Aetiology and diagnostic techniques in intensive care acquired pneumonia: A Spanish multi-centre study

F. Alvarez-Lerma, M. Palomar, A. Martinez-Pellus, B. Alvarez-Sanchez, E. Perez-Ortiz, R. Jorda, F. Barcenilla, E. Maravi, B. Galvan, M. Palomar, J. Serra, B. Bermejo, A. Mateu, E. Quintana, M. Sanchez Palacios, R. Giral, V. Gonzalez, J. Lopez Mesa, J. A. Melgarejo, J. MartinezJ. Insausti, P. Olaechea, M. Chanovas, A. Gilabert, C. Junquera, J. Valles, F. Palacios, R. Calvo, E. Mesalles, J. Nava, A. Santos, S. Armengol, D. Marzo

Research output: Contribution to journalArticleResearchpeer-review

22 Citations (Scopus)


Objective: To determine the aetiology and diagnostic techniques routinely used in clinical practice in patients admitted to intensive care units (ICUs). Design: A prospective multi-centre study. Setting: Medical or surgical ICUs of 30 different hospitals in Spain. Subjects: All patients (n = 16,872) admitted to ICUs of the participating hospitals over a one-year period who developed ICU-acquired pneumonia and fulfilled Centers for Disease Control definitions for nosocomial infection. Main results: A total of 530 patients developed 565 episodes of pneumonia, with an incidence of 3.3 per 100 patients discharged from the ICUs. The incidence of pneumonia in mechanically ventilated patients was 8.74% compared with 0.37% in patients not needing ventilatory assistance (odds ratio 22.8). In 215 patients (40.6%), pneumonia developed during the first five days of ICU stay. At the time of clinical diagnosis of respiratory infection, 91.8% of patients were mechanically ventilated. The techniques used for the diagnosis of 565 episodes of pneumonia were endotracheal aspirates in 91.3% cases, blood culture in 84.6%, protected specimen brush in 30.1%, bronchoalveolar lavage in 12.9%, endotracheal aspirates taken immediately after intubation in 9.4%, culture of pleural fluid in 8%, serology in 6.9%, and histopathological examination in 1.4%. In 185 episodes, pathogens isolated were considered highly likely to be the cause of pneumonia, whereas in 264 episodes pathogens identified were considered of low probability. Pseudomonas aeruginosa, Staphylococcus aureus and Acinetobacter calcoaceticus were the most frequent. S. aureus, Streptococcus pneumoniae and Haemophilus influenzae accounted for 50% of isolates in early pneumonia, while late onset pneumonia was mainly due to P. aeruginosa infection. Crude mortality was 35.1% and attributable mortality 20.3 %. Conclusions: Invasive methods for the routine diagnosis of ICU-acquired pneumonia were only used in one-third of patients with suspected clinical manifestations. Forty per cent of patients developed pneumonia during the first five days of ICU stay. Gram-positive microorganisms predominated in early onset pneumonia, whereas nonfermenting Gram-negative pathogens occurred more frequently in later-onset pneumonia.
Original languageEnglish
Pages (from-to)164-170
JournalClinical Intensive Care
Issue number4
Publication statusPublished - 1 Jan 1997


Dive into the research topics of 'Aetiology and diagnostic techniques in intensive care acquired pneumonia: A Spanish multi-centre study'. Together they form a unique fingerprint.

Cite this