CASE REPORT
Kluyvera cryocrescens Finger
Infection: Case Report and Review of
Eighteen Kluyvera Infections in
Human Beings
Burton C. West, Hemalatha Vijayan, and
Raja Shekar
We report a case of soft tissue infection with Kluyvera cryocrescens and a critical review of Kluyvera infections. A 31year-old diabetic man used a new chemical for stripping the
floor with his bare hands. Two days later he developed a blister
on a finger which progressed to tenosynovitis in spite of intravenous nafcillin therapy. After 11 days culture and sensitivity
results dictated treatment with intravenous ticarcillin/clavulanic acid. The wound was debrided twice, and later a skin
flap was done. Wound cultures became sterile after 7 days of
treatment with ticarcillin/clavulanic acid, and he recovered.
This case represents the fourth clinical infection with K. cryocrescens and the eighteenth of Kluyvera to be reported. Four
others were K. ascorbata, and the remaining ten Kluyvera
infections in humans were not identified beyond genus. Our
case and review of the 17 previous cases emphasize that while
Kluyvera rarely cause disease, these opportunistic Gramnegative bacilli may be virulent in a variety of sites under as
yet poorly defined host conditions. Sites of infection varied,
but the brain and meninges were not among them. Two patients had diabetes mellitus, none had AIDS, and four died.
Once shown clinically to be the cause of an infection,
Kluyvera deserve aggressive treatment which acknowledges
their ampicillin resistance.
© 1998 Elsevier Science Inc.
INTRODUCTION
CASE REPORT
The genus Kluyvera has been isolated from a variety
of human clinical specimens, but it is rarely implicated as the significant pathogen. Only 17 cases of
Kluyvera infection have been documented in the literature (Table 1). We report a patient with a softtissue finger infection caused by Kluyvera cryocrescens, the fourth K. cryocrescens human infection to be
reported.
A 31-year-old black man, who had insulindependent diabetes mellitus for 6 years, used a new
chemical for stripping the floor with his bare hands.
Two days later he developed a painful blister on the
right middle finger. He was treated with dicloxacillin. Three days later, because the finger was swollen
and its range of motion decreased, the patient was
admitted to Meridia Huron Hospital.
His temperature was 37.2°C, pulse 76 beats/min,
respiratory rate 20/min and blood pressure 130/80
mm Hg. The finger was erythematous, tender, and
swollen from the proximal interphalangeal joint to its
tip. There was no sensory loss, lymphangitis, or
lymphadenopathy. The rest of the physical examination was normal. The white count was 5,240/mL,
erythrocyte sedimentation rate 7 mm/h and blood
glucose 367 mg/dL. An X-ray showed soft-tissue
From the Department of Medicine, Meridia Huron Hospital,
Cleveland, Ohio, USA.
Address reprint requests to Dr. Burton C. West, Meridia Huron
Hospital Department of Medicine, 13951 Terrace Road, Cleveland OH 44112.
Received 27 February 1998; accepted 10 July 1998.
DIAGN MICROBIOL INFECT DIS 1998;32:237–241
© 1998 Elsevier Science Inc. All rights reserved.
655 Avenue of the Americas, New York, NY 10010
0732-8893/98/$19.00
PII S0732-8893(98)00087-X
Author and
Reference
Braunstein et al.
(1980)
Fainstein et al.
(1982)
Aevaliotis et al.
(1985)
Wong (1987)
Thaller et al.
(1988)
Year
Location
Age/Sex
79 San Bernardino, 63/F
California,
USA
81 Houston, Texas, 71/M
USA
52/M
75/M
72/M
67/M
85 Brussels,
3 wks/F
Belgium
17 months/M
87 Los Angeles,
California,
USA
88 Rome, Italy
76/F
Tristram and
Forbes (1988)
88 Syracuse, New
York, USA
11 months/F
Luttrell et al.
(1988)
37/F
88 Johnson City,
Tennessee,
USA
90 Jerusalem, Israel 5/F
Underlying Disease
Clinical
Presentation
238
TABLE 1 Clinically Significant Cases of Kluyvera Infection in Humans
Source of the
Organism
Kluyvera
Species
Treatment
Outcome
None
Acute pancreatitis Gallbladder fluid
ND
Not stated
Not stated
None
Diarrhea (severe)
Stool
ND
Symptomatic
Cured
AML
Bladder carcinoma
Bladder carcinoma
CML
None
Diarrhea (severe)
Diarrhea (severe)
Diarrhea (mild)
Diarrhea (mild)
Diarrhea
Stool
Stool
Stool
Stool
Stool
ND
ND
ND
ND
ascorbata
TMP/SMZ
Symptomatic
Symptomatic
TMP/SMZ
Symptomatic
Cured
Cured
Cured
Cured
Cured
Congenital heart
disease
Blood
Sepsis/Broviac
catheter
infection
Gallbladder pus
Acute
emphysematous
cholecystitis
Sepsis & UTI
Blood and urine
Chronic cholecystitis
Bilateral
vesicoureteral
reflux
None
cryocrescens Ampicillin and gentamicin
Died
cryocrescens Gentamicin
Cured
ND
Tobramycin and cefoxitin
Cured
Soft tissue
infection
Wound culture
ND
Cefaperazone, TMP/SMZ, Cured
ticarcillin/clavulanic acid
None
Pyelonephritis
Urine
ND
Cefazolin
90 Chicago, Illinois, 13/M
USA
Friedreich’s ataxia
Peritonitis and
urinary tract
infection
Sierra-Madero
et al. (1990)
90 Cleveland, Ohio, 74/M
USA
Diabetes mellitus,
CABG, and CRF
Mediastinitis
Sanchis Bayarri
et al. (1992)
92 Valencia, Spain 18/M
Prolapsed colostomy, Gross purulent
discharge
urethral-rectal
fistula (posttraumatic)
None stated
Acute
pyelonephritis
Chronic renal failure, Peritonitis
CAPD
Endotracheal tube ascorbata
aspirate,
peritoneal fluid,
urine and lung
tissue
Blood, mediastinal ND
wound, and
sputum
Purulent exudate ascorbata
Dollberg et al.
(1990)
Yogev and
Kozlowski
(1990)
36/M
96 Antalya, Turkey 57/F
Present report
96 Cleveland, Ohio, 31/M
USA
Moxalactam and gentamicin Died
Ceftizoxime and gentamicin Died
Not stated
Not stated
Not stated
Not stated
Urine
ascorbata
CAPD fluid
cryocrescens Vancomycin and amikacin, Died with
multiple
then ciprofloxacin, then
organ
imipenem and amikacin
failure
cryocrescens Dicloxacillin, then nafcillin, Cured
then ticarcillin/clavulanic
acid
Wound culture
Soft tissue and
Diabetes mellitus
(insulin dependent), flexor tendon
sheaths (finger)
local chemical and
physical trauma
AML, Acute myelogenous leukemia; CABG, coronary artery bypass graft; CAPD, chronic ambulatory peritoneal dialysis; CML, chronic myelogenous leukemia; CRF, chronic renal failure;
ND, not done; TMP/SMZ, trimethoprim/sulfamethoxazole.
B.C. West et al.
Sezer et al.
(1996)
Cured
Kluyvera Finger Infection and Review
swelling of the right middle finger without evidence
for osteomyelitis. Cellulitis was diagnosed, but a
Gram stain and culture of an aspirate of the finger
were negative.
On admission, nafcillin, 1 g, q.4.h., was administered intravenously. Because the infection had
spread to the flexor tendon sheaths, on hospital day
6 the finger was debrided and the drainage from the
wound was aerobically and anaerobically cultured.
The Gram stain showed moderate Gram-positive
cocci in pairs, chains and clusters, a few Grampositive rods, and many neutrophils. The culture
revealed moderate growth of Kluyvera and diphtheroids after 5 days (hospital day 11). On the same day
the patient underwent the second debridement because he had not improved. Gram stain of pus obtained from the surgical debridement showed many
Gram-negative bacilli, few Gram-positive cocci in
pairs, and moderate neutrophils. On the same day
(hospital day 11) intravenously administered ticarcillin/clavulanic acid 3.1 g, q.6.h. and vancomycin, 1 g,
q.12h., were substituted for nafcillin. Aerobic and
anaerobic cultures once again grew Kluyvera now
accompanied by Bacteroides spp.
After 7 days of treatment with ticarcillin/clavulanic acid, a wound culture showed no growth. Skin
grafting with a groin flap was done. The patient
remained afebrile throughout the hospital course
and was discharged on ciprofloxacin after 21 days. A
month later the groin flap was detached, and the
donor site was closed.
Kluyvera was identified to genus using the API 20
E system (Analytab, Inc., Plainview, NY, USA) and
Vitek AMS (Vitek Systems, Hazelwood, MO, USA) at
the hospital and to species at the Cleveland Biological Institute; K. cryocrescens was confirmed by the
Ohio Department of Health. It demonstrated intermediate susceptibility to ampicillin and cefazolin. It
was sensitive to cefotaxime, cefoxitin, ciprofloxacin,
gentamicin, piperacillin, trimethoprim/sulfamethoxazole, and ticarcillin.
RESULTS, DISCUSSION, AND REVIEW
In 1956, Asai and coworkers identified a group of
flagellated Gram-negative bacilli that produced large
amounts of alpha-ketoglutaric acid during glucose
fermentation, excluding them from classification as
Pseudomonas. Asai named these organisms Kluyvera
because of the similarity to an organism predicted by
Kluyver and van Niel in 1936, but later proposed
Kluyvera be transferred to the genus Escherichia, since
they have peritrichous flagella. The genus Kluyvera
was abolished and did not appear on the “Approved
Lists of Bacterial Names” published in 1976 (Farmer
et al. 1981).
239
In 1979 Braunstein et al. (1980) isolated the same
type of organism from the sputum of a 6-year-old
boy with pulmonary tuberculosis. It was considered
a contaminant. They also isolated it from the gallbladder drainage of a 63-year-old woman with acute
pancreatitis. The Center for Disease Control (Atlanta,
GA, USA) identified them as “Enteric Group 8” or
“CDC Group 8” or “Citrobacter-Enterobacter intermediate” (Braunstein et al. 1980).
Based on extensive investigations, in 1981 Farmer
et al. placed these microorganisms formerly known
as CDC Group 8 in a separate genus Kluyvera in the
family Enterobacteriaceae. They can be identified by
biochemical criteria, including positive reactions for
catalase, ornithine decarboxylase, indole, citrate, malonate, and esculin. They ferment arabinose, mannitol, maltose, melibiose, raffinose, rhamnose, glucose,
and sucrose. They have negative reactions for inositol, adonitol, urea, H2S, arginine, Voges-Proskauer,
and tryptophan deaminase (Farmer et al. 1981, 1985).
At present the genus Kluyvera is composed of two
species: K. ascorbata and K. cryocrescens and some
strains designated Kluyvera species 3. In contrast to
the observation that K. ascorbata was more frequently
isolated from clinical specimens while K. cryocrescens
was isolated primarily from the environment and
rarely from clinical specimens (Farmer et al. 1985),
we tabulated four well-documented cases of K. ascorbata and four of K. cryocrescens (including our case) as
causes of human disease (Table 1). The number of
isolates of each species remains small, making generalization risky. The species are differentiated by
the ability of K. ascorbata to utilize ascorbate and K.
cryocrescens to grow and ferment d-glucose at 5°C.
Our isolate grew luxuriantly at 5°C, which was the
key to consideration of this species. They are also
differentiated by differences in the zone of inhibition
around disks of carbenicillin (100-mg disk) and cephalothin (30-mg disk). The carbenicillin disk usually
produces a zone of inhibition .17 mm with K. cryocrescens, but carbenicillin produces a much smaller
zone of inhibition with K. ascorbata (Farmer et al.
1985). In our case, carbenicillin produced a zone of
inhibition of 26 mm.
In addition, a positive reaction for lysine decarboxylase is observed in 97% of isolates of K. ascorbata,
while only 23% of K. cryocrescens have a positive
reaction (Farmer et al. 1995, Sanchis Bayarri et al.
1992). Dulcitol fermentation occurs with 25% of K.
ascorbata, while zero isolates of K. cryocrescens demonstrate this fermentation (Farmer et al. 1995). The
strains can be differentiated by their response to
irgasan (Altwegg et al. 1986). Kluyvera ascorbata
shows no zone of inhibition while K. cryocrescens
shows a zone of .15 mm around a disk containing 5
mg of irgasan. In our case the organism was identified at both reference laboratories as K. cryocrescens
240
by utilizing all biochemical tests except irgasan disk
testing.
In our patient, the chemical and physical injury to
his hand contributed to his local susceptibility and
implies an environmental source. The cellulitis at
first was treated with dicloxacillin that might have
controlled streptococci or staphylococci in a mixed
infection including K. cryocrescens or a primary infection that later became infected with K. cryocrescens.
The various Gram-positive cocci on the smear from
hospital day 6 suggest that despite dicloxacillin and
nafcillin treatment, both streptococci and staphylococci had been present. At the same time, Grampositive rods on the Gram stain suggest that the
diphtheroids isolated in mixed culture with K. cryocrescens were also part of a mixed infection. Later as
the infection spread into the tendon sheath and after
five more days of nafcillin therapy, K. cryocrescens
emerged as the dominant pathogen by Gram stain
and by culture, although Bacteroides spp. accompanied its growth. No diphtheroids grew again and
never did streptococci or staphylococci grow. One
might designate this an opportunistic infection
emerging from a mixed infection. We were impressed by the severity of the cellulitis and tenosynovitis that progressed during treatment with semisynthetic penicillins. We did not think this was a
trivial secondary infection because at presentation
there was neither a break in the skin nor purulent
drainage from the cellulitis. Inoculation of K. cryocrescens must have occurred primarily, i.e., at the time
of injury. Moreover, the emergence of K. cryocrescens
as the dominant pathogen was a response to antimicrobial pressure in killing other pathogens or potential pathogens in the infectious inoculum, leaving K.
cryocrescens less opposed by microbial competition.
Seventeen clinical cases of significant Kluyvera infection reported in the literature and the present case
are summarized in Table 1. The sex ratio is 11:7
(M:F). The age ranges from 3 weeks to 74 years.
There is no specific site of infection favored by this
genus of bacteria and no favored body fluid or tissue
infected by these organisms. However, no brain or
meningeal infections are reported. Deaths occurred
in 4 of the 15 patients for whom we know an outcome. Of the four deaths, three had blood cultures; in
two, the blood cultures were positive. In one death
associated with peritonitis, multiorgan failure was
observed, but this patient did not have blood cultures reported (Sezer et al. 1996). Other blood isolates
are known, but clinical details are lacking (Brenner
1984). Treatment with ampicillin to which Kluyvera is
usually resistant may have contributed to the death
B.C. West et al.
of one case (Table 1). Our isolate was typical in
lacking susceptibility to ampicillin.
Host factors and a poor underlying condition
might be more important in determining susceptibility or a fatal outcome than the infecting organism,
but no host defect has been defined. No acquired
immunodeficiency syndrome associated cases have
been reported to our knowledge. Our nonfatal case
and a fatal case (Sierra-Madero et al. 1990) had diabetes mellitus, a common predisposition to serious
infections, but two cases hardly define a predisposition to Kluyvera.
Kluyvera was isolated from stools of 12 persons, of
whom five were asymptomatic cancer patients (Aevaliotis et al. 1985; Fainstein et al. 1982). Seven patients had diarrhea, six of whom are listed in Table 1.
One was omitted because of the concurrent isolation
of Salmonella enteritidis, the probable cause of diarrhea. Because of those reports, it appeared that
Kluyvera might become an important separate cause
of bacterial diarrhea, but no cases have been reported
since 1985. An association with a diseased biliary
tract was made in two other cases, but stool was not
cultured (Braunstein et al. 1980; Thaller et al. 1988).
Neither was stool cultured in cases of urethral-rectal
fistula or pyelonephritis where there might have
been an intestinal source (Sanchis Bayarri et al. 1992).
Although more than one organism was isolated
from the tenosynovitis in our case, K. cryocrescens
was isolated consistently and became dominant by
Gram stain and culture over time. We think it was
the significant pathogen because the patient did not
improve until he received ticarcillin/clavulanic acid
to which the K. cryocrescens was susceptible. Our case
and review emphasize that Kluyvera may cause serious human infections. Whether Kluyvera are “simple” opportunistic invaders or are simply rare but
virulent remains unresolved. Microbiologists should
be aware of the significance of Kluyvera and identify
them fully when they are suspected. Physicians
should be aware of the potential virulence of
Kluyvera species and treat them, aware of their resistance to ampicillin. They must not be dismissed as
saprophytes, whenever found in a clinically significant infection.
We thank Linda Church, M.S., of Meridia Huron Hospital
and Mario Markowic, M.S., of Cleveland Biological Institute,
Cleveland, Ohio, for isolating and identifying the organism.
We thank Billy Brown, M.D., Damian Laber, M.D., Ann Tekancic, and Cindy Storm for assistance. Financial support is
from The Meridia Huron Hospital Department of Medicine.
Kluyvera Finger Infection and Review
241
REFERENCES
Aevaliotis A, Belle AM, Chanione JP, Serruys E (1985)
Kluyvera ascorbata isolated from a baby with diarrhea.
Clin Microbiol Newsl 7:51.
Altwegg M, Zollenger-Iten J, von Graevenitz A (1986) Differentiation of Kluyvera cryocrescens from Kluyvera ascorbata by irgasan susceptibility testing. Ann Inst Pasteur
Microbiol (Paris) 137A:159–168.
Braunstein H, Tomasulo M, Scott S, Chadwick MP (1980) A
biotype of Enterobacteriaceae between Citrobacter and Enterbacter. Am J Clin Pathol 73:114–116.
Brenner DJ (1984) Enterobacteriaceae. In: Bergey’s Manual of
Systematic Bacteriology. Vol. 1. Eds, Kreig NR, Holt JG.
Baltimore: Williams and Wilkins, pp. 511–513.
Dollberg S, Gandacuu A, Klar A (1990) Acute pyelonephritis due to Kluyvera species in a child. Eur J Clin Microbial
Infect Dis 9:281–283.
Fainstein V, Hopfer R, Mills K, Bodey G (1982) Colonization by or diarrhea due to Kluyvera species. J Infect Dis
145:127.
Farmer JJ III, Fanning GR, Huntley-Carter GP, Homes B,
Hickman FW, Richard C, Brenner DJ (1981) Kluyvera, a
new (redefined) genus in the family Enterobacteriaceae:
Identification of Kluyvera ascorbata sp. nov. and Kluyvera
cryocrescens sp. nov. in clinical specimens. J Clin Microbiol 13:919–933.
Farmer JJ III, Davis BR, Hickman-Brenner FW, McWhorter
A, Huntley-Carter GP, Asbury MA, Riddle C, WathenGrady HG, Elias C, Fanning GR, Steigerwalt AG,
O’Hara CM, Morris GK, Smith PB, Brenner DJ (1985)
Biochemical identification of new species and bio-
groups of Enterobacteriaceae isolated from clinical specimens. J Clin Microbiol 21:46–76.
Farmer JJ III (1995) Enterobacteriaceae: Introduction and
identification. In: Manual of Clinical Microbiology. 6th ed.
Eds, Murray PR, Baron EJ, Pfaller MA, Tenover FC,
Yolken RH. Washington, DC: ASM Press, pp. 438–450.
Luttrell RE, Rannick GA, Soto-Hernandez JL, Verghese A
(1988) Kluyvera species soft tissue infection: Case report
and review. J Clin Microbiol 26:2650–2651.
Sanchis Bayarri V, Sanchez Sánchez R, Marcaida Benito G,
Llucian Rambla R (1992) Infecciones por Kluyvera ascorbata. A propósito de dos casos. Rev Clin Esp 190:187–188.
Sezer MT, Gultekin M, Gunseren F, Erkilic M, Ersoy F
(1996) A case of Kluyvera cryocrescens peritonitis in a
CAPD patient. Dialysis Internat 16:326–327.
Sierra-Madero J, Pratt K, Hall GS, Stewart RW, Scerbo JJ,
Longworth DL (1990) Kluyvera mediastinitis following
open-heart surgery: A case report. J Clin Microbiol 28:
2848–2849.
Thaller R, Berlutti F, Thaller MC (1988) A Kluyvera cryocrescens strain from a gallbladder infection. Eur J Epidemiol
4:124–126.
Tristram DA, Forbes BA (1988) Kluyvera: A case report of
urinary tract infection and sepsis. Pediatr Infect Dis J
7:297–298.
Wong VK (1987) Broviac catheter infection with Kluyvera
cryocrescens: A case report. J Clin Microbiol 25:1115–1116.
Yogev R, Kozlowski S (1990) Peritonitis due to Kluyvera
ascorbata: Case report and review. Rev Infect Dis 12:399–
402.