The diagnosis of ventilator-associated pneumonia (VAP) at the bedside is difficult, but no more so than the diagnosis of meningitis in patients with external intraventricular devices, or of urinary tract infections in patients with a bladder catheter. In VAP, somewhat surprisingly, the exceptions have been overemphasized. In addition, most investigators have neglected to evaluate the key role of the quality of the sample provided. Others have expended considerable effort on evaluating the representativity of diagnostic tools in autopsy studies, but information gathered from moribund patients receiving antibiotics for a nonrecent episode is not representative of the case of other patients in the correct clinical setting.

In recent decades, much has been learned about the diagnosis of VAP, but little about the impact of diagnostic techniques on outcome. The growing number of articles about diagnosis, far from improving our understanding, have only added to the confusion. Differences in populations studied or in the timing of the performance of the technique, poor standardization of techniques, the absence of any appraisal of the quality of the samples, and the absence of a representative gold standard have made it very difficult to compare the different studies reliably, and preclude the generalization of findings. In 1999, despite almost 20 years of clinical practice with “invasive diagnostic techniques,” only five recent investigations<sup>2Alvarez-Lerma F and ICU-Acquired Pneumonia Study Group Modification of empiric antibiotic treatment in patients with pneumonia acquired in the intensive care unit Intensive Care Medicine 1996 ;  22 : 387-394

Cliquez ici pour aller à la section Références, 22Kollef M.H., Ward S. The influence of mini-BAL cultures on patient outcome: Implications for the antibiotic management of ventilator-associated pneumonia Chest 1998 ;  113 : 412-420 [cross-ref]

Cliquez ici pour aller à la section Références, 24Luna C.M., Vujacich P., Niederman M., et al. Impact of BAL data on the therapy and outcome of ventilator-associated pneumonia Chest 1997 ;  111 : 676-685 [cross-ref]

Cliquez ici pour aller à la section Références, 43Rello J., Gallego M., Mariscal D., et al. The value of routine microbial investigation in ventilator-associated pneumonia Am J Respir Crit Care Med 1997 ;  156 : 196-200

Cliquez ici pour aller à la section Références, 52Sanchez-Nieto J.M., Torres A., García-Cordoba F., et al. Impact of invasive and noninvasive quantitative culture sampling on outcome of ventilator-associated pneumonia: A pilot study Am J Respir Crit Care Med 1998 ;  157 : 371-376

Cliquez ici pour aller à la section Références</sup> have focused on evaluating the impact on outcome. As the authors noted in a recent update, <sup>16Gallego M., Vallés J., Rello J. New perspectives in the diagnosis of nosocomial pneumonia Curr Opin Pulm Med 1997 ;  23 : 116-119 [cross-ref]

Cliquez ici pour aller à la section Références</sup> more time has been spent on looking for “El Dorado” than on evaluating the impact on outcome of diagnostic tests.

With these precedents, and with the confusion created, it is not surprising that most clinicians have missed the real nature of the problem. In this article, the authors review the controversial topic of diagnostic testing of pneumonia in intubated patients as presented in clinical practice, based on their own experience. Specifically, they reply to the following key questions: (1) When should an etiologic investigation be performed? (2) Should an intubated patient with suspicion of pneumonia be treated? (3) How should the microbiologic findings be interpreted? (4) What is the potential impact of diagnostic techniques on outcome?