Pulmonary Infiltrates in the ICU
abstract & commentary
Source: Singh, et al. Pulmonary infiltrates in the surgical ICU: Prospective assessment of predictors of etiology and mortality. Chest 1998; 114:1129-1136.
In a prospective study of 1148 consecutive patients admitted to the surgical ICUs of two teaching hospitals, Singh and colleagues identified 129 patients who subsequently developed pulmonary infiltrates. Eighty-three percent of the infiltrates occurred in patients receiving mechanical ventilation. Using predetermined definitions, patients were classified as having acute lung injury (including adult respiratory distress syndrome), pneumonia, pulmonary edema, or atelectasis. They collected clinical and physiologic data on patients, including diagnoses, co-morbidities, and Apache III and Organ System Failure Index.
There were 39 pneumonias (30% of infiltrates), of which six occurred within 72 hours of admission and were considered community acquired. Of the 33 nosocomial pneumonias, 13 were caused by gram-negative bacilli, four were caused by MRSA, and six were caused by Streptococcus pneumonia or Haemophilus spp. Haemophilus/pneumococcal pneumonia occurred significantly earlier after admission (4 days) than did pneumonia due to gram-negative bacilli (36 days) or MRSA pneumonia (42). In transplant patients, pneumonia was due to gram-negative bacilli, Staphylococcus aureus, or opportunistic pathogens. Interestingly, 38 out of 57 patients who had neither pneumonia nor an extrapulmonary site of infection received systemic antibiotics.
Mortality was 84% for patients with acute lung injury, 28% with patients with pneumonia, and 27% for patients with pulmonary edema. By multivariate analysis, only APACHE III score and acute lung injury were significantly associated with mortality.
Comment by robert muder, MD
Determining the etiology of acute pulmonary infiltrates in a patient in the ICU can be problematic. Using reasonably strict criteria for diagnosis, Singh et al determined that 30% of infiltrates were due to pneumonia; the remainder were due to noninfectious etiologies. Pneumonias occurring early in the ICU admission were typically due to community-acquired pathogens, whereas those occurring later were due to "typical" nosocomial bacteria such as gram-negative bacilli and MRSA. This should be helpful in guiding initial antibiotic therapy; broad spectrum therapy appears to be unnecessary in immunocompetent patients who have been recently admitted.
It is troubling, but not surprising, that the majority of patients with no evidence of pulmonary or extra-pulmonary therapy received antibiotic therapy for their pulmonary infiltrates. This suggests a major area for practice improvement. One might suggest that ICU patients with pulmonary infiltrates be critically evaluated for etiology so that antibiotics could be avoided, or at least discontinued after a brief course of initial therapy once true diagnosis of the infiltrate is clear. Determining the "true" diagnosis of pulmonary infiltrates is somewhat hampered by the lack of unequivocal "gold standards" for some of the major diagnostic entities, particularly pneumonia. However, studies such as this one form an important foundation for improvement in the management of pulmonary infiltrates in the ICU.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content