Clinical Characteristics, Antimicrobial Susceptibilities ...

1 520*Corresponding author: Mailing address: Department ofCritical Care Medicine, Tainan Municipal Hospital, , Chung-Te Road, East District, Tainan, Taiwan : 886-6-2609926 ext. 21396, Fax: 886-6-2606351,E-mail: choudw J. Infect. Dis., 64, 520-524, 2011 Short CommunicationClinical Characteristics, Antimicrobial Susceptibilities ,andOutcomesofPatientswi thChryseobacterium indologenesBacteremia in an Intensive Care UnitDeng-Wei Chou1,2*, Shu-Ling Wu2, Chao-Tai Lee3, Fan-Ting Tai4,andWen-LiangYu5,61 Department of Critical Care Medicine,3 Department of Clinical Laboratory, and4 Committee of Infection Control, Tainan Municipal Hospital, Tainan;2 Department of Nursing, Chung-Hwa University of Medical Technology, Tainan;5 Department of Intensive Care Medicine, Chi-Mei Medical Center, Tainan; and6 Department of Medicine, Taipei Medical University, Taipei, Taiwan(Received June 8, 2011. Accepted September 2, 2011)SUMMARY: Ten patients with intensive care unit (ICU)-acquiredChryseobacterium indologenesbac-teremia between January 2004 and December 2008 were studied.

2 The primary site of infection wasunknown for 80zof the cases. The known primary sites of infection were empyema (10z)andcatheter-related bacteremia (10z). Eight patients (80z) had polymicrobial bacteremia, spent more than21 days in the ICU, and received more than 14 days of broad-spectrum antibiotic therapy prior to theonset ofC. indologenesbacteremia. All isolates were 100zsusceptible to minocycline andtrimethoprim/sulfamethoxazole. Vancomycin, imipenem, piperacillin/tazobactam, ciprofloxacin, andlevofloxacin exhibited 0z,10z,20z,30z,and30z, respectively, susceptibility against this patho-gen. All isolates were 100zresistant to ceftazidime, cefepime, meropenem, piperacillin, and 14-day mortality rate was 40z. Our findings suggest that this pathogen should be included amongthe causes of ICU-acquired bacteremia, especially in patients with a prolonged stay in an ICU or whohad received long-term broad-spectrum antibiotic therapy. Extended-spectrum penicillins, third- andfourth-generation cephalosporins, and quinolones had very little or no effect against this , choosing an appropriate antibiotic therapy for this pathogen is very indologenesis a non-motile, ox-idase-positive, indole-positive, and glucose-nonfermen-tative Gram-negative rod that is widely distributed innature (1).

3 The microorganism had not been reported asa cause of bacteremia in humans until 1996. Hsueh et 12 patients withC. indologenesbacteremiaover a 3-year period (2). Since that report, there havebeen relatively few reports ofC. indologenesbacteremiain humans (3 10), and most of them have occurred inTaiwan (2,5,11,12). Clinical manifestations includenosocomial pneumonia, biliary tract infection, peritoni-tis, surgical wound infection (2), intravascular catheter-related bacteremia (3,9), cellulitis (4), and primary bac-teremia (2,5 8,10).C. indologenesinfections that areassociated with various indwelling devices have beenreported (12). AlthoughC. indologenesis a relativelyuncommon human pathogen, we have observed an in-creasing incidence ofC. indologenesbacteremia in ourintensive care unit (ICU) over the last few years (0 in2003, cases per 1,000 patient-days in 2004, 2005, in 2006, in 2007, and in2008). The incidence of bacteremia caused by otherpathogens was cases per 1,000 patient-days in 2003, in 2004, in 2005, in 2006, in 2007, and 2008.

4 However, thorough examination of existingliterature revealed no publications that specifically fo-cused on ICU-acquiredC. , we elucidate the Clinical characteristics, an-timicrobial Susceptibilities , and outcomes of patientswith ICU-acquiredC. study was conducted at Tainan MunicipalHospital, which is a referral teaching hospital located insouthern Taiwan. The mixed medical-surgical ICU is a38-bed unit with individual rooms and about 2,000 an-nual admissions. Admissions are two-thirds medical andone-third surgical. All patients hospitalized in the ICUduring the study period, which was between January2004 and December 2008, with proven positive bloodcultures forC. indologeneswere included. Collectedpatient data included age, gender, any immunocom-promised diseases, condition while in the ICU, use of in-dwelling devices (including central venous catheter, in-vasive mechanical ventilation, and urinary catheter),length of stay in the ICU prior to the onset of infection,other sites ofC.

5 Indologenesisolation, primary site ofinfection, polymicrobial bacteremia, antibiotic therapy,and hospital indologenesbacteremia was definedwhen a patient hadC. indologenesbacteremia afterstaying in the ICU for more than 48 h. Catheter-relatedbacteremia caused byC. indologeneswas defined whena patient had a vascular catheter in place for longer than48 h and who experienced an unexplained fever. In ad-521521dition, at least one peripheral blood culture and one cul-ture of the catheter tip had to be positive forC. in-dologenes. Clinical diagnoses of sepsis and septic shockwere based upon the criteria specified by the AmericanCollege of Chest Physicians and the Society of CriticalCare Medicine (13). Polymicrobial bacteremia was de-fined as the identification of more than one microorgan-ism from a single set of blood cultures (14). Appropriateantibiotic therapy was defined as one of the prescribedantibiotics with minimal inhibitory concentrations(MICs) below the corresponding breakpoint for non-Enterobacteriaceaein isolates ofC.

6 Indologeneswere identified bythe commercial API 20NE identification system(bioM erieux, Marcy-l'Etoile, France) during the studyperiod. Results were interpreted using the API LAB IDcomputer software. The positive blood culture isolateswere stored at 709C and subcultured before furtherconfirmation. All isolates were confirmed asC. in-dologenesby the BD Phoenix Automated MicrobiologySystem (software version ; Becton Dickinson,Sparks, Md., USA).The individual MICs of the Antimicrobial agents forC. indologeneswere determined by the Phoenix systemfor Antimicrobial susceptibility testing. The tested an-timicrobial agents from the Phoenix panels were gen-tamicin, amikacin, ciprofloxacin, levofloxacin, piper-acillin, piperacillin/tazobactam, ceftriaxone, ceftazi-dime, cefepime, imipenem, meropenem, and trime-thoprim/sulfamethoxazole (TMP/SMZ). Vancomycinand minocycline were tested with Etest strips (AB Bio-disk, Solna, Sweden). The percentage of isolates thatwere susceptible to these Antimicrobial agents was deter-mined by applying the Clinical and Laboratory Stand-ards Institute susceptibility breakpoints used for non-Enterobacteriaceae(15).

7 Our study included 10 patients (1 patient in 2004, 2 in2005, 2 in 2006, 2 in 2007, and 3 in 2008).C. in-dologenesaccounted for the total number ofICU-acquired bloodstream infections over the 5-yearstudy period. Table 1 shows the Clinical Characteristics, antibiotic therapy, and outcome of these patients. Fivepatients were men (50z), and 5 were women (50z).The mean age was years (range, 48 84 years). Sixpatients (60z) had underlying immunocompromiseddiseases, 4 patients (40z) had malignancies, 3 (30z)had diabetes mellitus, and 2 (20z) had cirrhosis. Theprimary site ofC. indologenesinfection was unknownin 80zof the cases. The known primary sites of infec-tion were empyema (10z) and catheter-related bactere-mia (10z). Eight patients (80z) had polymicrobial bac-teremia (2 patients withAcinetobacter baumannii,2patients withKlebsiella pneumoniae,1patientwithChryseobacterium meningosepticum,1patientwithEnterococcus faecalis,1patientwithEnterobacter cloa-cae,1patientwithBurkholderia cepacia,1patientwithStaphylococcus aureus,and1patientwithCandida al-bicans).

8 Eight patients (80z)hadstayedintheICUformore than 21 days. Eight patients (80z) had receivedmore than 14 days of broad-spectrum antibiotics priorto the onset ofC. MIC ranges and the MICs at which 50zand90zfor the 10 isolates ofC. indologenesare shown inTable 2. All isolates were 100zsusceptible to minocy-cline and TMP/SMZ. Vancomycin, imipenem, piper-acillin/tazobactam, ciprofloxacin, and levofloxacin ex-hibited 0z,10z,20z,30z,and30z, respectively,susceptibility against this pathogen. All isolates were100zresistant to ceftazidime, ceftriaxone, cefepime,meropenem, piperacillin, gentamicin, and amikacin. Nopatients had received appropriate antibiotic therapyprior toC. patients presented with sepsis. Four patients(40z) developed septic shock and died within 6 was determined to be the cause of death inall of these cases. The 14-day mortality rate in our studywas were no overlapping dates of hospitalizationamong our patients, and they were treated by differenthealth care workers.

9 Therefore, all patients had sporad-ic infections. Although they had all received invasivemechanical ventilation, only one patient hadC. in-dologenesisolated from the pleural effusion. None oftheC. indologeneswas isolated from the trachealaspirate or bronchoalveolar SENTRY Antimicrobial Surveillance Programshowed that the quinolones, including levofloxacin(100zsusceptible) and ciprofloxacin ( ), had the highest potency againstC. ( ), piperacil-lin ( ), ceftazidime ( ), and cefepime ( ) were the mostactive agents among theb-lactams (16). Here,C. in-dologeneswas uniformly resistant to third- and fourth-generation cephalosporins, piperacillin, meropenem,and aminoglycosides. Piperacillin/tazobactam, imipen-em, ciprofloxacin, and levofloxacin exhibited little ef-fect against this pathogen. Our findings confirm thatthe isolates collected from patients hospitalized in theICU had much lower Susceptibilities to various an-timicrobial agents than that suggested by previous inves-tigations.

10 Most of our patients had received more than14 days of broad-spectrum antibiotics, which mighthave led to selective pressure for resistance in this patho-gen. As indicated here, it is very difficult to choose anappropriate antibiotic therapy for this minocycline and TMP/SMZ had the highestpotency against this pathogen in vitro, Clinical treatmentofC. indologenesbacteremia with minocycline andTMP/SMZ awaits further research. In addition, a highproportion of our patients had polymicrobial bactere-mia, including Gram-negative bacilli, Gram-positivecocci, orCandida. The above results severely compli-cate the choice of an effective antibiotic therapy forICU-acquiredC. of our patients who were free from underlyingimmunocompromised diseases (patients 1, 3, 5, and 7)recovered well, despite the initial administration of an-tibiotics that lacked activity againstC. outcome suggests a low virulence that is relativelyineffective in immunocompetent patients. In contrast,four immunocompromised patients (patients 2, 4, 6,and 10) who had not received appropriate antibiotictherapy died of septic shock within 6 days.