A double-blind randomized placebo-controlled study was carried out to evaluate the efficacy and the cost of selective digestive decontamination (SDD) to prevent nosocomial pneumonia in multiple-trauma patients. Nosocomial infections, particularly pneumonia, were more frequent in the placebo group. The most common infectious agent was Staphylococcus: Staphylococcus aureus in the placebo group and Staphylococcus epidermidis in the SDD group. Methicillin-resistant Staphylococcus epidermidis was detected more often in the SDD group. No methicillin-resistant Staphylococcus aureus was observed in this study. Fewer patients in the SDD group required antibiotherapy. SDD resulted in a saving of about 41% in drug expenditure.