2006 OPEN FORUM Abstracts
BRONCHOSCOPIC AND NON-BRONCHOSCOPIC LAVAGE IN CRITICALLY ILL CANCER PATIENTS WITH LUNG INFILTRATES: DIAGNOSTIC YIELD AND COST-EFFECTIVENESS
S. Egbert Pravinkumar MD, FRCP, FCCP, Clarence G.
Finch BS, BA, RRT , Natalie Binns BS, Thomas Kanneath,
RRT, Isaac Denning, RRT, Joseph L. Nates, MD, MBA-HCA, FCCM, Kristen J. Price
MD, FCCP
University of Texas- MD Anderson Cancer Center, Houston,
Texas
Introduction:
Ventilator Associated Pneumonia (VAP) is the most frequent ICU-acquired
nosocomial infection among ventilated patients with a mortality exceeding 50%.
Pulmonary infiltrates in cancer patients can be caused by infection, acute lung
injury and also by tumor, radiation and chemotherapy. Diagnosis of VAP by chest
x-ray has a sensitivity of 92% and specificity of only 33%, rendering
micobiological diagnosis a challenge [1]. So far, there is no gold standard
investigation for diagnosis of VAP. We investigated the diagnostic yield and
cost effectiveness of bronchoscopic bronchoalveolar lavage (B-BAL) and non-bronchoscopic
BAL (NB-BAL).
Methods: All
critically ill mechanically ventilated cancer patients with pulmonary
infiltrates and clinical suspicion of VAP, who underwent B-BAL or NB-BAL were
surveyed for positive microbiology and cost analysis, in the respiratory care
database. The B-BAL was performed by physicians and washings were obtained from
the radiographic sites of infiltrates. The NB-BAL from both lower lobe segments
was performed by respiratory therapist (RT) according to departmental protocol;
using 16 Fr Kimberly Clark's protected BAL catheter.
Results: Between 9/1/2004 and 1/31/2006, a total
of 327 patients underwent B-BAL and 109 patients had NB-BAL. The diagnostic
yield obtained by invasive B-BAL and non-invasive, NB-BAL techniques were 53%
and 49% respectively. Non-diagnostic yield with B-BAL and NB-BAL were 47% and
51% respectively. The total patient cost of performing B-BAL was US$2427 and
for NB-BAL was US$128. The common organism isolated were, Staphyloccocal
aureus, Escheria coli, Coagulase negative staphylococcus, Stenotrophomonas,
Candida species, and Mold.
Conclusion: The
study showed that the microbiological diagnostic yield obtained by RT performed
NB-BAL is comparable to physician performed B-BAL in selected group of
critically ill cancer patients. The availability of RT at the bed side, the
ease of NB-BAL procedure with similar diagnostic yield while costing 20 times
less than B-BAL, makes NB-BAL not only useful in critically ill cancer patients
with pulmonary infiltrates but also a cost effective alternative for B-BAL.