Hepatic Artery Origin and Course at Scale

See also article by Choi et al in this issue.

Patrick D. Sutphin, MD, PhD, is an interventional radiologist at Massachusetts General Hospital in Boston, Massachusetts, and assistant professor at Harvard Medical School. His clinical and research interests include CT angiography, hepatic angiography, and liver-directed therapies for hepatocellular carcinoma and secondary liver tumors.

Sanjeeva Kalva, MD, is chief of the division of interventional radiology at Massachusetts General Hospital and professor of radiology at Harvard Medical School in Boston, Massachusetts. His research interests include interventional oncology, portal hypertension, inferior vena cava filters, and pulmonary arteriovenous malformations. He serves as an associate editor of Radiology: Cardiothoracic Imaging and international editor of Journal of Clinical Interventional Radiology.

In the classic textbook description of hepatic artery anatomy, the common hepatic artery (CHA) arises from the celiac axis. The CHA then branches into the gastroduodenal artery and proper hepatic artery. The proper hepatic artery then divides into the right and left hepatic arteries. This simplified description of the arterial supply to the liver fails to capture the anatomic diversity one may encounter in the angiography suite or in the operating room. A greater appreciation for variant anatomy was spurred by early surgical complications associated with variant anatomy in the setting of cholecystectomy and upper abdominal surgery. Deadly complications related to variant anatomy such as descriptions of ischemic liver necrosis secondary to severance of a replaced left hepatic artery during gastric resection inspired Nicholas Michels to address the great void in knowledge of hepatic artery anatomy. At the time, there was limited documentation of the anatomic variants and their associated prevalence. Michels devoted 20 years to the dissection of 200 cadavers to document hepatic artery anatomy and the associated rates of each variant (1,2).

Michels identified 10 distinct anatomic variants and provided a baseline probability for each of the anatomic variants. Only 110 of the 200 patients (55%) exhibited the classic anatomy, which was termed Michels type 1.

The Michels classification system separated the anatomic variants into replaced and accessory arteries; this classification was later modified by Hiatt et al in 1994 in the context of liver transplantation. Hiatt et al evaluated the hepatic artery anatomy in 1000 liver donors during surgical harvesting of livers for transplant. Based on this anatomic data, an update to the Michels classification was proposed in which the replaced and accessory arteries were considered as equivalent. Though both classification systems note the origins of the hepatic arteries, neither address the anatomic course of the arteries and the relationship to critical adjacent structures.

The study by Choi et al in the current issue of Radiology: Cardiothoracic Imaging addresses hepatic arterial anatomy through a large-scale analysis of imaging (3). The authors examine not only the origins of the hepatic arteries but also characterize the three-dimensional course of the hepatic arteries with respect to the portal vein and biliary system. The sheer size of the data set is an imaging tour de force with analysis of digitally subtracted images from transarterial chemoembolization (TACE) procedures combined with CT studies in each of the 5625 patients. Two radiologists retrospectively reviewed and documented the origins of the right and left hepatic arteries as well as painstakingly examined CT images to document the course of the hepatic arteries relative to adjacent anatomic structures, including the portal vein and bile duct for the right hepatic artery and in the fissure for the ligamentum venosum for the left hepatic artery.

In this study an aberrant hepatic artery is defined as originating from an artery other than the CHA or as arising directly from a subdivided region of the CHA or gastroduodenal artery. The authors identified the classic, type I, anatomy in 72.6% of patients, similar to other recent studies, but notably greater than the original study by Michels.

The large size of the data set enabled the authors to examine the relationship between artery origin and course not possible in smaller studies. A significant association was found between an aberrant right hepatic artery and an aberrant left hepatic artery. The authors found a retroportal course of the right hepatic artery when it arose from the superior mesenteric artery, aorta, celiac axis, and proximal to middle CHA in all cases.

Recognition of variant hepatic artery anatomy is critically important for transarterial tumor therapies such as hepatic artery infusion including with infusion pumps, TACE, and radioembolization. Maximizing the therapeutic benefit depends on identifying each of the branch arteries that supply the liver and tumor tissue. Failure to recognize variant anatomy in transarterial therapies puts the patient at risk for inadequate tumor treatment and at a higher risk for disease progression. Prior to hepatic artery infusion pump placement, patients undergo thorough evaluation of hepatic arterial supply with CT angiography to determine the number of vessels to cannulate or the need for ligation of anatomic variants. A study by Allen et al found a higher rate of complication associated with multiple arterial cannulations, as well as cannulation of vessels other than the gastroduodenal artery, factors most likely related to variant hepatic artery anatomy (4).

Hepatic artery variant anatomy also has significant implications for hepatobiliary surgery. In the case of hepatobiliary surgery, not only is the origin of the artery important, but also the three-dimensional course of the hepatic arteries. In the case of pancreaticoduodenectomy, an aberrant right hepatic artery may have an intraparenchymal pancreatic course and may require en bloc resection with resection of the pancreatic head mass (5). The loss of arterial supply from a compromised aberrant right hepatic artery may then result in anastomotic breakdown secondary to ischemia, or possibly stricture of the hepaticojejunostomy (6).

Liver transplantation also requires the recognition of aberrant hepatic artery anatomy. At the time of graft harvest, the aberrant hepatic arteries must be carefully identified with precision to avoid injury that may prevent complete graft arterialization (7). Loss of hepatic arterial supply to the transplanted liver may lead to biliary necrosis or liver parenchyma complications that may ultimately result in graft failure (8).

In summary, an in-depth understanding of variant anatomy is critical for radiologists, interventional radiologists, and surgeons to ensure adequate tumor treatment via transarterial therapies and avoid catastrophic complications related to hepatic arterial injury during surgery. The large-scale analysis of hepatic artery variants reported in this issue by Choi et al provides a wealth of information on the prevalence of variants and the associations between anatomic variants and hepatic artery anatomic course (3).