A New Classification of Infections in Ventilated Patients in the ICU?
ABSTRACT & COMMENTARY
Synopsis: A more accurate distinction between nosocomial and community-acquired infections could be made if, instead of simply applying the conventional 48 h cut-off point, carriers of potential pathogens were detected on admission to an ICU.
Source: Murray AE, et al. Infections in patients requiring ventilation in intensive care: Application of a new classification. Clin Microbiol Infect 1998;4:94-99.
Over a three-month period, 104 patients in a mixed medical and surgical ICU were screened on admission for the presence of potentially pathogenic micro-organisms (PPM) acquired in the community (Staphylococcus aureus, Haemophilus influenzae, Streptococcus pneumoniae, Escherichia coli) or the hospital (Klebsiella spp., Enterobacter spp., Serratia spp., Citrobacter spp., and Pseudomonas spp.) by a stool sample and a throat swab. These surveillance cultures were repeated every Monday and Thursday during admission, and other specimens were obtained for diagnostic purposes as and when required. An infection was considered exogenous when the PPM involved had not been recovered from surveillance cultures. Otherwise, an infection was considered primary endogenous when the PPM had been present in surveillance cultures on admission (the carrier state) or secondary endogenous when infection was caused by PPMs acquired in the ICU. Twenty-one patients stayed in the ICU for longer than three days and eight developed 12 infections, of which five were considered primary endogenous and the remainder secondary endogenous (See Table).
Only three episodes developed within 48 h of admission and would have been classified as "community acquired" using the conventional CDC criteria, and three episodes of secondary endogenous infections would have been classified as "hospital acquired" even though the PPMs were already present on admission to the ICU. This alternative means of classifying infections in the ICU might be more appropriate than the conventional CDC criteria, especially as endogenous infections will not be prevented by handwashing since, unlike exogenous infections, the patient is the source of his own infection.
COMMENT BY J. PETER DONNELLY, PhD
This paper has already provoked a vigorous criticism in the form of a letter in the same issue (Spencer RC. Clin Microbiol Infect 1998;4:100-101) because Murray and associates use definitions of carriage and infections that are regarded as insecure and assert that handwashing has never contributed to infection control-a comment guaranteed to dismay and even infuriate, most if not all, of hospital hygiene and infection control practitioners. But is this fair? If patients in the ICU are, in fact, infecting themselves with the PPMs they already carry on their mucosal surfaces, then alternative measures are going to be necessary. Handwashing is clearly good clinical practice and helps limit the spread of infectious agents but not if they are carried in the recesses of the patient's own body. This is analogous to the neutropenic patient who is also more often than not infected by PPMs from among his own commensal flora. The analogy can be carried further. Selective decontamination of the digestive tract has long been practiced successfully in Europe by hematologists and many in the intensive care fraternity as a means of controlling infection caused by the endogenous PPMs, such as Escherichia coli, and other enterobacteria, Pseudomonas aeruginosa and Staphylococcus aureus. However, the success is limited to lowering the rate of microbiologically defined infections due to these PPMs and does not extend to reducing overall morbidity and mortality or in lowering the rate of other opportunistic infections. Hence, the data fail to convince the skeptic and devotee alike, but the latter still believe in selective digestive tract decontamination. By contrast, the "48 h rule" does not address the source of PPMs at all and arbitrarily assigns the cause of infections to either the community or hospital. It may be a coincidence, but adherents of the conventional wisdom also tend to be fierce critics of SDD. They may be right in both respects, but there are no convincing data either way. The "48 h rule" does seem neither appropriate nor useful for classifying infections in immunocompromised patients, such as those with chemotherapy-induced neutropenia and those in the ICU, but it is the only one widely recognized and approved. The challenge to those with a different view is to conduct studies of sufficient quality to prove the alternative approach beyond reasonable doubt.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content