Category: Surgical ICU

Pugin et al have developed a clinical pulmonary infection score that combines six different individually weighted clinical indexes to determine the likelihood of pneumonia; its predictive value for the diagnosis of pneumonia approached that of bronchoscopic criteria. However, in the study by Pugin et al, the sample size was small; 13 patients with pneumonia were compared with 15 patients without pneumonia. No study, to our knowledge, has validated the specificity of the Pugin score in a large number of patients stratified by the etiology of pulmonary infiltrate. Our data show that a Pugin score > 6 virtually excluded pulmonary edema, acute lung injury, atelectasis, and contusion as likely etiologies of pulmonary infiltrates in the ICU patients. …Read the rest of this article

Pulmonary Infiltrates in the Surgical ICU: OutcomeThese data have implications for the election of empiric antibiotic therapy for patients with pneumonia in the ICU. Patients with pneumonia upon admission to the ICU or shortly thereafter are likely to have a community-acquired pneumonia pathogen, particularly if they are immunocompetent. Thus, ampicillin-sulbactam or a second-gene ration cephalosporin such as cefuroxime alone, as recommended in the American Thoracic Society Consensus Guidelines for community-acquired pneumonia, would suffice as therapy in such patients. However, patients hospitalized in the ICU for a prolonged period of time are likely to have Gram-negative bacteria or MRSA such that empiric therapy with third-generation cephalosporins, quinolones, aminoglycosides, or vancomycin would be appropriate. …Read the rest of this article

Haemophilus and pneumococcal carriage is common in healthy patients in the community. Trauma patients, unlike patients with chronic illnesses admitted to the ICU, are previously healthy. A lower incidence of Haemophilus/pneumococcal pneumonia in patients with chronic illnesses (and therefore prior or prolonged hospitalizations) may be reflective of the modification of the resident flora of the oropharynx and/or colonization with more resistant organisms as dominant flora as compared with trauma patients who may still be colonized with Haemophilus or pneumococcus. Data further corroborating this hypothesis is the fact that H influenzae/^ne\xmococcsl pneumonia in our study occurred significantly earlier during the ICU stay than all other pneumonias. …Read the rest of this article

Pulmonary Infiltrates in the Surgical ICU: DiscussionIt has been suggested that different ICU patient populations may have different risks for nosocomial pneumonia. In one study, patients with cardiotho-racic surgery were more likely to have ventilator-associated pneumonia than medical patients. We found that different ICU populations not only have unique predispositions to particular types of pulmonary infiltrates, but the causative agent for pneumonia may also differ between ICU populations. …Read the rest of this article

The following variables were significant predictors of mortality at 30 days (Table 3): male gender (p = 0.021), renal disease (p = 0.003), liver disease (p = 0.0001), absence of ulcer prophylaxis (p = 0.01), absence of enteral nutrition (p = 0.014), inotropic support (p = 0.0001), vasopressor support (p = 0.0001), lower hematocrit (p = 0.027), lower platelet count (p = 0.02), higher prothrombin time (p = 0.05), higher bilirubin (p = 0.001), higher creatinine (p – 0.001), APACHE III score (p = 0.0001), APACHE vital signs score (p = 0.0001), APACHE age/chronic illness scores (p = 0.023), OSFI score (p = 0.0001), acute lung injury as etiology or pulmonary infiltrate (p = 0.0001), and length of hospitalization (p = 0.02). By multivariate analysis (with gender, liver disease, serum creatinine, APACHE III score, acute lung injury as etiology, and OSFI in the model), only APACHE III score (odds ratio, 1.03; 95% confidence interval, 1.00 to 1.06, p = 0.003) and acute lung injury (odds ratio, 5.6; 95% confidence interval, 1.3 to 23.7, p = 0.01) were independently significant predictors of mortality. …Read the rest of this article

Pulmonary Infiltrates in the Surgical ICU: Extrapulmonary InfectionsExtrapulmonary Infections
Extrapulmonary infections occurred in 18% (7/39) of the patients with pneumonia, 24% (9/37) with edema, 20% (5/19) with acute lung injury, and 18% (3/17) of the patients with atelectasis. Bacteremia was observed in 7 of 39 patients with pneumonia and included 4 patients with MRS A, 1 with Enterobacter cloacae, and 1 with P aeruginosa bacteremia. Only one of seven bacteremias in patients with pneumonia was due to an isolate not associated with pneumonia (MRSA line sepsis in a patient with Legionella). Five percent (2/37) of the patients with pulmonary edema, 21% (4/19) with acute lung injury, and 6% (1/17) with atelectasis had bacteremia. …Read the rest of this article

A vast majority of the nonpneumonic pulmonary infiltrates were also in ventilated patients; 89% of acute lung injury, 88% of atelectasis, 70% of edema, and 75% of the contusion cases were in ventilated patients.
Nosocomial Pneumonia in the ICU
For this analysis, overt community-acquired pneumonias, ie, those occurring < 72 h after hospitalization, were excluded. Thirty-three of 39 pneumonias in the ICU occurred > 72 h after hospital admission; enteric Gram-negative bacteria (24%, 8/33) P aerus-ingosa (15%, 5/33), Haemophilus (15%, 5/33), and MRSA (12%, 4/33) were the most frequently occurring pathogens in nosocomial pneumonia in our study. Pneumococcal pneumonia occurred in 1 of 33 patients (3%). …Read the rest of this article