Factors affecting Outcomes of Thoracic Duct Embolization in Management of Iatrogenic Chylothorax after Thoracic Oncologic Surgery

To assess the factors affecting the outcomes of thoracic duct embolization (TDE) in management of iatrogenic chylothorax after thoracic oncologic surgery.

Materials and Methods:

In this IRB approved single institution retrospective analysis, the medical records of 36 patients who underwent TDE for management of iatrogenic chylothorax after thoracic oncologic surgery between March 2018 and March 2024 were reviewed. Patient demographics, oncologic diagnosis (cancer type, stage) and prior oncologic treatments (systemic treatments, thoracic radiation), surgical procedural details (number of lymph nodes resected, benign vs. malignant lymph nodes, body mass index at time of surgery) and TDE procedural details (time of lipiodol transit) were collected. Chest tube output prior to and following TDE, the fluid triglyceride level prior to and following TDE, the number of days between TDE and chest tube removal, and between TDE and hospital discharge were assessed. TDE procedure images were reviewed for assessment of the site of lymphatic leak (main thoracic duct vs. tributary) and post TDE cross-sectional images were reviewed to assess the diameter of the thoracic duct. Linear regression was used to assess factors affecting outcomes.

Results:

The median time from thoracic duct embolization to chest tube removal was 5 days (IQR: 3-11). A significant change was noted in the volume of output between pre-intervention (median of 1305, 687.5-1559 ml/ day), versus the output 2 days post-TDE (325, IQR: 360-1075 ml/day), p=0.0003. The time to chest tube removal post intervention was noted to be shorter for lymphatic leaks occurring at a branch level (median of 4, IQR:3-5 days) compared to leaks originating from the main thoracic duct (median of 6, IQR:5-12 days), p=0.024. A shorter time of lipiodol transit was significantly correlated with higher chest tube output (coefficient: -7.33, 95% CI: -11.99, -2.63, p=0.003) but did not correlate with a longer time to chest tube removal. No other factors assessed were statistically significant.

Conclusion: Thoracic duct embolization was associated with a significant decrease in chest tube output and time to chest tube removal was shorter for branch leaks vs. main duct leaks. Shorter lipiodol transit time was correlated with higher chest tube output suggesting that flow rate within the lymphatic system correlates with chest tube output.