Background: Patients with nosocomial infections in the intensive care unit (ICU) seem to have a poor prognosis. In this retrospective cohort study we investigated the relationship between weaning outcome, in-hospital mortality and the microbiological proof of nosocomial pathogens from secretions in mechanically ventilated patients in the years 2002 and 2006.
Patients and methods: 311 patients with long term (> 14 days) invasive (tube or tracheostomy) mechanical ventilation (MV) were enrolled in to the study when they had failed at least two weaning attempts prior to transfer. Microbiological proof of nosocomial pathogens from secretions sampled by the bronchoscope and an X-ray of the chest on admission day (in the transferring ICU and in our ICU) was collected from all patients.
Results: There was a significant decline of the weaning success rate between 2002 and 2006 (p = 0.001). The In-hospital mortality was higher in 2006 (p = 0.03). The microbiological proof of nosocomial pathogens had no influence on the weanability (exception: MRSA patients in 2002). In both years, patients with infiltrates on X-ray of the chest showed no increased mortality. But in 2006 it took longer to liberate these patients from invasive MV. In 2002 microbiological proof of pathogens was related to higher in-hospital mortality. In 2006, there was no difference concerning mortality in both groups.
Conclusions: Proof of nosocomial pathogens and infiltrates had no influence on the weanability of long-term mechanically ventilated patients. For in-hospital mortality, the results are contradictory.