Diagnostic Yield and Complications of Ultrasound-Guided Trans-Thoracic Biopsies by Pulmonologists: A Single-Center Retrospective Study of Lesions with Pleural Contact.

Introduction: Ultrasound-guided trans-thoracic needle biopsy (US-TTNB) is a method of choice for the diagnostic management of peripheral lung lesions and pleural masses for pulmonologists. If complication risk factors and diagnostic yield have been well reported for CT-guided biopsies, publications for US-TTNB in this field are very scarce.

Methods: The primary objective of this study was to describe the diagnostic yield of US-TTNB carried out by pulmonologists defined by a definitive histopathological diagnosis. Secondary objectives were to identify factors that may influence diagnostic yield and to describe complications of this procedure. Between September 2015 and December 2022, charts of consecutive patients presenting peripheral lesion with pleural contact and having undergone US-TTNB were retrospectively analyzed. ROC curves were performed to assess the probability of having a contributing biopsy (definitive histology) depending on scannographic measurements and the number of punctures. Univariate and then multivariate analyses were performed to look for variables associated with complications.

Results: One hundred and fifty-nine patients were enrolled in this study. Among them, diagnosis was obtained for 140 patients (88% success rate). The histology was in favor of a neoplastic process in 96% of cases (135/140). Analysis of the ROC curves showed that the depth of the lesion (AUC 75%; 95% CI: 65-85; cut-off 3.0 cm), the width of the lesion (AUC 73%; 95% CI: 63-86; cut-off 3.9 cm), the pleural contact (AUC 68%; 95% CI: 57-80; cut-off 3.8 cm), and the number of biopsies (AUC 70%; 95% CI: 59-81; cut-off 3 biopsies) were the variables associated with diagnostic yield. Complications occurred for 27 patients (17%), mainly pneumothorax (6%), hemoptysis (6%), and sepsis (6%). Univariate analysis showed a significant association between pneumothorax and lesion depth (OR 0.68; 95% CI: 0.65-0.92; p = 0.03) with a predicted probability >5% for a depth <4 cm according to general linear model analysis. Univariate and multivariate analysis revealed a significant association between the number of biopsies and the risk of sepsis (OR 1.90; 95% CI: 1.19-3.26; p = 0.01) with a predicted probability >5% for more than 4 biopsies.

Conclusion: US-TTNB is a reliable diagnostic procedure that can be performed by pulmonologists. The depth, width, pleural contact of the lesion, and the number of biopsies are key variables associated with diagnostic yield. The depth of the lesion is associated with the risk of pneumothorax, and the number of biopsies with the risk of pulmonary sepsis.