Optimized Computed Tomographic Angiography Vessel Evaluation Protocol (OCTAVE) Prior to Transarterial Radioembolization

Abstract: Purpose: To evaluate optimized edge-enhanced 1 mm to 1.5 mm thin-slice arterial-phase computed tomography (CT) vs standard 3.0 mm to 5.0 mm slice arterial-phase CT in identification of clinically relevant mesenteric vessels prior to mesenteric mapping. Materials and Methods: An IRB-approved retrospective review of 50 consecutive patients (male-to-female ratio 37:13, mean age 62.6, range 32-80) undergoing yttrium-90 radioembolization was conducted. A McNemar test was applied for statistical analysis comparing standard 3.0 mm to 5.0 mm (series A) and edge-enhanced filtered 1.0 mm to 1.5 mm sections (OCTAVE protocol) (series B) sections in identification and measurement of the common hepatic artery (CHA), right/middle/left hepatic (RHA, MHA, LHA), gastroduodenal (GDA), cystic, left/right gastric (LGA, RGA), falciform, phrenic and supraduodenal arteries. Results: The mean diameter of the MHA, cystic artery, and RGA ranged from 1.4 mm to 2.4 mm. The MHA was identified in 70% vs 96%, cystic artery 44% vs 86%, RGA 14% vs 79%, and other sub 3 mm arteries such as the falciform, phrenic, and supraduodenal arteries 26% vs 45% of the time in series A vs series B. Identification of the MHA, cystic artery, RGA, and other arteries of concern was statistically significant (P<.05). The mean diameter of the CHA, RHA, and LHA ranged from 3.6 mm to 6.2 mm and was identified in 100% of both series. The LGA and GDA ranged from 2.9 mm to 3.9 mm and were identified in 97% in series A and 100% in series B. No statistically significant difference was noted in the detection of vessels >3 mm. Conclusion: OCTAVE (series B) is superior in the detection of clinically significant vessels. The detection of these vessels prior to catheterization may lead to reductions in contrast use, radiation, procedural time, and complications associated with radioembolization.

Key words: radioembolization, mesenteric mapping, computed tomography,
interventional oncology, yttrium-90


Complex embryologic origins of the biliary and hepatic systems allow for a high degree of local variation and adaptation in arterial anatomy.1 61% of patients have conventional hepatic arterial anatomy, with the main right and left hepatic branches arising from the proper hepatic artery.2 Variant “replaced” origins of the left and right hepatic artery (LHA and RHA), sometimes with accessory arterial branches, and other anatomic variants including a middle hepatic artery (MHA), double replaced hepatic artery, and replaced common hepatic artery (CHA), are routinely noted.3 In addition, serosal liver tumors have a 17% likelihood of having parasitized supply from the right phrenic, right adrenal, right internal mammary, omental, renal, adrenal, left gastric, or cystic arteries, depending on the location of the tumor.4

With newer generations of hepatic embolotherapy, such as yttrium-90 (Y90) radioembolization, complications may be more severe (as compared to chemoembolization) and primarily associated with nontargeted delivery of microspheres to the gut, skin/abdominal wall, gallbladder, pancreas, and diaphragm.5-8 The need to optimize variant anatomy and parasitization has led to meticulous pretreatment angiography assessment and aggressive skeletonization of the hepatic artery. Accurate assessment of the hepatic arterial supply prior to the delivery of radioembolic material is key to reducing complications and optimizing patient outcome. 

The aim of the present study is to evaluate an Optimized Computed Tomographic Angiography Vessel Evaluation (OCTAVE) protocol for preprocedural detection of significant mesenteric vessels in a noninvasive manner prior to catheter-based angiography.


This retrospective study was approved by the institutional review board of our hospital, and informed consent was waived. 

Study Population

A retrospective review of 69 patients with consultation for radioembolization from June 2011 to November 2014 was conducted. Inclusion criteria were a mandatory pretreatment contrast-enhanced arterial-phase computed tomography (CT) with reconstructions of the dataset at 3.0 mm to 5.0 mm and 1.0 mm to 1.5 mm axial/coronal sections on a 64-slice Siemens Somatom Sensation scanner with Optiray 320 contrast at 2cc/kg (to a maximum of 150 cc of contrast). Patients who did not undergo this CT protocol were excluded from the study. Any patients who did not consent to radioembolization after CT, and thus did not proceed to mesenteric mapping, were also excluded. For each patient, any vessels that were previously embolized or resected (as a result of partial hepatectomy or previous Y90 treatment) were also excluded from analysis. Select patients underwent OCTAVE protocol at the discretion of the protocoling diagnostic radiologist. Of the 69 patients who underwent OCTAVE protocol, 50 patients (male-to-female ratio 37:13, mean age 62.6, minimum age 32, maximum age 80) satisfied the inclusion criteria.

Each patient’s CT study was analyzed by two independent board-certified radiologists, initially reviewing the standard filter 3.0-5.0 mm slice images followed by the edge enhanced 1.0-1.5 mm slice image to identify the presence of the stated nontarget vessels using a checklist form. Reconstructions were obtained in the coronal and axial plane. For the purposes of the study, no postprocessing reconstruction or volume rendering (e.g., maximum intensity projections [MIPs], 3D workstation, or additional planes) were used, in order to best represent standard workflow. Three-dimensional and MIP reconstructions were utilized for the illustrative figures within the manuscript but were not utilized for primary identification or diagnosis by the interpreters. The location of each vessel was then compared to the catheter-based angiograms, which were considered the gold standard. Any discrepancies with respect to the presence of the vessel on CT were resolved by consultation with a third board certified radiologist. The average resulting diameter for each vessel from all readers is presented in table format. 

Statistical Analysis

A McNemar test was applied for statistical analysis comparing 3.0 mm to 5.0 mm axial sections (series A) and OCTAVE protocol utilizing 1.0 mm to 1.5 mm axial sections (series B) for the identification and measurement of the CHA, RHA, MHA, LHA, gastroduodenal (GDA), cystic, left gastric artery (LGA), and right gastric artery (RGA) in addition to the falciform, phrenic, and supraduodenal arteries. The McNemar test is based on the number of discordant cases (where the vessel was identified on the 1.0 mm to 1.5 mm series but not on the 3.0 mm to 5.0 mm series) and attempts to reject the null hypothesis that the proportions of positive results for the two CT datasets are due to chance vs the alternative hypothesis that they are in fact more accurate. Statistical significance was defined as P<.05.


All scanned series were devoid of motion artifact. It was observed that in highly tortuous vessels, simultaneous review of both axial and coronal planes allowed for proper tracking and visualization of the vessels to their origins (e.g., RGA, cystic artery). Figures 1-6 demonstrate images of the right gastric (Figure 1), cystic (Figure 2), phrenic (Figure 3), supraduodenal (Figure 4), middle hepatic (Figure 4A) (Figure 5), and falciform (Figure 6) arteries outlining the close correlation between the thin-cut edge-enhanced CT study vs the angiographic findings. 

Common vascular anatomy such as the CHA, RHA, LHA, GDA, and LGA were identified in both series A and series B, without statistical significance, with mean vessel diameter ranging from 2.6 mm to 6.2 mm (absolute range 1.0 mm to 8.0 mm). 

When mean vessel diameter fell below 2.5 mm, series B demonstrated a significant improvement in detection across this range, representing the MHA, cystic, RGA, and other smaller significant vessels. The range of mean vessel diameters was 1.8 mm to 2.3 mm (absolute range 1.0 mm to 3 mm). The use of OCTAVE protocol (series B) allowed for overall better visualization of all vessels, and in particular, visualization of the RGA in 79% of cases, as compared to 14%, cystic artery noted in 86% vs 44% and visualization of the MHA was possible in 96% vs 70%. Table 1 summarizes the results of the analysis.

Discussion and Conclusion

Gastrointestinal ulceration rates after radioembolization average 8% in larger multicenter cohorts.9-11 With increased awareness, reported rates have largely decreased to <5% but remain a concern.12-14 With advances in CT technology, early identification of clinically significant vessels can be performed on a noninvasive basis, allowing for improved preprocedural planning. This study demonstrates the superiority of the OCTAVE protocol in the detection of significant but small vessels, including the RGA, MHA, and cystic artery. While detection of vessels using CT does not guarantee technical success in selection and embolization, knowledge of these communications even without successful embolization may influence treatment planning and may prompt more distal catheter positioning for microsphere administration. Furthermore, proper identification could allow for improved efficiency in mapping, preprocedural planning for skeletonization for hepatic redistribution, overall reduction in procedural radiation dose, contrast utilization, and procedural time by facilitating the rapid identification and selection of the best viewing angle to reach the target vessel. Preprocedure identification of clinically relevant vessels further complements and reduces cone-beam CT acquisitions during mesenteric angiography by optimizing safety and patient outcome.15 

Limitations of this study include its retrospective nature, small sample size, and two reader interpretations for each case.

A potential drawback of obtaining edge-enhanced 1.0 mm thin cuts in CT is the tradeoff in radiation dose. Patients with hepatic primary or metastatic malignancies are often followed serially with multiphase liver imaging. Increased dose (higher milliampere-seconds [mAs]) is required for delineation during arterial acquisition for thinner slices to prevent noisy images as a result of proton starvation. In general, an increase of approximately 100 mAs is required during arterial phase acquisition. This exposure translates into approximately an additional 3 millisieverts per study; however, it may only be required during the planning phases and thus may be a one-time radiation penalty. This may be slightly more compounded as a CTA would need to be performed for patients with metastatic disease prior to mapping. In general, patients with metastatic disease (e.g., colorectal cancer metastasis) generally only undergo single-phase portal venous phase imaging for staging. In our experience, radiation dose relating to the enhanced CT acquisition can potentially result in significantly lower procedural radiation doses, fluoroscopy times, and room time due to the ability to strategize mapping and administration in an a priori fashion.

In summary, the OCTAVE protocol dramatically improves the ability to detect mesenteric vessels that are relevant to radioembolization in a noninvasive fashion prior to catheterization. The described simple technique has the potential to dramatically increase procedural efficiency and contribute to greater safety when performing radioembolization.

Editor’s note: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr. Liu reports consultancy to and grants from BTG and Sirtex Medical.

Manuscript received August 1, 2016; manuscript accepted September 22, 2016. 

Suggested citation: Ng EH, Chung JB, Klass D, Ho SG, Patel R, Chou FY, Legiehn GM, Liu DM. Optimized computed tomographic angiography vessel evaluation protocol (OCTAVE) prior to trans­arterial radioembolization. Intervent Oncol 360. 2016;4(11):E183-E193.


  1. Liu DM, Salem R, Bui JT, et al. Angiographic considerations in patients undergoing liver-directed therapy. J Vasc Interv Radiol. 2005;16(7):911-935. 
  2. Lee AJ, Gomes AS, Liu DM, Kee ST, Loh CT, McWilliams JP. The road less traveled: importance of the lesser branches of the celiac axis in liver embolotherapy. Radiographics. 2012;32(4):1121-1132. 
  3. Song SY, Chung JW, Lim HG, Park JH. Nonhepatic arteries originating from the hepatic arteries: angiographic analysis in 250 patients. J Vasc Interv Radiol. 2006;17(3):461-469.
  4. Lee EW, Thakor AS, Tafti BA, Liu DM. Y90 selective internal radiation therapy. Surg Oncol Clin N Am. 2015;24(1):167-185.  
  5. Schelhorn J, Ertle J, Schlaak JF, et al. Selective internal radiation therapy of hepatic tumors: Procedural implications of a patent hepatic falciform artery. Springerplus. 2014;3:595. 
  6. Burgmans MC, Too CW, Kao YH, et al. Computed tomography hepatic arteriography has a hepatic falciform artery detection rate that is much higher than that of digital subtraction angiography and 99mTc-MAA SPECT/CT: Implications for planning 90Y radioembolization? Eur J Radiol. 2012;81(12):3979-3984.
  7. Abdelmaksoud MH, Hwang GL, Louie JD, et al. Development of new hepaticoenteric collateral pathways after hepatic arterial skeletonization in preparation for yttrium-90 radioembolization. J Vasc Interv Radiol. 2010;21(9):1385-1395.
  8. Prince JF, van den Hoven AF, van den Bosch MA, Elschot M, de Jong HW, Lam MG. Radiation-induced cholecystitis after hepatic radioembolization: Do we need to take precautionary measures? J Vasc Interv Radiol. 2014;25(11):1717-1723.  
  9. Murthy R, Brown DB, Salem R, et al. Gastrointestinal complications associated with hepatic arterial yttrium-90 microsphere therapy. J Vasc Interv Radiol. 2007;18(4):553-561.
  10. Carretero C, Munoz-Navas M, Betes M, et al. Gastroduodenal injury after radioembolization of hepatic tumors. Am J Gastroenterol. 2007;102(6):1216-1220.
  11. Neff R, Abdel-Misih R, Khatri J, et al. The toxicity of liver directed yttrium-90 microspheres in primary and metastatic liver tumors. Cancer Invest. 2008;26(2):173-177. 
  12. Naymagon S, Warner RR, Patel K, et al. Gastroduodenal ulceration associated with radioembolization for the treatment of hepatic tumors: an institutional experience and review of the literature. Dig Dis Sci. 2010;55(9):2450-2458. 
  13. Riaz A, Lewandowski RJ, Kulik LM, et al. Complications following radioembolization with yttrium-90 microspheres: a comprehensive literature review. J Vasc Interv Radiol. 2009;20(9):1121-1130.
  14. Konda A, Savin MA, Cappell MS, Duffy MC. Radiation microsphere-induced GI ulcers after selective internal radiation therapy for hepatic tumors: an under recognized clinical entity. Gastrointest Endosc. 2009;70(3):561-567. 
  15. Tacher V, Radaelli A, Lin M, Geschwind JF. How I do it: Cone-beam CT during transarterial chemoembolization for liver cancer. Radiology. 2015;274(2):320-334.