Contribution of blinded, protected quantitative specimens to the diagnostic and therapeutic management of ventilator-associated pneumonia

Chest. 2005 Aug;128(2):533-44. doi: 10.1378/chest.128.2.533.

Abstract

Objective: Sampling techniques for microbiological diagnosis of ventilator-associated pneumonia (VAP) remain debated, and it is unclear to what extent invasive diagnostic techniques impact the management of patients.

Design: A prospective observational study of 68 first episodes of suspected pneumonia in which specimens were obtained blindly (endotracheal aspirate [EA] and blinded protected telescoping catheter [PTC]) and via bronchoscopy (directed PTC bronchoscopy and BAL), and in sequence, and the results were provided to the attending physicians in the same order. Therapeutic plans resulting at each step were examined, and their adequacy was assessed using quantitative BAL fluid culture as the diagnostic standard.

Participants: Sixty-eight patients with clinically suspected VAP hospitalized in two ICUs in a tertiary care university hospital.

Results: There were 35 patients (51%) with VAP confirmed by BAL fluid culture (13 early onset and 22 late onset). EA specimens grew organisms (light growth or more) in all BAL-confirmed VAP cases and 59% of nonconfirmed cases, whereas the sensitivity and specificity of blinded PTC quantitative cultures were 77% and 97%, and did not differ from those of directed PTC cultures (77% and 94%, respectively). Antibiotic therapy based on the clinical severity and likelihood of VAP, Gram stain results, and early blinded PTC culture results was adequate in 54% (19 of 35 VAP patients) within 2 h of sampling and 80% (28 of 35 patients) within 24 h; therapy was revised in only 3 more patients following BAL culture results. New antibiotics were introduced within the first 24 h in 14 of 33 nonconfirmed episodes (42%), and antibiotics were withheld or withdrawn within 48 h in 23 episodes (70%); three of these patients-with both blinded PTC and BAL growing organisms below the threshold-had early subsequently confirmed pneumonia with the same organism.

Conclusions: A therapeutic approach guided by quantitative cultures of blinded specimens helps achieve early adequate management of approximately 90% of patients suspected of having VAP.

MeSH terms

  • Bronchoalveolar Lavage Fluid
  • Bronchoscopy
  • Humans
  • Middle Aged
  • Pneumonia, Bacterial / diagnosis*
  • Pneumonia, Bacterial / etiology
  • Pneumonia, Bacterial / therapy*
  • Reproducibility of Results
  • Specimen Handling*
  • Ventilators, Mechanical / adverse effects*