Pancreatic Venous Anatomy for Trans-portal treatment of pancreatic cancers using pressure enabled drug delivery (PEDD)

Wednesday, March 8, 2023

4:21 PM – 4:30 PM

Location: Phoenix Convention Center, 222AB

Abstract Speaker(s)

Michael G. Kwong, BA

Medical Student
University of Colorado Anschutz Medical Campus

Disclosure(s): AKICept: Consultant (); Becton Dickenson: Advisory Committee/Board Member (); Trisalus: Stock Shareholder (excluding mutual funds) ()

Author/Co-author(s)

HS

Hannah R. Spears, MPH

Statistician
Colorado School of Public Health, University of Colorado-Denver Anschutz Medical Campus
JP

Jack Pattee, PhD

Statistician
University of Colorado Anschutz Medical Campus
EW

Erika Worley, BS

Research Assistant
University of Colorado Anschutz Medical Campus
GM

Gordon McLennan, MD FSIR

Vice Chairman for Clinical Research
University of Colorado School of Medicine

Disclosure(s):

Michael G. Kwong, BA: AKICept: Consultant (); Becton Dickenson: Advisory Committee/Board Member (); Trisalus: Stock Shareholder (excluding mutual funds) ()

Gordon McLennan, MD FSIR: AKICept: Advisory Committee or Review Panel Member (Ongoing); Becton Dickenson: Advisory Committee or Review Panel Member (Ongoing); Trisalus: Ownership Interest (Ongoing)

Purpose: To define variability of pancreatic venous drainage in preparation for Pancreatic Retrograde Venous Infusion (PRVI^TM) trials with Pressure Enabled Drug Delivery (PEDD^TM) for locally advanced pancreatic ductal adenocarcinoma.

Materials and Methods: 117 triple-phase liver CT scans from November 2020 to October 2021 were reviewed. One mm axial images, coronal & sagittal reconstructions were reviewed from portal, arterial, & venous phases. The presence of pancreatic cancer was noted. Diameter, visible length, angle of insertion into draining vein, tortuosity, and presence of intra-parenchymal collateralization were recorded for each pancreatic vein seen. To identify veins that might be appropriate targets for PRVI via catheters with PEDD, veins greater than 10 or 20 mm in length were noted. The diameter of veins at the origin and 20 mm from the origin were measured if applicable.

Results:

350 veins were seen in 117 CT scans. The mean number of pancreatic veins visible per patient was 2.99 with a standard deviation of 1.00. 285 veins were best seen in the portal phase, 14 in the arterial phase, 41 in the venous phase, & 10 with a combination of arterial & portal phase. 172 veins drained the pancreatic head, 69 the body & 109 the tail. The pancreatic head drained into the portal vein (70) or SMV (90). The tail drained primarily into the splenic vein (95) while the body had more variable drainage with 15 veins draining into the portal, 12 into the SMV & 38 into the splenic vein. 10 of 22 patients with pancreatic tumors had veins draining tumors.

293 veins were between 2 & 6 mm in diameter at the origin (83.7%) Of those, 208 were at least 10mm long (59.4%). 118, 32, & 58 drained the head, body & tail respectively. 119 veins were at least 20 mm long (34%). Mean number of pancreatic veins longer than 20 mm visible per patient was 1.42 with a standard deviation of 0.56. 92 veins were seen in the portal, 6 in the arterial, & 18 in the venous phase. 85 veins drained the pancreatic head, 16 the body & 18 the tail. Mean lengths were 34.47 mm, 23.44 mm & 29.52 mm from the head, body & tail.

Conclusion: Based on CT findings, vein diameter and angle are consistent with the ability to canulate the veins from portal access. 83.7% of veins had adequate diameters & 59.4% were of at least 10mm length.