• Luis Fernando Aranha Camargo, Fernando Vinícius De Marco, Carmen Sílvia Valente Barbas, Cristiane Hoelz, Marco Aurélio Scarpinella Bueno, Milton Rodrigues, Verônica Moreira Amado, Raquel Caserta, Marinês Dalla Valle Martino, Jacyr Pasternak, and Elias Knobel.
• Intensive Care Unit, Hospital Israelita Albert Einstein, São Paulo, Brasil. luis_camargo@einstein.com.br
• Crit Care. 2004 Dec 1;8(6):R422-30.

IntroductionDeferred or inappropriate antibiotic treatment in ventilator-associated pneumonia (VAP) is associated with increased mortality, and clinical and radiological criteria are frequently employed to establish an early diagnosis. Culture results are used to confirm the clinical diagnosis and to adjust or sometimes withdraw antibiotic treatment. Tracheal aspirates have been shown to be useful for these purposes. Nonetheless, little is known about the usefulness of quantitative findings in tracheal secretions for diagnosing VAP.MethodsTo determine the value of quantification of bacterial colonies in tracheal aspirates for diagnosing VAP, we conducted a prospective follow-up study of 106 intensive care unit patients who were under ventilatory support. In total, the findings from 219 sequential weekly evaluations for VAP were examined. Clinical and radiological parameters were recorded and evaluated by three independent experts; a diagnosis of VAP required the agreement of at least two of the three experts. At the same time, cultures of tracheal aspirates were analyzed qualitatively and quantitatively (10(5) colony-forming units [cfu]/ml and 10(6) cfu/ml)ResultsQuantitative cultures of tracheal aspirates (10(5) cfu/ml and 10(6) cfu/ml) exhibited increased specificity (48% and 78%, respectively) over qualitative cultures (23%), but decreased sensitivity (26% and 65%, respectively) as compared with the qualitative findings (81%). Quantification did not improve the ability to predict a diagnosis of VAP.ConclusionQuantitative cultures of tracheal aspirates in selected critically ill patients have decreased sensitivity when compared with qualitative results, and they should not replace the latter to confirm a clinical diagnosis of VAP or to adjust antimicrobial therapy.

42 keywords...

hide…