[Diagnosis and treatment of nosocomial pneumonia: bronchial fibroscopy, protected brushing and/or bronchial lavage is not indispensable].

Nosocomial pneumonia in ventilated patients is a frequent and serious complication of ventilatory assistance. The causal role of these lung infections in the overmortality observed in ventilated patients remains a question of debate, but the therapeutic cost (antibiotics and longer stay in intensive care) is considerable. Unlike the spontaneously ventilating patient, fever and new radiologic opacities in ventilated patients can be caused by many non-infectious conditions often concomitant with bacterial pneumonia. The association of clinical signs (fever, purulent tracheal secretions) and laboratory (leukocytosis, leukopenia, hypoxia) or radiographic (recent persistent alveoloar opacities) findings are suggestive of bacterial pneumonia but do not provide bacteriological proof. In order to avoid unjustified treatments (in patients without bacterial pneumonia) or poorly adapted treatments (broad spectrum empiric antibiotics) that can generate high costs in terms of therapy and epidemiology (emergence of resistance), a certain number of fibroscopic techniques have been proposed to improve the diagnosis of nosocomial pneumonia in ventilated patients. These sophisticated and attractive techniques have however provided rather disappointing results and do not allow sufficiently sure positive diagnosis nor bacteriological proof. Moreover, despite their cost and the difficulty of implementing fibroscopic techniques, no reduction in the mortality of nosocomial pneumonia in ventilated patients has been achieved.