SEARCH

SEARCH BY CITATION

Keywords:

  • Identification;
  • differentiation;
  • susceptibility;
  • Citrobacter

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. References

Objective: Recently a publication of Brenner et al. introduced 11 genetically distinct species within the genus Citrobacter. These newly recognized Citrobacter species can be classified by means of their biochemical characteristics. The aim of this study was to examine the distribution and susceptibility of Citrobacter isolates in our patient population.

Method: A total of 126 samples—containing a Citrobacter species—was collected from 116 hospitalized patients during a 6-month period. Organisms were identified according to standard procedures. Antimicrobial susceptibility testing was performed by agar dilution on Mueller-Hinton agar, and interpretation was based on NCCLS criteria.

Results: C. freundii was the most common organism isolated (n=59), followed by C. braakii (n=25) and C. koseri (n=23). The urinary tract and the respiratory tract were found to be the predominant sites of colonization or infection, accounting for 45% and 32% of all isolates respectively. It appeared that young children (<12 months old) and the elderly were most at risk of acquiring Citrobacter. Two-thirds of all specimens contained other organisms in addition to Citrobacter. Most Citrobacter isolates were related with a predisposing factor. Species-related differences were found in the susceptibility pattern.

Conclusions: These findings suggest that citrobacteria are important opportunistic pathogens contributing to colonization or infection in our hospital population.

The genus Citrobacter consists of Gram-negative, facultative anaerobic, motile bacilli with growth on Simmons citrate medium (hence its name). These bacilli are commonly found in water, soil, food and the intestinal tracts of animals and humans. A new publication of Brenner et al. [1] proposed the introduction of 11 genetically distinct species within the genus Citrobacter. The 11 species are: C. koseri (formerly described as C. diversus), C. amalonaticus, C. farmeri, C. freundii, C. youngae, C. braakii, C. werkmanii, C. sedlakii and the three unnamed groups (Citrobacter species 9, 10 and 11). The term ‘C. freundii complex’ covers the last eight Citrobacter species mentioned. Citrobacteria have been associated with a variety of infectious diseases. Amongst others, hospital-acquired bacteremias, endocarditis, urinary tract infections, neonatal meningitis and brain abscess have been reported [2–9]. The etiologic significance of citrobacteria in diarrheal illnesses has been put forward in some studies, but is still unclear [10,11]. We conducted this study to evaluate some aspects of Citrobacter isolates of patients staying at our university hospital.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. References

Bacterial strains and patients

A total of 126 samples was collected during a 6-month period (from 1 November 1994 to 30 April 1995) from clinical specimens submitted to the microbiology laboratory of the University Hospital of Leuven. This study involved 116 patients from whom a Citrobacter species was isolated. In eight patients citrobacteria were obtained from two or more different isolation sites. The information obtained from the medical records of the patients included the patient's age, sex, additional underlying diseases and possible risk factors.

Biochemical tests

A preliminary subdivision of the Citrobacter isolates into C. koseri, members of the C. amalonaticus group (this group consists of C. amalonaticus and C. farmeri, previously described as C. amalonaticus biogroup 1) or members of the C. freundii complex was performed with the following tests: H2S and indole production, malonate utilization, and acid production from adonitol and D-arabitol [12]. Further biochemical identification of the Citrobacter strains to the species level was performed according to standard procedures based on the criteria of Brenner et al. [1]. Sugar fermentation was performed in Phenol Red Broth. Tests were read daily, and final results were recorded after 7 days of incubation. All media were incubated at 36±1°C.

Antimicrobial susceptibility testing

The antimicrobial susceptibility was determined by agar dilution on Mueller-Hinton agar. For each antibiotic one or two fixed concentrations, as indicated, were used. From each strain approximately 104 CFU/spot was inoculated with the multipoint inoculator. After overnight incubation at 36±1°C the Mueller-Hinton plates were examined. Susceptibility patterns were determined by means of inhibition of visible growth of the organism after incubation. All strains were tested for their sensitivity according to the NCCLS breakpoints (amoxiclavulanate, piperacillin/tazobactam and co-trimoxazole are combinations of two antibiotics. The concentrations of both antibiotics are given): ampicillin (8 mg/L), amoxiclavulanate (8 mg/L), piperacillin (16 mg/L), piperacillin/tazobactam (16/2 mg/L), cefazolin (8 mg/L), cefuroxime (8 mg/L), cefotaxime (8 mg/L), ceftazidime (8 mg/L), cefepime (8 mg/L), imipenem (4 mg/L), co-trimoxazole (2/38 mg/L), gentamicin (4 mg/L), tobramycin (4 mg/L), amikacin (16 mg/L), nitrofurantoin (32 mg/L), and ofloxacin (2 mg/L). For temocillin 16 mg/L was taken as the breakpoint.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. References

Identification and frequency distribution of Citrobacter species

During the 6 months of study, 126 Citrobacter isolates were observed in specimens of 116 hospitalized patients. From six patients and two patients we isolated the same Citrobacter species from two and three different specimen sources, respectively, and hence they were counted only once in the analysis of the biochemical characteristics (Table 1) and in the frequency distribution. C. koseri accounted for 19.8% (n= 23) of all Citrobacter strains. Within the C. freundii complex, C. freundii was the most frequently identified species (50.9%, n= 59) followed by C. braakii (21.6%, n=25). From the 116 strains tested, only four strains of C. youngae, two strains of C. werkmanii and one strain of C. sedlakii were identified. No strains of C. amalonaticus, C. farmeri, Citrobacter species 9, Citrobacter species 10 and Citrobacter species 11 were found.

Table 1.  Biochemical characteristics of Citrobacter species found in this study
 % of positive strains
TestC. koseri (n=23)C. freundii (n=59)C. braakii (n=25)C. youngae (n=4)C. werkmanii (n=2)C. sedlakii (n=1)
  1. ADH=arginine dihydrolase; ODC=ornithine decarboxylase. Results were read after 2 days of incubation. The values in parentheses are the percentages of strains which gave delayed-positive reactions.

  2. NT=not tested.

Indole10020250100
H2S071961001000
Citrate1009892100100100
Malonate100300100100
AcetateNT3736501000
ADHNT92100100100100
ODCNT39200100
Acid produced from:      
 Adonitol10000000
 D-Arabitol10000000
 DulcitolNT14481000100
 MelibioseNT10096 (100)250100
 α-Methyl-D-glucosideNT2(7)13 (54)000
 RaffinoseNT10012 (16)000
 SalicinNT0 (7)12 (16)2500 (100)
 SucroseNT9802500

Specimen source

Fifty-seven of the 126 Citrobacter isolates (45.2%) were obtained from the urinary tract. Twenty-nine (50.9%) urine specimens involved catheterized patients with serious debilitating conditions. From the 28 non-catheterized patients, only six isolates were associated with the presence of symptoms referable to the urinary tract. The remaining 22 isolates were obtained from patients without symptoms or patients with a recent history of catheterization.

Forty-one of the 126 Citrobacter isolates (32.5%) were obtained from the respiratory tract. All isolates came from bronchus aspirates. Thirteen patients were intubated and one patient had a tracheostomy. Only seven patients (17.1%) showed signs consistent with pneumonia. Among the remaining 34 Citrobacter isolates, only four were not associated with an apparent predisposing factor such as intubation, tracheostomy, chronic pulmonary disease or some other debilitating disease. Sometimes a combination of these predisposing factors was observed.

Nineteen of the 126 Citrobacter isolates (15.1%) were obtained from wound exudates originating from soft tissue or skin sites. Ten isolates of Citrobacter were associated with underlying gastrointestinal diseases and/or invasive procedures. The other specimens originated from a variety of superficial wounds, including foot ulcers, an injured leg stump, perianal abscesses, pus expressed from an ear, a decubitus ulcer and paronychia.

Nine of the 126 Citrobacter isolates (7.1%) were obtained from blood cultures. In six patients the origin was probably the gastrointestinal tract, since positive blood cultures followed abdominal surgery. In all these cases Citrobacter was the only bacterium present in the blood culture.

No clear relation was detected between any one of the Citrobacter species and a particular specimen source.

Age, sex and medical department distribution of patients with Citrobacter isolates

The age distribution of patients with Citrobacter isolates is shown in Figure 1. Of all Citrobacter specimens, 9% were obtained from patients before the age of 12 months, while 77% of all isolates occurred at the age of > 50 years.

image

Figure 1. Age distribution of patients with Citrobacter isolates.

Download figure to PowerPoint

Citrobacter isolates were obtained from men more often than from women in reported specimen sources, except for the urinary tract specimens. No pattern of Citrobacter species by sex of patients was apparent.

The various Citrobacter species appeared to be randomly distributed over the medical departments in which the patients were hospitalized (geriatrics, pediatrics, surgery, intensive care unit and internal medicine).

Additional organisms present in specimens

Forty of the 57 urine specimens (70.2%) contained several other bacteria, mostly Enterobacteriaceae and non-fermenters. Catheterized patients were more likely to have polymicrobial bacteriuria (82.8% versus 57.1%, p=0.04 by chi-square). Twenty-six of the 41 respiratory tract specimens (63.4%) occurred as mixed cultures, mostly with yeasts, Enterobacteriaceae, staphylococci and non-fermenters. Fourteen of the 19 wound specimens (73.7%) contained additional bacteria, particularly Enterobacteriaceae, non-fermenters and Gram-positive cocci. Four of the nine blood cultures (44.4%) had other pathogens in addition to Citrobacter. The additional organisms encountered were Staphylococcus aureus, Candida albicans, Escherichia coli and Enterococcus species.

Antimicrobial susceptibility

Antimicrobial sensitivities of all Citrobacter strains were tested (Table 2). Species-related differences can be noted. All strains of C. koseri were resistant to ampicillin, but sensitive to cefazolin, and 91% of the strains were sensitive to amoxiclavulanate. Within the C. freundii complex, on the other hand, the majority of isolates were resistant to ampicillin (96.7%), amoxiclavulanate (98.9%) and cefazolin (96.7%). The sensitivities to antimicrobial agents of C. freundii and C. braakii were comparable. Approximately 19% of the C. freundii complex strains were multiply resistant to ampicillin, amoxiclavulanate, first-, second-, and third-generation cephalosporins and piperacillin.

Table Table.  Sensitivities to antimicrobial agents of Citrobacter species found in the study
 % of strains sensitive
Antimicrobial agentC. koseri (n=23)C. freundii (n=59)C. braakii (n=25)C. youngae (n=4)C. werkmanii (n=2)C. sedlakii (n=1)
Ampicillin0240500
Amoxiclavulanate9104000
Temocillin10098100100100100
Piperacillin8775841001000
Pipera/Tazobactam10078841001000
Cefazolin10028000
Cefuroxime836672100500
Cefotaxime1008088100100100
Ceftazidime1008088100100100
Cefepime100100100100100100
Imipenem1009896100100100
Co-trimoxazole10097100100100100
Gentamicin100100100100100100
Tobramycin10097100100100100
Amikacin10097100100100100
Nitrofurantoin87931001001000
Ofloxacin10093100100100100

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. References

A recent article from Brenner et al. introduced 11 genomospecies within the genus Citrobacter based on DNA-relatedness studies. This study deals with 126 Citrobacter isolates—collected during six successive months—from 116 patients staying at our university hospital. Identification of these Citrobacter strains was performed by means of conventional biochemical tests proposed by Brenner et al. [1]. The reported biochemical characteristics of the isolated Citrobacter species in this study (Table 1) correspond well with the results of Brenner et al. [1]. The most striking differences are the larger percentage of indole-negative C. freundii and C. braakii and the larger percentage of H2S-positive C. braakii and C. youngae. Within the C. freundii complex, C. freundii was the most common organism isolated, followed by C. braakii. This is in contrast to the results of Janda et al. [12]; in their studies C. youngae came second instead of C. braakii. No strains of C. amalonaticus or C. farmeri were found in this study, suggesting that these species are less pathogenic for humans. The urinary tract and the respiratory tract were predominant sites of isolation, accounting for 45.2% and 32.5% of all isolates respectively. Citrobacter infections were seldom encountered between the age of 1 and 40. More than 77% of all Citrobacter isolates occurred after the age of 40. One more group that appears to be at a higher risk comprises young children (<12 months old). The majority of isolates came from patients with previous surgery, who were admitted to an intensive care unit, or who had serious underlying diseases or malignancies. A striking experience in this study was the large amount of cultures containing other organisms in addition to Citrobacter. Two-thirds of all isolates occurred as part of a mixed infection. Our findings suggest that citrobacteria, like several other species of Enterobacteriaceae, are important opportunistic microorganisms contributing to colonization or infection. The opportunistic character of citrobacteria is suggested by the age of affected patients, the large proportion of mixed cultures, and the predisposing factors which were often present (e.g. intubation, mechanical ventilation, indwelling catheters, invasive procedures, serious compromising diseases).

As mentioned in previous reports, we also found marked differences in the susceptibility patterns of C. koseri and the members of the C. freundii complex [2,9]. C. koseri strains were generally sensitive to amoxiclavulanate and cefazolin, but resistant to ampicillin. Members of the C. freundii complex, on the other hand, were generally resistant to ampicillin, amoxiclavulanate and cefazolin. There is sometimes a discordance between the activity of ampicillin and amoxiclavulanate for the members of the C. freundii complex. This is probably due to the intrinsic higher activity of ampicillin resulting from the production of an inducible chromosomally encoded type I cephalosporinase by these citrobacteria, which is not inhibited by clavulanic acid. A worrisome fact is the isolation of several strains of C. freundii and C. braakii multiply resistant to the newer β-lactam antibiotics.

References

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. References
  • 1
    Brenner DJ, Grimont PAD, Steigerwalt AG, Fanning GR, Ageron E, Riddle CF. Classification of Citrobacteria by DNA hybridization: designation of Citrobacter farmeri sp. nov., Citrobacter youngae sp. nov., Citrobacter braakii sp. nov., Citrobacter werkmanii sp. nov., Citrobacter sedlakii sp. nov., and three unnamed Citrobacter genomospecies. Int J Syst Bacteriol 1993; 43: 64558.
  • 2
    Drelichman V, Band JD. Bacteremias due to Citrobacter diversus and Citrobacter freundii: incidence, risk factors, and clinical outcome. Arch Intern Med 1985; 145: 180810.
  • 3
    MacCulloch D, Menzies R, Cornere BM. Endocarditis due to Citrobacter diversus developing resistance to cephalothin. NZ Med J 1977; 85: 1823.
  • 4
    Carruthers M. Endocarditis due to enteric bacilli other than salmonellae: case reports and literature review. Am J Med Sci 1977; 273:20311.
  • 5
    Fields BN, Uwaydah MM, Kunz LJ, Swartz MN. The so-called ‘paracolon’ bacteria: a bacteriologic and clinical reappraisal. Am J Med 1967; 42: 89106.
  • 6
    Williams WW, Mariano J, Spurrier M, et al. Nosocomial meningitis due to Citrobacter diversus in neonates: new aspects of the epidemiology. J Infect Dis 1984; 150: 22935.
  • 7
    Graham DR, Anderson RL, Ariel FE, et al. Epidemic nosocomial meningitis due to Citrobacter diversus in neonates. J Infect Dis 1981; 144: 2039.
  • 8
    Curless RG. Neonatal intracranial abscess: two cases caused by Citrobacter and a literature review. Ann Neurol 1980; 8: 26972.
  • 9
    Lipsky BA, Hook ER III, Smith A, et al. Citrobacter infections in humans: experience at the Seattle Veterans Administration Medical Center and a review of the literature. Rev Infect Dis 1980; 2: 74660.
  • 10
    Guarino A, Capano G, Malamisura B, et al. Production of Escherichia coli St, heat-stable enterotoxin by Citrobacter freundii isolated from humans. J Clin Microbiol 1987; 25: 11014.
  • 11
    Guerrant RL, Dickens MD, Wenzel RP, Kapikian AZ. Toxigenic bacterial diarrhoea: nursery outbreak involving multiple bacterial strains. J Pediatr 1976; 89: 88591.
  • 12
    Janda JM, Abbott SL, Cheung WKW, Hanson DF. Biochemical identification of Citrobacteria in the clinical laboratory. J Clin Microbiol 1994; 32: 18504.