for people ages 18 years and up (full criteria)
at San Francisco, California
study started
estimated completion
Principal Investigator
by Jae Ho Sohn, MD
Headshot of Jae Ho Sohn
Jae Ho Sohn



Pneumothorax is the most common complication after CT-guided lung biopsy, and several techniques have been proposed to reduce the risk. Among them, rapid rollover is the maneuver to immediately reposition the patient, with biopsy-side down after removal of biopsy needle. It has been theorized that the technique reduces the size of alveoli surrounding the needle tract, leading to airway closure and reduction in the alveolar-to-pleural pressure gradient, thereby preventing pneumothorax. The aim of this study is to evaluate the effectiveness of rapid rollover in reducing the risk of radiographically detectable pneumothorax and the rate of chest tube insertion. Patients undergoing CT-guided lung biopsy for any indication will be recruited and randomized into either rapid rollover group or control group. In the control group, CT guided lung biopsy will be performed per standard protocols; in the rapid rollover group, the biopsy will also be performed per the same protocol with the addition of rapid rollover at the end of the procedure. For both groups, the primary outcome would be new or enlarging pneumothorax detected on post-biopsy radiographs, and the secondary outcome would be the risk of pneumothorax necessitating chest tube insertion, all complications associated with CT guided lung biopsy, time to complication development, and patient experience in each arm.

Official Title

Effect of Post-Procedural Rapid Rollover on Pneumothorax After Percutaneous CT-Guided Lung Biopsy: A Randomized Controlled Trial


All patients will undergo CT-guided lung biopsy by the best approach decided by the primary operator. In patients with the rapid rollover group, patient will be repositioned on the CT scanner table immediately after the co-axial needle is removed such that the biopsy site is down. Patient will be maintained in the same position for the post-biopsy CT scanner as well as in the post-procedure recovery area for at least 2 hours if patients can tolerate. All patients will be monitored for at least 2 hours, when a post-biopsy radiograph will be obtained to assess for pneumothorax. Interpreting radiologists of the post-biopsy radiographs will be blinded to whether the patient is in the control or intervention group. Multiple patient-, lesion- and technique-specific factors will be recorded: Patient-specific factors: - Age - Gender - Indication of lung biopsy - History of underlying lung disease - History of tobacco use - Prior lung surgery or radiation Lesion-specific factors: - Location - Size - Morphology - Shortest distance from pleura Technique-specific factors - Approach - Patient positioning during procedure - Number of times crossing the pleura - Number of times crossing the fissure - Number of fine needle aspirations (FNAs) - Number of core biopsies - Needle-path length - Needle-pleura angle - Co-axial size - Duration of biopsy site down time and number of turns as tolerated by the patient - Use of blood patch


Lung Cancer Pneumothorax CT-guided lung biopsy Chest tube Rapid Rollover


You can join if…

Open to people ages 18 years and up

  • All patients who are referred to the Department of Radiology and Biomedical Imaging Cardiac and Pulmonary Imaging section for CT-guided lung biopsy will be screened for eligibility. All eligible patients will be enrolled. Patients who develop intra-procedural pneumothorax will be analyzed but excluded from the final analysis.

You CAN'T join if...

  • Patients with chest wall, subpleural or mediastinal lesions
  • Patients who develop intra-procedural pneumothorax
  • Patients who are biopsied by a prone approach
  • Patients who are repositioned intra-procedurally
  • Patients who are unable to tolerate reposition or have spinal precautions
  • Patients who develop hemothorax and a chest tube is inserted for this indication


  • University of California, San Francisco
    San Francisco California 94143 United States

Lead Scientist at UCSF

  • Jae Ho Sohn, MD
    I am interested in the intersection of big data and radiology, specifically lung cancer screening, natural language processing, and other cardiothoracic imaging topics. As a physician with engineering background, I have been broadly involved in projects that use mathematical techniques to tackle research questions in radiology.


not yet accepting patients
Start Date
Completion Date
University of California, San Francisco
Study Type
Expecting 126 study participants
Last Updated