Study objective: To assess the usefulness of quantification of infected cells (ICs) in BAL fluid for the diagnosis of ventilator-associated pneumonia (VAP).
Design: A prospective study.
Setting: A medico-surgical ICU in a tertiary health-care institution.
Patients: One hundred thirty-two patients (mean age, 52 ±19 years). The suspicion of nosocomial pneumonia was strong in these patients: all had fever (>38.5°C), purulent tracheal aspirates, leukocytosis (^10,000 cells per cubic millimeter), and new or persistent radiographic lung infiltrates. Interventions: One hundred sixty-three samples (BAL and protected specimen brushes [PSB]) were obtained.
Results: VAP was present in 56 cases. The diagnosis was excluded in the remaining 107 cases. The IC count was performed on 100 cells in BAL fluid. The percentage of IC was significantly higher (12.6±12.4 vs 1.14±3.39; p<0.0001) in patients with pneumonia: the area under the receiver operating characteristic (ROC) curve was 0.888 and a threshold of 2% of IC corresponded to a sensitivity of 84%, a specificity of 80%, a positive predictive value of 69%, and a negative predictive value of 90%.
Conclusions: It is possible to define a threshold of IC in BAL fluid with a good reliability by using an ROC curve. This technique is useful for the early diagnosis (<2 h) of nosocomial bacterial pneumonia in mechanically ventilated patients and allows a rapid and appropriate treatment of most of the patients with suspected VAP.
Nosocomial bacterial pneumonia is a frequent and severe complication in mechanically ventilated patients. This event carries a high mortality rate” (30 to 71%). The diagnosis of these pneumonias is difficult. The clinical signs (temperature >38.5°C, purulent tracheal secretions, leukocytosis, new or persistent infiltrates on the chest radiograph) are neither sensitive nor specific.
A number of diagnostic methods have been evaluated. The optimal technique for the diagnosis of nosocomial pneumonia has not been determined, but there are some acceptable methods. The use of a protected specimen brush (PSB) is a method with a sensitivity of 90% and a specificity of 75% using 103 colony-forming units (cfu)/mL as a threshold.2 Quantitative culture of bacteria in BAL collected during a bronchoscopy is also a reliable method.
Early diagnosis is necessary to begin appropriate treatment as soon as possible. However results of quantitative cultures are not available until 24 to 72 h after the procedure. To solve the problem of delay, Chastre and colleagues/ in a study of 21 patients, showed that all patients with bacterial pneumonia had a count of infected cells (ICs) of 25% or greater in their BAL/PSB fluids. In 1989, in a new study^ of 61 patients using a cutoff value of 7%, these authors found a sensitivity of 86% and a specificity of 96%.
In another study of 72 patients and 80 BAL procedures, Aubas and coworkers tried to define a threshold of intracellular microorganisms using a receiver operating characteristic (ROC) curve: the area under the ROC curve (AUC) was 0.718; no cutoff point could be determined. The aim of our study was to define a diagnostic threshold of IC in 163 BAL samples (132 patients) by applying a ROC curve analysis.