[Comparative study of three selective methods for nosocomial pneumonia surveillance in non-critical adult patients]

Among nosocomial infections, pneumonia is one of the most important ones due to the mortality, morbidity and the increase of costs that it causes. The implementation of a cost-effective surveillance system for these infections is a challenge for hospitals. The objective of this work was to evaluate three selective methods of surveillance for nosocomial pneumonias in non-critical adult patients.

A prospective surveillance of nosocomial pneumonias was carried out during 4 months in an university hospital of the Valencian Community, by means of three methods based on reports of chest x-ray, consumption of several antibiotics and health-care workers' liaison. Subsequently, possible cases were confirmed through revision of patient's charts, comparing each method with the reference one.

Of 541 possible cases of pneumonia, 27 were confirmed. The cumulative incidence was 3.67 cases/1,000 admissions and the incidence density was 5.9 cases/10,000 patient-days. The method with the highest sensitivity was that dealing with the consumption of antibiotics (85.2%), followed by x-rays (70.4%). The method based on the personnel's liaison had 99.4% specificity, while that of x-rays was 75.5%. Negative predictive values were above 95% for the three methods.

A surveillance system for nosocomial pneumonia based on reports of chest x-rays may be very efficient, providing the necessary information to program and to evaluate the prevention and control activities for these infections in the hospital.