusion, thorough microbiological assessment of earthquake victims upon admission is extremely important, and quarantine should be considered for patients infected, or suspected to be infected, with ESBL producers, MDR nonfermenting Gram-negative bacilli and MRSA. Acknowledgements We thank D. Lin and L. Qin for discussing the manuscript and editing it. We also thank P. Cai and Y. Bai for contributing to the ideas and statistical analysis. References 1. Hiransuthikul N, Tantisiriwat W, Lertutsahakul K, Vibhagool A, Boonma P. Skin and soft-tissue infections among tsunami survivors in southern Thailand. Clin Infect Dis 2005; 41: 93 96. 2. Centers for Disease Control and Prevention. Rapid health response, assessment, and surveillance after a tsunami Thailand, 2004 2005. MMWR 2005; 54: 61 64. 3. Clinical Laboratory Standards Institute. Principle and procedures for blood culture; proposed guideline M47-P. Wayne, PA: CLSI, 2007. 4. Clinical and Laboratory Standards Institute. Performance standards for antimicrobial susceptibility testing, 17th informational supplement. Wayne, PA: CLSI, 2007. 5. Eron LJ, Lipsky BA, Low DE, Nathwani D, Tice AD, Volturo GA. Managing skin and soft tissue infection: expert panel recommendations on key decision points. J Antimicrob Chemother 2003; 52 (suppl): 13 17. 6. Kazancioglu R, Cagatay A, Calangu S et al. The characteristic of infections in crush syndrome. Clin Microbiol Infect 2002; 8: 202 206. 7. Vally H, Whittle A, Cameron S, Dowse GK, Watson T. Outbreak of Aeromonas hydrophila wound infections associated with mud football. Clin Infect Dis 2004; 38: 1084 1089. 8. Semel JD, Trenholme G. Aeromonas hydrophila water-associated traumatic wound infections: a review. J Trauma 1990; 30: 324 327. 9. Siegel JD, Rhinehart E, Jackson M, Chiarello L, The health-care infection control practices advisory committee. Management of multidrugresistant organisms in health-care settings. Atlanta, GA: Centers for Disease Control and Prevention, 2006. 10. Scott PT, Petersen K, Fishbain J et al. Acinetobacter baumannii infections among patients at military medical facilities treating injured US service members, 2002 2004. JAMA 2004; 292: 2964 2966. 11. Jane F, Turton M, Kaufmann E et al. Comparison of Acinetobacter baumannii isolates from the United Kingdom and the United States that were associated with repatriated casualties of the Iraq conflict. J Clin Microbiol 2006; 44: 2630 2634. 12. Joly GM. Clinical impact and pathogenicity of Acinetobacter. Clin Microbiol Infect 2005; 11: 868 873. 13. Uckay I, Sax H, Harbarth S, Bernard L, Pittet D. Multi-resistant infections in repatriated patients after natural disasters: lessons learned from the 2004 tsunami for hospital infection control. J Hosp Infect 2008; 68: 1 8. 14. Bowler PG, Duerden BI, Armstrong DG. Wound microbiology and associated approaches to wound management. Clin Microbiol Rev 2001; 14: 244 269. 15. Geertruid M, Marres H, Dylan W et al. Wound infections in repatriated survivors of the tsunami disaster. Wounds 2006; 18: 92 100. Transparency Declaration The authors declare that they have no conflicting interests.