The article by Dalal et al1 published in this issue of Infectious Diseases in Clinical Practice raises awareness of an oftentimes-forgotten gastrointestinal pathogen: Staphylococcus aureus. Although stories of Clostridium difficile diarrhea and infection have become ubiquitous in the media because of its rampant spread and attributable deaths in 2004 and 2005, S. aureus, particularly methicillin-resistant S. aureus, assumed a similar role with respect to recurrent skin and soft tissue infections and necrotizing pneumonias. For these reasons, and the fact that S. aureus enteritis is usually associated with food poisoning, many clinicians have neglected the potential role of S. aureus as a gastrointestinal pathogen that causes a clinical presentation very similar to infection by C. difficile. Whether there is a correlation between skin colonization of S. aureus and gastrointestinal colonization or diarrheal illness is not known.
As reported by Dalal et al,1 diarrhea due to S. aureus is not uncommon. Multiple citations have demonstrated its role as a pathogen, either methicillin-resistant or methicillin-sensitive and antibiotic-associated or sporadic.2-5 Presentation can be very similar, if not identical, to a typical patient with C. difficile infection with multiple liquid stools, leukocytosis, fever, and abdominal pain. The difficulty lies in that C. difficile is much more common as an antibiotic-associated infectious diarrhea pathogen than S. aureus. It is also problematic that S. aureus is often a member of the normal microbiologic flora of the gastrointestinal tract.6,7 In addition, clinicians who use oral vancomycin as their primary treatment of choice for C. difficile infection may be concurrently and inadvertently treating S. aureus and C. difficile. Dalal et al1 report their use of Gram stains of fecal smears in conjunction with fecal cultures. Because of the presence of S. aureus as normal enteric flora, the stain must be interpreted with caution. Most microbiology laboratories will not report growth of S. aureus in the stool or the presence of organisms resembling S. aureus on fecal stains unless specifically requested, and it cannot be assumed that they will. In addition, because there are no specific guidelines to help determine the amount of growth that should be considered sufficient or likely to be causative for diarrheal disease, these cultures must also be interpreted with caution. Again, the same must be said of fecal stains. Few studies, if any, have made direct correlations of the colony counts with the incidence of diarrhea. However, abstracts and case series that have been published suggest that growth of moderate to predominant S. aureus should be considered a pathogen in the appropriate clinical setting especially in those cases where patients have had presumptive C. difficile infection with nonresponse to metronidazole and persistently negative C. difficile toxin assays.2,4,5 These same studies did report the presence of enterotoxin; however, testing for the entire panel of S. aureus enterotoxins is not routinely available in most hospital laboratories.
With the difficulties mentioned and the relative uncertainty as to the prevalence of the pathogen as a cause of antibiotic-associated diarrhea, S. aureus should be considered a pathogen if there are supporting clinical factors. These should include patients who (1) have a diarrheal disease, (2) have at least moderate S. aureus reported on their fecal cultures, (3) have a diarrheal disease that is unresponsive to metronidazole, and (4) may have persistently negative C. difficile toxin assays. Guidelines in development on the treatment of C. difficile infection will likely make the recommendation that severe C. difficile infection should be treated with oral vancomycin; this will effectively cover S. aureus as a pathogen even if it is not considered to be one. In addition, starting doses of 125 mg vancomycin orally every 6 hours should provide sufficient intracolonic levels to treat S. aureus because it is not well absorbed. The use of C. difficile common antigen testing may also be beneficial because its high negative predictive value may allow clinicians to exclude C. difficile and to look for other causes earlier.
At present, because of the apparent relative infrequency of S. aureus antibiotic-associated diarrhea, it is not necessarily practical to test every patient with antibiotic-associated diarrhea or known C. difficile infection for S. aureus. Certainly, any growth of S. aureus in stool specimens may confuse the issue. Circumstances mentioned earlier may help identify those patients who should be tested. It may provide more support, however, for the argument that oral vancomycin should be the treatment of choice for patients with suspected antibiotic-associated diarrhea due to an infectious agent.
1. Dalal A, Urban C. Enterocolitis caused by methicillin-resistant Staphylococcus aureus
. Infect Dis Clin Pract
2. Boyce JM, Havill NL. Nosocomial antibiotic-associated diarrhea associated with enterotoxin-producing strains of methicillin-resistant Staphylococcus aureus
. Am J Gastroenterol
3. Flemming K, Ackermann G. Prevalence of enterotoxin producing Staphylococcus aureus
in stools of patients with nosocomial diarrhea. Infection
4. Gravet AM, Rondeau C, Harf-Monteil F, et al. Predominant Staphylococcus aureus
isolated from antibiotic-associated diarrhea is clinically relevant and produced enterotoxin A and the bicomponent toxin LukE-lukD. J Clin Microbiol
5. Ledoux D, Labombardi V, Turett G, et al. Oral Abstract 111: Relationship of Enterotoxin Producing
Staphylococcus aureus With Antibiotic Associated Diarrhea
. Paper presented at: 42nd Annual Meeting of IDSA; September 30-October 3, 2004; Boston, MA.
6. Murray PR, Rosenthal KS, Pfaller MA. Commensal and pathogenic microbial flora in humans. In: Murray PR, Rosenthal KS, Pfaller MA, eds. Medical Microbiology
. 5th ed. Philadelphia, PA: Elsevier Mosby; 2005:83-87.