Evaluation of the hepatic artery in potential donors for living donor liver transplantation by computed tomography angiography using multidetector-row computed tomography: comparison of volume rendering and maximum intensity projection techniques.

PURPOSE

To compare the volume rendering (VR) and maximum intensity projection (MIP) computed tomography angiography (CTA) techniques using multidetector-row CT for hepatic artery evaluation of potential donors for living donor liver transplantation (LDLT).

METHOD

Over a 9-month period, CTA using multidetector-row CT and conventional angiography was performed in 62 consecutive potential donors for LDLT. Acquisition of arterial phase scans was initiated within 5 seconds after reaching enhancement of the descending aorta up to 70 HU as measured by a bolus-tracking technique (collimation = 1.25 mm, table speed = 7.5 mm, reconstruction interval = 0.625 mm). Postprocessing was performed on a commercially available workstation. Computed tomography angiography images of the hepatic artery were made using the VR and MIP techniques. A total of 10 to 30 images of various planes were generated to reveal the origins and branching patterns of hepatic arteries. With a 2-week interval, two reviewers separately evaluated CTA using VR and MIP techniques, specifically evaluating anatomic variations and hepatic arterial conspicuity. The results of conventional angiography were considered to represent the gold standard. The difference in postprocessing time between the two techniques was statistically analyzed by the Student t test, and the differences in arterial conspicuity and in the identified number of the hepatic artery to segment IV of the liver were analyzed by the chi2 test.

RESULTS

Two CT examinations (3%) were technically inadequate for generating CTA because of respiratory motion artifact. The average times for postprocessing of VR and MIP images were 8.1 +/- 3.8 minutes and 5.0 +/- 0.7 minutes, respectively. Variations of hepatic arterial anatomy were present in 19 patients (32%) on conventional angiography. In 8 patients (13%), there was discrepancy in the variations of hepatic arterial anatomy between the two techniques: MIP was correct in 7 patients, and VR was correct in 1 patient. The dominant artery supplying segment IV was identified in 51 patients (85%) with MIP and in 39 patients (65%) with VR. There was no significant difference in conspicuity of the hepatic arteries using either the MIP or VR technique.

CONCLUSION

In CTA techniques using multidetector-row CT, MIP is superior to VR for the hepatic arterial evaluation of potential LDLT donors in terms of depicting anatomic variations and postprocessing time.