PARIS, FRANCE--In patients with suspected ventilator-associated pneumonia (VAP), a more invasive diagnostic approach involving fiberoptic bronchoscopy or bronchoalveolar lavage may reduce mortality, decrease antibiotic use, and stave off organ failure, a new study suggests. But some experts question whether this study's results are strong enough to warrant routine adoption of the more invasive approach.

In an uncontrolled multicenter trial, researchers led by Jean-Yves Fagon, MD, of the Hôpital Broussais in Paris randomized 413 patients with suspected VAP to either the more invasive approach or the traditional, diagnostic method (clinical assessment and microscopy of endotracheal aspirates).[1] Patients were then followed for 28 days.

Of the study subjects, 204 were randomized to more invasive diagnosis and 209 to traditional diagnosis. In the invasive group, 90 cases of VAP were confirmed after samples underwent quantitative culture. In the traditional group, endotracheal aspirate cultures were positive in 179 patients.

At day 14, mortality was lower in the invasive group than in the traditional group (16.2% vs 25.8%). In addition, organ failure developed less often in the invasive group during the study's first week; the mean Sepsis-related Organ Failure Assessment (SOFA) scores were lower in this group than in the traditional group on day 3 (6.1 vs 7.0) and day 7 (4.9 vs 5.8 ). All these differences were significant.

At the end of the study, mortality was still lower in the invasive group (30.9% vs 38.8%), but the difference was no longer significant. And the mean SOFA scores in the two groups were identical (3.1). But the invasive group still had a markedly lower mean number of days of antibiotic treatment (12.8 vs 14.9) and number of antibiotics received per day (1.0 vs 1.3). This group also had a lower rate of Candida colonization or infection (11.3% vs 22.6%). Shorter and more appropriate antibiotic therapy resulting from greater diagnostic accuracy may have been responsible for the invasive group's better outcomes.

PROJECTING TRENDS

"This was a provocative, well-done study," Michael Matthay, MD, recently told PULMONARY REVIEWS. "However, it probably won't change practice for most clinicians in the United States in terms of convincing them to go right to bronchoscopy or bronchoalveolar lavage to diagnose VAP."

The more invasive approach, he explained, did not significantly lower mortality at 28 days, an end point most clinicians consider more important than 14 days. "However, there was a trend toward reduced mortality at 28 days for the more invasive approach," acknowledged Dr. Matthay, associate director of the intensive care unit at the University of California at San Francisco Medical Center.

This trend did reach significance in a multivariate proportional hazards regression analysis. However, this analysis is somewhat unconventional, said Dr. Matthay. The real take-home message of this study, he suggested, is that clinicians should look harder for urinary tract, venous catheter, and other nonpulmonary infections even when initial cultures for VAP are positive. Indeed, Fagon et al documented 22 nonpulmonary infections in the invasive group between study days 1 and 3.

--Timothy Begany

Reference
1. Fagon J-Y, Chastre J, Wolff M, et al. Invasive and noninvasive strategies for management of suspected ventilator-associated pneumonia. Ann Intern Med. 2000;132:621-630.

Return to table of contents