Efficacy and safety of surgical lung biopsy for interstitial disease. Experience of 161 consecutive patients from a single institution in Italy.

ABSTRACT

BACKGROUND: The role of surgical biopsy for interstitial lung disease (ILD) is controversial, because of possible postoperative morbidity and mortality. We aimed to assess the efficacy and safety of surgical biopsy for ILD. METHODS: We retrospectively analysed the diagnostic performance and the postoperative complications of 161 consecutive surgical lung biopsy procedures carried out in suspected ILD cases that were undefined after multidisciplinary clinico-radiological evaluation. In 151 cases (93.8%) the biopsy was performed by video-assisted thoracoscopic surgery (VATS), in 6.2% by limited thoracotomy. RESULTS: A specific histological diagnosis was obtained in 154 (95.7%) of the surgically biopsied patients, while 4.3% remained histologically unclassified. The predominant histological patterns were sarcoidosis (29.8 %), usual interstitial pneumonia/idiopathic pulmonary fibrosis (UIP/IPF) (24.2%), cryptogenic organizing pneumonia (18.6%) and nonspecific interstitial pneumonia (8.1%). The postoperative course was uneventful in 142 cases. In 19 patients (11.8%) we observed postoperative complications, predominantly prolonged air leakage (5.0% of all cases). Thirty-day postoperative mortality was 3.1%, mostly due to acute exacerbation of respiratory insufficiency. Postoperative mortality independently correlated with preoperative need of oxygen therapy (OR, 5.21; 95% CI, 1.19-22.95) and with UIP/IPF histology (OR, 5.67; 95% CI, 1.27-25.25). CONCLUSIONS: Lung biopsy was performed mostly by VATS, with limited morbidity, and was effective in yielding a specific histologic diagnosis in the vast majority of undefined ILD cases. To optimize the outcome of surgical biopsy for specific diagnosis of ILD, this procedure should be performed only exceptionally in patients with critical respiratory illness as postoperative mortality risk in these subjects is exceedingly high.