C-arm Cone-beam CT-guided Needle Biopsies through the Erector Spinal Muscle for Posterior Thoracic Pulmonary Lesions

This study investigated retrospectively the diagnostic yield and complication rate of transthoracic needle biopsies for posterior thoracic pulmonary lesions using C-arm cone-beam computed tomography （CBCT）. The risk factors for pulmonary hemorrhage were evaluated. Our study included 113 patients with 113 posterior pulmonary lesions （mean longest diameter: 30.6mm, and mean depth: 4.7mm） through the erector spinal muscles using a 19/20-gauge coaxial system. The diagnostic performances of procedures for malignant lesions and the incidence of complications after biopsies were also assessed. The patient-related and procedure-related variables were investigated. Risk factors for pulmonary hemorrhage were analyzed with a multivariate logistic regression analysis. Findings revealed 99 malignant, 13 benign, and one intermediate lesion. Sensitivity, specificity, and diagnostic accuracy rates were 100％ （99/99）, 92.3％ （12/13）, and 99.1％ （111/112）, respectively. Air embolization, hemothorax, hemoptysis, pneumothorax, and pulmonary hemorrhage, occurred in 0, 2, 12, 48, and 70 procedures. The averaged spinous process-pleura depth and the traversed lung parenchyma depth achieved by the introducer needles were 54.2mm and 27.4mm, respectively. The needle position at the pleural puncture site within the intercostal space was in middle （31％） and inferior （69％） areas. The incidence of pulmonary hemorrhage was significantly higher in smaller lesions （p＝0.001）. Manual evacuation was performed in five procedures for patients with pneumothorax. The chest tube placement （trocar＞8 Fr） was performed in two procedures in patients with hemothorax and pneumothorax. In conclusion, the biopsy method with a posterior intercostal approach for posterior thoracic pulmonary lesions yielded high diagnostic accuracy and few major complications.