Microbiology specimens obtained at the time of surgical lung biopsy for interstitial lung disease: clinical yield and cost analysis.

OBJECTIVES

In efforts to obtain complete results, current practice in surgical lung biopsy (LB) for interstitial lung disease (ILD) recommends sending lung tissue samples for bacterial, mycobacterial, fungal, and viral cultures. This study assesses the value of this practice by evaluating the microbiology findings obtained from LB for ILD and their associated costs.

METHODS

A total of 296 consecutive patients (140 women, 156 men, median age=61 years) underwent LB for ILD from 2002 to 2009. All had lung tissue sent for microbiology examination. Microbiology results and resultant changes in patient management were analyzed retrospectively. A cost analysis was performed based upon nominal hospital charges adjusted on current inflation rates. Cost data included cultures, stains, smears, direct fluorescent antibody studies, and microbiologist consulting fees.

RESULTS

As many as 25 patients (8.4%) underwent open LB and 271 (91.6%) underwent thoracoscopic LB. A total of 592 specimens were assessed (range 1-4 per patient). The most common pathologic diagnoses were idiopathic pulmonary fibrosis in 122 (41.2%), cryptogenic organizing pneumonia in 31 (10.5%), and respiratory bronchiolitis ILD in 16 (5.4%). Microbiology testing was negative in 174 patients (58.8%). A total of 118 of 122 (96.7%) positive results were clinically considered to be contaminants and resulted in no change in clinical management. The most common contaminants were Propionibacterium acnes (38 patients; 31%) and Penicillium fungus (16 patients; 13%). In only four patients (1.4%), the organism cultured (Nocardia one, Histoplasma one, and Aspergillus fumigatus two) resulted in a change in clinical management. The cost of microbiology studies per specimen was $984 (€709), with a total cost for the study cohort being $582,000 (€420,000).

CONCLUSIONS

The yield and impact on clinical management of microbiology specimens from LB for ILD is very low. Its routine use in LB is questionable. We suggest it should be limited to those cases of ILD with a high suspicion of infection. Substantial cost savings are possible with this change in clinical practice.