[Diagnosis and treatment of nosocomial pneumonia: bronchial fibroscopy, protected brushing and/or bronchial lavage is indispensable]

Abstract

In case of suspected nosocomial lung infection in an intensive care patient, a strictly "clinical" strategy based on the physical exam and qualitative culture of tracheal secretions has three potential drawbacks: i) Numerous patients are given unnecessary antibiotic regimens unwarranted on the sole basis of colonization of the upper airways. In addition to the high cost of such regimens, these antibiotics also have important adverse effects favoring the emergence of multiresistant strains, and in addition, considerably complicating the search for the real cause and the treatment of the fever and the new lung images. ii) This strategy exposes the patient to diagnostic errors by omission if all patients with suspected infection are not treated. iii) Finally, even if the diagnosis of pneumonia is exact, the antibiotic prescribed on the basis of necessarily imprecise information can be most inappropriate or non-optimal in many cases. It is particularly difficult or often impossible to taper off the regimen on day 3 when tracheal cultures become available since the prescription would have to cover all the germs present in the upper airways even though only certain strains are undoubtedly causing the infection in the lung parenchyma. We prefer another strategy based on systematic bronchial fibroscopy before changing the antibiotic regimen in patients presenting signs suggestive of nosocomial pneumonia. The fibroscope allows us to obtain minimally-contaminated or non-contaminated specimens of the distal lung secretions in the suspected zone using the "Wimberley" brush for a protected sample and/or bronchoalveolar lavage. Although data in the literature on the usefulness of protected brushings and bronchoalveolar lavage specimens is rather confusing due to methodological differences, all the published work confirms that these techniques do allow identification of the germs present in the lung at the time of the sampling. Thus, on the basis of the culture results, it would be possible to discontinue any unnecessary antibiotic in patients with negative samples and inversely to use the narrowest possible treatment spectrum for those with positive samples, adapting the regimen to the susceptibility pattern or identified strains. The strongest arguments favoring this invasive strategy come from a multicentric French study on 413 patients with clinically suspected nosocomial pneumonia under artificial ventilation. The results demonstrated lower mortality in the fibroscopy group associated with a major reduction in the consumption of antibiotics.