Although most liver tumors are fed by hepatic vessels, a small but significant portion have extrahepatic supply. In this Q&A, William Rilling, MD, FSIR, of the Medical College of Wisconsin in Milwaukee, discusses pre-procedure imaging, risks/benefits of treatment, and communicating about treatment possibilities with patients.
What is important to keep in mind regarding treating extrahepatic tumor supply?
Physicians who are treating patients with liver tumors need to know that extrahepatic tumor supply is something that occurs on a fairly regular basis, so it is important to have a high index of suspicion. My rough estimate is that extrahepatic supply occurs in 8% to 12% of patients.
There are certain vessels that are the major culprits. The inferior phrenic artery is by far the most common in all literature on this phenomenon. If tumors have extra hepatic supply and it is not diagnosed and treated, patient outcomes may be adversely affected.
What should physicians remember about risk/benefit ratios of treatment?
The risk of treating these extrahepatic vessels is different than treating the hepatic vessels. The extent and the implications of these risks should be discussed with the patient ahead of time. For example, when we are treating the inferior phrenic artery, the patient needs to know that there is risk of an irritation to the diaphragm, or that they could get referred pain to the shoulder from diaphragm irritation. They could have fluid around the lung after treatment, and they could have non-target embolization to the pulmonary arteries. All these risks are well-described in the literature and known to medical professionals, but patients will be surprised and alarmed post treatment if they have severe shoulder pain and that possibility was not discussed.
What are some of the risks involved with treating different vessels?
Damage to the kidney or the adrenal gland can occur when treating the vessels off the adrenal or the renal artery. When treating from the internal mammary artery into the superior epigastric artery, there is risk mostly of injury to the skin. The most feared complication is a spinal cord injury, which can occur rarely when treating intercostal or lumbar arteries that feed some of these tumors.
Do you have any tips for approaching the consent process with patients and explaining the risks and benefits?
The most important point is to tell a patient that it is possible that the treatment may include some vessels outside the liver to try to achieve complete treatment of the tumor. The goal is to completely cover all of the target tumor. In order to do that, we have to sometimes treat these extrahepatic vessels, which involves different risks.
Good pre-procedure imaging can help with guiding the discussion because it is difficult to talk to a patient about possibilities involving a big list of 10 vessels that might need to be treated. The pre-procedure imaging and location of the tumor can help physicians predict with reasonable accuracy which extrahepatic vessels would come into play. Tumors in certain locations, such as a posterior tumor that is high up in liver, would often derive some flow from the phrenic artery, whereas an anterior tumor next to the heart is in a classic location to parasitize from the internal mammary artery. An exophytic tumor on the bottom of the liver is more likely to pull some extra flow from the renal artery, or the adrenal artery.
What should be considered when selecting patients?
All factors should be considered within the context of the patient’s overall performance status and prognosis. Careful consideration is important since extrahepatic vessels involve a different risk/benefit profile. We might not typically perform these procedures in someone with a very guarded prognosis, but if we are trying to obtain a complete response from treatment, then it may be necessary to proceed.
Sometimes there is a game-time decision as to the risk/benefit. Sometimes we can select branches that go more exclusively to tumor and spare more normal tissue, or we can use coil protection methods for some branches going to normal tissues to force treatment to be more on target. Those approaches are not usually revealed until mapping of the vessels. Plan as much as possible in advance, but realize that some decisions will need to be made in the middle of the procedure.
Do you have any other technical tips or comments?
I think we're fortunate now to have cone-beam CT as a tool to assess in real time how much targeted versus non-targeted embolization there might be from treating a particular branch. I also think that choosing the type of therapy is important. For example, conventional (lipiodol) chemoembolization allows real-time visualization of the actual therapeutic agent. We can see in real-time under fluoroscopy whether we are going to target or into non-target branches, and we can make adjustments in the moment.
What are some imaging tips for the pre-procedure?
We need to have good arterial phase imaging to help visualize some of the extrahepatic branches and follow them to see if they are contributing to tumor perfusion. Having really good arterial phase imaging is key to being able to plan.
Are there any future techniques or developments in this field that you are looking forward to within the next three to five years?
There are a lot of new drug-eluting platforms that are in development. There is certainly potential that some of these can be useful in situations with extrahepatic perfusion, but there are no data at this point in time.
Is extrahepatic tumor supply seen more frequently in tumors of a certain size?
It’s important to note that this phenomenon is not just seen in large tumors. I know from experience that a patient can have a 2.5-cm tumor fed entirely by extrahepatic vessels. In either large or small tumors, the clinician needs to problem solve and figure out ways to treat the patient. If you do not see perfusion from hepatic vessels, then the tumor must be perfused from somewhere else. Additionally, the vast majority of the literature discusses extrahepatic tumor vessels in the context of hepatocellular cancer, but this can happen in metastatic tumors as well.