A wide range of pathogens can cause hospital-associated (HA) infections in small animal hospitals. Among these is Enterobacter cloacae, which is one of the most clinically relevant Enterobacter spp and a common cause of HA infection in humans. Recently, multidrug resistance has become a concern, particularly with emergence of extended-spectrum beta-lactamase and extended spectrum cephalosporinase producing strains. An infection control investigation was initiated at the Ontario Veterinary College Teaching Hospital (OVCTH) in the fall of 2007 in response to anecdotal concerns about Enterobacter cloacae infections in hospitalized animals.
Enterobacter cloacae was isolated from 45/36719 animals from January 1, 2005 to October 31, 2007, for an overall incidence of 1.2/1000 admissions. The monthly incidence rate ranged from 0 to 4.3/1000 admissions. Twenty-one (47%) cases were classified as community-associated, while 17 (38%) were hospital associated. Seven (15%) were community-onset but hospital associated, with three of these associated with other veterinary hospitals. There was no increase in the incidence of overall or hospital-associated infections during the study period.
The urinary tract was the most common site of infection (n=11, 24%). Wound infections (excluding surgical site infections) accounted for 8 (18%) of infections with superficial and deep surgical site infections accounting for 7 (16%) and organ/space surgical site infections accounted for another 2 cases. Urinary tract infections were most common among animals with CA infection, accounting for 8/21 (38%) cases with wound infections accounting for 4 (19%) cases.
Of the 24 cases associated with the OVCTH, 17 (71%) had surgery, 15 (63%) were hospitalized in the intensive care unit, 10 (42%) had indwelling urinary catheters placed and 20 (83%) had received antimicrobials prior to onset of infection. Risk factors for E. cloacae infection could not be determined because a non-infected control group was not evaluated. Surgical site infections accounted for 9 (38%) HA cases. Overall, only 2/11 (18%) urinary tract infections were associated with prior placement of a urinary catheter. Nine (20%) animals died or were euthanized and E. cloacae was implicated as a causative or contributing factor in 5 (56%) of those cases.
Two main antimicrobial phenotype patterns were identified. One (n=25) was characterized by susceptibility to fluoroquinolones, tetracycline and trimethoprim with variable susceptibility to cefoxitin while the other (n=14) was characterized by resistance to these antimicrobials. Prior administration of antimicrobials was associated with presence of the more resistant phenotype (P=0.044) but there was no association between this phenotype and origin of infection (P=0.74) and no increase in the prevalence of this phenotype from 2005 to 2007 (P=0.97). Infections with this phenotype were not associated with non-survival (P=0.74).
There was no evidence of a HA outbreak or increase in prevalence, yet identification of multidrug resistant E. cloacae in both CA and HA infections is concerning and requires ongoing surveillance.