Surgical biopsy of suspected interstitial lung disease is superior to radiographic diagnosis

Background: Different modalities are used to diagnose interstitial lung disease. We compared the effectiveness of minimally invasive surgical biopsy versus high-resolution computed tomography for the diagnosis of interstitial lung disease and report the mortality of the procedure. Methods: We reviewed 194 patients undergoing video-assisted thoracoscopic lung biopsies for the suspicion of interstitial lung disease from January 2003 to February 2012 at Emory University. Demographics and patient characteristics were analyzed in addition to final diagnoses and clinical outcomes. Results: Concordance of radiographic diagnosis with final diagnosis was poor, matching pathologic diagnosis in 15% of cases, and specific diagnoses were included in the radiographic differential in only 34% of cases. A specific diagnosis was made after surgical biopsy in 88% of cases. Overall mortality of surgical biopsy was 6.7% (13/194). Major risk factors for death were preoperative supplemental oxygen, ventilator dependence, and age (p < 0.0001, p < 0.0001, and p = 0.03, respectively). Among patients with ventilator dependence preoperatively, the mortality rate was 100% versus 4.8% in patients not ventilator dependent. All biopsy specimens were concordant 91% of the time, and the first two biopsy specimens were concordant 96% of the time. Conclusions: Surgical biopsy should remain the gold standard for diagnosis of interstitial lung disease. The mortality is low with proper patient selection. More than two surgical biopsy specimens may not be needed because the concordance rates among pathologic specimens are very high.