Room: Room 207
Purpose: To assess the hemodynamic effects of hemodialysis (HD) on the liver using CT perfusion (CTP) imaging and to explore dialysate cooling as a protective intervention against intradialytic circulatory stress.
Methods: 15 HD patients were randomized to receive either standard (36.5Â°C) or cooled (35Â°C) HD first in a 2-visit crossover study design. For each visit, CTP imaging was performed at three timepoints (before, 3 hours into, and after HD) on a 256-slice CT scanner (GE Healthcare) without interruption to HD. Each scan was done without breath-hold for 2 minutes immediately following a bolus injection of iodinated contrast agent. Misalignment among CTP images was minimized using non-rigid registration software, and parametric hepatic perfusion maps for total liver (TL), hepatic arterial (HA), and portal venous (PV) blood flow were generated from the registered CT images.
Results: The average TL, HA and PV perfusion for the standard HD case increased to 7%, 1% and 11% above baseline, respectively, by 3 hours into HD and remained almost unchanged (fluctuations <2%) after HD. For the cooled HD case, the average TL perfusion remained unchanged by 3 hours into HD, while the HA and PV perfusion changed to 4% below and above baseline, respectively. By the end of HD, the TL perfusion changed to 2% above baseline, while the HA and PV perfusion continued their trends and changed to 6% below and 7% above baseline, respectively.
Conclusion: There was a difference between the standard and cooled HD perfusion results, where by 3 hours into HD, the cooled HD values were 6-7% less than the standard HD values. Additionally, the magnitude of perfusion changes from baseline was smaller for cooled HD compared to standard HD. These findings suggest that the cooled dialysate intervention successfully reduced the hemodynamic impact of HD on the liver.
Funding Support, Disclosures, and Conflict of Interest: Dr. Ting-Yim Lee has a licensing agreement with GE Healthcare for the CT Perfusion software used in this work.