Results of culture form colonoscopically obtained specimens for bacteria and fungi in HIV-infected patients with diarrhea

Klebsiella oxytoca as an agent of antibiotic-associated hemorrhagic colitis

BACKGROUND & AIMS

Klebsiella oxytoca has been isolated from stools and colonic biopsy specimens of patients with Clostridium difficile-negative antibiotic-associated hemorrhagic colitis (AAHC), but the pathogenic role of the germ has not been established. The purpose of this study was to investigate the presence of K. oxytoca in patients with AAHC from a prospective cohort of patients with acute colitis, and to test the cytotoxicity on HEp-2 cells of K. oxytoca strains from patients with AAHC and healthy carriers.

METHODS

Colonic biopsy specimens and a sample of colonic fluid from 93 consecutive patients with acute colitis were cultured on selective media for 7 established pathogens and K. oxytoca. The 2 K. oxytoca strains isolated in the 4 patients with C. difficile-negative AAHC of this cohort and 105 additional K. oxytoca strains from patients with C. difficile-negative AAHC (n = 15) and healthy carriers (n = 90) were tested for cytotoxicity using a HEp-2 cell culture assay.

RESULTS

K. oxytoca was isolated in 50% (2 of 4) of the patients of the prospective cohort with C. difficile-negative AAHC compared with 2% (1 of 41) of the patients with acute colitis caused by established pathogens (P = 0.02). The rate of cytotoxic strains of K. oxytoca was higher in patients with AAHC (82%) than in healthy carriers (42%, P = 0.003).

CONCLUSIONS

We conclude that K. oxytoca is isolated with a significant high rate in patients with C. difficile-negative AAHC, and that K. oxytoca strains from patients with AAHC are cytotoxic more frequently on HEp-2 cells than strains from healthy carriers. These results strengthen the hypothesis of a causative role of K. oxytoca in some of the patients with AAHC.

Infectious diarrhea in human immunodeficiency virus

Chronic HIV-associated diarrhea is currently a field in flux. Improved noninvasive diagnostic tests, improved pathogen-specific regimens, and better empiric therapies may change some of the assumptions used to select algorithms for diagnostic evaluation and management. Any shift in the cause of diarrhea from pathogen-associated to idiopathic or a reduction in the overall incidence of diarrhea would have considerable impact. It is unclear how significant the problem of pathogen relapse in previous responders will become. Existing studies reviewed in this article show that the high diagnostic yield of endoscopy when stool tests are negative, coupled with significantly better outcomes when pathogens are identified, support the current practice of routine endoscopic evaluation. There currently are scant data on the economic impact of HIV-associated diarrhea as it relates to pathogen-specific and empiric therapy in the era of protease inhibitors. Such data would be integral to future evaluation of the impact of diagnostic and therapeutic strategies.