[Selective digestive decontamination in patients under reanimation].

Nosocomial infections increase morbidity and mortality in hospitalized patients. ICU patients are at high risk of sustaining them, due to the high rate of invasive procedures and their poor health state. Conventional methods for decreasing the incidence of infection in ICU patients include handwashing, catheter care, strict antibiotic policy, and reduction of environmental sources of infection. Despite these measures, the colonization in these patients is always high, because of the presence of pathogens in the own patients' flora. Nosocomial pneumonia which is a major cause of mortality in ICU patients arises from retrograde colonization of the lung by pathogens originating from oro-pharyngeal and gastric secretions. Since 1984, selective decontamination of the digestive tract (SDD) has been advocated in ICUs to prevent from bacterial and fungal gastrointestinal/oropharyngreal colonization, nosocomial infection, subsequent multiple organ failure (MOF) and death. The SDD regimen is usually an extemporaneously prepared suspension of antimicrobial agents. Appropriate antibiotics for this regimen should ideally be nonabsorbable, to prevent from the development of resistant pathogens and avoid systemic toxicity. They should also be able to selectively eliminate enterobacteriaceae and yeasts, without decreasing the protective anaerobic flora. The most used combination is a suspension of colistin, amphotericin B and aminoglycoside, administered four times day through the nasogastric tube, in association with a paste consisting of 2 p. 100 colistin/amphotericin B/aminoglycoside, applied to the oropharynx. A parenteral antibiotic is also often co-administered during the first four days to prevent from early infections until the SDD regimen reaches its full effect; cefotaxime is usually used for this. SDD significantly decreases colonization rates in the oropharynx, gastrointestinal (GI) tract and trachea. This effects is primarily attributable to a decrease of Gram-negative bacilli (GNB) and yeasts, although several studies also reported decreased isolates of Gram-positive cocci (GPC). Oropharyngeal and GI colonization significantly decrease after four days of such a regimen, but tracheal decontamination in uncertain. Several studies recognized an emergence of GPC during or after SDD and resistance occurrence in GNB (especially against aminoglycosides). Recolonization occurs rapidly, about 4 to 8 days after the discontinuation of SDD. SDD decreases significantly the nosocomial infections, especially Gram-negative pneumonia. This benefit is most obvious in trauma patients, severely burned patients and after orthopic liver transplantation. Several studies reported a significant decrease in the overall rate of infections, especially extrapulmonary infections, including blood, urinary tract, wounds, abdominal, and catheter related infections. Despite a major decrease in infection rates with SDD, most studies did not show lowered mortality rates.(ABSTRACT TRUNCATED AT 400 WORDS)