Dear Sir: The authors address two issues in the recent European experts' statement on nosocomial pneumonia.

  1. (1)

    The statement questions the term “VAP” since it appears misleading to refer to the ventilator when addressing nosocomial pneumonia in intubated patients. However, we are not aware of data supporting disease severity as the main determinant in the development of VAP. It is also questionable whether pneumonia severity is associated with distinct pathogen and resistance patterns. The suggestion to change "ventilator-associated" into "ventilation-associated" and to differentiate further intubation- and tubus-associated pneumonia closely follows pathogenetic pathways, reflects different potential pathogen and resistance patterns and, therefore, helps the clinician to select initial empiric antimicrobial treatment. However, it is crucial to remember that this concept is based on data showing that nosocomial colonization begins at hospitalization and that comorbid conditions as well as previous antimicrobial treatment may change normal endogenous colonization. Therefore, the timetable of intubation- versus tubus-associated pneumonia must begin with hospitalization, not intubation, and comorbidity and antimicrobial treatment have to be taken into account.

  2. (2)

    Selective digestive decontamination (SDD) remains a matter of controversy. Evidently, systemic antimicrobial treatment is not “selective,” and, therefore, the three-stage approach to prevent nosocomial pneumonia described by the authors of the letter should not be addressed as SDD.

We agree that a reduction in mortality has only been found for SDD; however, this is only true for “SDD” including systemic antimicrobial treatment. The reasons why other measures could not be shown to reduce mortality are complex and probably not exclusively related to their limited effects. Concerns about selection pressure and its long-term consequences (particularly by systemically administered cephalosporins) preclude general acceptance of SDD. Whereas it still may applicable in countries with low MRSA prevalence, there are serious concerns in the large number of countries with high prevalence. In addition, the growing threat of pathogens producing ESBL in many countries and regions makes SDD even more problematic.