Breath Hold Not Linked to Better Lung Lesion Biopsy Yield

PHOENIX — Sustained breath holding while taking lung nodule biopsies by robotic bronchoscopy didn’t improve diagnostic yield, a small randomized trial showed.

Diagnostic yield by strict American Thoracic Society/American College of Chest Physicians (ACCP) criteria was 86.8% (46 of 53 patients) with breath hold compared with 92.2% (59 of 64) using conventional ventilation (P=0.338).

Localization rates too were similar at 98.1% with breath hold and 95.3% with conventional ventilation (P=0.625), reported Roshen Mathew, MD, of WVU Medicine Camden Clark Medical Center in Parkersburg, West Virginia, at the CHEST annual meeting hosted by the ACCP.

“Breath hold after you have achieved integrated imaging with your robot really doesn’t improve diagnostic yield,” he said. “What we really should focus on is it’s not static imaging, it’s the dynamic imaging that’s really important when you try to biopsy sub-centimeter nodules. And then optimizing the tool-lesion alignment is the second step once you reach that area that you should focus on.”

Dynamic imaging with integrated augmented fluoroscopy and C-arm-based tomography enables real-time localization of the tool in relation to the lesion as well as divergence recognition and tool correction, he said.

In the study, diagnostic yield was 95.9% with tool-in-lesion confirmation versus 85.0% with tool-touch-lesion. Sustained breath hold did not significantly improve tool-in-lesion positioning.

Rather than simply holding the lung still, clinicians need to reassess the tool-lesion relationship dynamically, Mathew emphasized. “Even if it’s moving, you still do a second biopsy … making sure the tool is actually aligned to the target.”

However, the study might not generalize to smaller lesions of 5 or 6 mm, given the small number of nodules of that size included in the study, he cautioned.

Mean lesion size was about 1.7 cm. Pneumothorax rate was about 1.7%.

The STABLE trial involved 117 patients with peripheral pulmonary lesions randomized to either conventional ventilation or breath hold during tissue biopsy. This comprised standardized end-inspiratory breath-hold ventilation during tissue acquisition to stabilize the target lung lesion. All cases used the Noah Galaxy robotic bronchoscopy with integrated augmented fluoroscopy and C-arm-based digital tomography imaging and tool-position correction.

Radiation exposure was comparable between study arms, averaging 5.8 minutes and about 38.6 mGy.

Mathew cautioned that the study represented a small exploratory analysis, and further study is needed.

Session moderator Hoi Yee Kwan, MBChB, chief of respiratory medicine at Kowloon Hospital in Hong Kong, who was not involved in the trial, questioned whether the breath hold was long enough or perhaps utilizing apnea ventilation might have achieved better yield.

Mathew responded that the biopsy took 5-7 minutes and how long the breath hold was within that timeframe wasn’t measured.

“Five to 7 minutes, I think, is long enough,” Kwan said.