The use of MDCT-based computer-aided pathway finding for mediastinal and perihilar lymph node biopsy: a randomized controlled prospective trial.

BACKGROUND

Mediastinal and perihilar lymph node samples can be acquired safely through the transbronchial approach during a bronchoscopic examination that is usually required as part of the evaluation of suspected lung cancer. Typically, needle aspiration samples are performed and needle cores can be sampled if the operator is confident that the needle is within the lymph node target, partly because of the risk of bleeding if a large blood vessel is sampled during core biopsy, especially in the perihilar region. Many bronchoscopists have difficulty assessing the three-dimensional (3D) positioning for needle sampling during these procedures, especially when relying on multidetector-row computerized tomography (MDCT) images displayed two-dimensionally seen prior to and usually during the procedure.

OBJECTIVE

We have developed and evaluated a process model and associated software for providing interactive 3D displays of the MDCT data for procedure planning and real-time virtual bronchoscopic pathfinding for these procedures.

METHODS

We undertook a prospective randomized clinical study for evaluating the computer-aided pathfinding assistance in mediastinal lymph node biopsies in 87 consenting subjects.

RESULTS

We demonstrate that the addition of this computer-aided pathfinding improved operator performance in perihilar and paratracheal lymph node sampling (100 vs. 69%) but not in subcarinal sampling (82 vs. 85%). Overall success with lymph node sampling is 92% using the computer-aided method and 77% using standard clinical practice.

CONCLUSIONS

The type of computer-aided pathway assistance described here, using 3D MDCT scanning information obtained before the procedure, but interacting with real-time bronchoscopic images during the bronchoscopic procedure, should improve the confidence of most bronchoscopists in performing these procedures, with improved clinical outcomes, and will add to the personalization of medicine through imaging.