Arteriovenous Malformation

AVM is referred to as a ‘Fast-flow’ vascular malformation, because multiple direct connections between artery and vein lead to a greatly increased localized blood flow.  Normally, arteries are at a much higher pressure than veins, and this pressure difference is maintained by the resistance to flow as blood flows through the capillary network connecting the arterial to the venous network.  Within an AVM these connections are larger, and so provide little resistance to flow (the same as fluid flows more easily through a hose than a straw).  As a consequence, much more blood than normal flows through the AVM making it warm and pink.  Also the pressure in the venous systemnearby rises, so that the veins which carry blood away from the AVM are enlarged and feel tense. 

The exact prevalence of AVM is not known, but they are thought to affect somewhere around 1 in 10,000 people

AVMs can occur in any part of the body, although for reasons that are not understood they are more common in some parts (such as the lips, nose, ears, hands and feet) than others.  AVMs which occur in the brain have traditionally been managed by Neurosurgeons because of the special problems in this location, in particular that even a small bleed can have major consequences.  It is not known whether AVMs in the brain are inherently different from AVMs elsewhere, but with increasing knowledge it seems likely that these are the same thing but in different locations.  Because of the unique consequences of brain AVM, they are not managed in vascular anomalies teams, but all other AVMs are. 

Most AVMs are ‘sporadic’, which means that they have occurred spontaneously in someone who does not have any underlying inheritedtendency and so is not likely to pass the risk on to their children.  

A proportion of AVMs, however, arise as part of a ‘syndrome’ (it is not clear yet exactly what proportion, but one in ten would not be too far off), in which there is an inherited tendency towards AVM formation.  The two best characterized syndromes are ‘CM-AVM’ and HHT.  

In CM-AVM syndrome the most characteristic feature is red patches on the skin, which may be found on other family members.  Not all people who have CM-AVM have AVMs, but the red patches are the best indication that there may be an inherited syndrome.  Not all red patches indicate CM-AVM, so if there is any question in someone with an AVM, then it is best if the family member(s) with the patches are seen by an AVM expert. Variants in two separate genes have been associated with CM-AVM (RASA1 and EPHB4) but some verified cases do not have variants identified in either gene and there may be more genes to be discovered.  AVMs in the brain can occur in CM-AVM syndrome, so if it is diagnosed, then screening tests (MRI brain and spine) are usually performed on affected family members (including those without a known AVM).  This, and genetic testing, should be discussed with your treating doctor.

HHT (hereditary haemorrhagic telangiectasia) otherwise known as ‘Rendu Ossler Weber’ disease is characterized by typical dilated vessels in the skin, which are often seen around the lips.  There may be a family history of severe nose bleeds, which can be difficult to manage.  AVMsin HHT can occur in the lungs, something not otherwise generally seen.  Once again genetic testing for one of the three types of HHT should be considered and management in a specialized centre is advised.

Children and adults with sporadic AVMs should be seen by a specialist with experience with AVMs, ideally someone in a multidisciplinary team which contains a Dermatologist, but genetic testing for a syndrome and screening for brain AVM is not generally indicated unless there are some identifiable clinical features.

In recent years there has been considerable progress in the understanding of the causes of AVM.  Like other vascular malformations, AVMs appear to be due to what is known as ‘somatic mutations’, that is mutations in the genetic code that are not inherited but are found in cells within the AVM.  Variants in a number of different genes have been found in AVM, and those genes are mostly in a signalling pathway within the cell called the Ras/MEK/ERK pathway.  In those AVMs which occur as part of an inherited condition, it is thought likely that the actual AVM is due to a ‘second hit’ somatic mutation in the same gene as the one where the inherited variant has occurred or in another related gene, which explains why not all people with CM-AVM syndrome have AVMs.

AVMs do enlarge over time, but generally this happens very slowly.  The exception is during pregnancy when, for reasons not well understood, but presumably due to circulating hormones, they can enlarge quite quickly.  AVMs can also enlarge at puberty.   AVMs that have enlarged during pregnancy sometimes go back to their original size after delivery, but not always. 

Apart from pregnancy, enlargement of AVM is usually very gradual and may be so slow as to be imperceptible. This means that there is generally no urgency to treat an AVM, but it can also be a trap because through slow, steady growth the AVM can eventually reach a size which is difficult to treat without there having been a moment when a decision to treat could be made. 

It used to be said that AVMs are ‘errors of development’ which led doctors to believe that all AVMs truly arose in the embryo, but only became evident later in life.  The implication of this was that it was thought that enlargement of an AVM was really just a manifestation of something which had been there all along, that ‘growth’ of the AVM was not really growth, but a process of something which had been there all along becoming more evident.  If this was the case then there would be no advantage to treating the AVM early, and it might be better to treat it later.  The finding that AVMs are caused by somatic mutations has changed this view, and most specialists have now come to the view that AVMs do truly grow, albeit slowly, and are better treated earlier rather than later.  

Many AVMs do not cause any complications.  The most important complication when one occurs is bleeding.  The blood vessels in and around an AVM are fragile and bleed easily, and because of the high pressure in the veins bleeding can be rapid and very alarming.  

Like any other bleeding, the best way to control bleeding from an AVM is to apply sustained pressure.  This can be difficult to do inside the mouth but is the best way to control bleeding. Pushing a wad of gauze or similar material onto the bleeding point is the best way, and keeping the pressure on for at least ten minutes or longer (without lifting every few minutes to see if it has stopped) or until emergency care can be found is best.  Direct pressure if it can be sustained is nearly always preferable to a proximal tourniquet.  Blood clots normally in an AVM so bleeding does stop, but there is always a risk of rebleeding so specialist assessment is advisable.  If bleeding occurs in the nose or throat, then ‘packing’ may be needed, something that any Emergency Department should be equipped to perform. 

Ulceration of the skin with or without bleeding is another complication which can be seen later in the development of the AVM.  This happens because although there is a lot of blood flowing through the tissue, the normal slowing of blood flow which allows oxygen and nutrient exchange does not occur and the tissues are effectively starved of blood supply.  Even nearby tissues, including skin, which are not part of the AVM may not be adequately oxygenated and nourished because the high pressure in the veins prevents the normal circulation of blood from the arterial to the venous system.  Treating the AVM itself is the only effective way to treat this problem. 

In a very advanced state, an AVM can cause heart failure, but this is rare.

Because such a large volume of blood flows through an AVM, the arteries which flow into it and the veins which drain away from it become quite large.  This is an inevitable consequence of the presence of the AVM, but has no particular significance beyond this.  Sometimes people develop the impression that the ‘feeding vessels’ are a cause ofthe AVM rather than the consequence, but this is false.  If the AVM is successfully treated, the feeding vessels will mostly gradually return to normal.

The two most commonly applied treatments for AVM are surgery and interventional radiology.  

In surgery, the aim is generally to remove the whole of the AVM which generally involves removing the tissues it infiltrates and inevitably some of the surrounding tissue.  There will often be some sort of reconstruction required to restore normal appearance.  AVM surgery is challenging, firstly because the amount of blood flow causes bleeding which can be difficult to manage, and because the margins of the AVM are difficult to define, so it is difficult to be sure that the whole of the AVM has been removed.  If the whole AVM can be removed then the long-term results can be excellent. This is most often possible with small, superficial, well-defined lesions or with ones that are within a structure like a muscle that can be removed, especially when the muscle is one which doesn’t need to be replaced.   A decision about whether an AVM can be completely removed is complex, and needs to be made by a surgeon with experience in operating on AVMs. 

Interventional radiology has the advantage that the radiologist can see the extent of the lesion by imaging at the time of treatment.  A long, flexible catheter is fed up into the AVM where an agent can be injected which obstructs the shunts and stops the abnormal blood flow.  This can be challenging to achieve because of the rapid flow of blood through the AVM into the venous circulation, and the risk that any injected material will pass through the AVM and out the other side.  Treatment therefore needs again to be performed by someone with experience and knowledge of AVM treatment who is competent in a range of specifictechniques.  Multiple, staged treatments are often required.  Various agents are used to treat AVM, which have different properties and risk profiles.  Best results are achieved where the AVM all drains into a single vein.  AVMs that involve the skin can be difficult to treat because the treatment may exacerbate ulceration.

Yes, there is a significant rate of recurrence of AVM after treatment, whether it is surgical, by interventional radiology or by a combination of the two.  This can happen even if angiography performed straight after treatment does not show any AVM still there and can become evident several years after the treatment.   It is important to follow up any AVM so that recurrences can be treated while they are small.

Technically, embolization means injecting glue or a solid implant into the artery which leads into an AVM (sometimes referred to as the ‘feeding artery’, see Q7), while sclerotherapy refers to injecting an agent into the actual shunts which join the arteries to the veins, which is to say the AVM itself.  In the early days of AVM treatment, surgeons would tie off the artery going to the AVM, which would have the immediate effect of making the AVM smaller, so it would appear to be effective, but nearby arteries would soon open up and restore the circulation so nothing had really been achieved unless the AVM itself was removed as well.  Sometimes, especially early in the development of interventional radiology treatment, embolization of the ‘feeding arteries’ has been performed for the same rationale and with the same outcome. Interventional radiologists who have experience and knowledge of AVM know to avoid this trap, but sometimes it still happens that feeding vessels of an AVM are embolized.  This is unfortunate because it blocks access to the AVM that is needed for any future interventional radiology treatment.  

There are some circumstances where embolization of an AVM is appropriate,  for instance embolization is sometimes performed prior to surgery, to reduce the amount of bleeding during the operation.  It can also be used as a temporary measure to control bleeding.

Given the rarity of AVM, there will never be a comparative study between these two types of treatment and so no clear way to provide a definitive answer to this question. Both treatments have advantages and disadvantages.  In some cases one or the other treatment method is clearly likely to be better, and sometimes a combination of treatments is best.  Apart from these few cases, however, there is a large grey area. Rather than one treatment being better than the other, what is probably more important is the experience and expertise of the treating radiologist or surgeon. The best care for AVM is provided by doctors working in a multidisciplinary team where the expertise of both specialties is brought to the table in a cooperative manner.

Given the limitations with conventional treatment, there is considerable interest in developing new treatments for AVM.  The finding that AVMs are caused by somatic genetic mutations, and that those mutations are similar to ones seen in cancer, has led to considerable interest in repurposing drugs, especially drugs used in cancer, to treat AVM.  There is considerable interest at present in use of the cancer drug Bleomycin which can be injected into the AVM with or without ‘electoporation’ to enhance uptake by the cells. Drugs called MEK inhibitors are being trialled as well as thalidomide. 

Given the slow rate at which AVMs grow, and it being a rare disease, it is challenging to obtain good long-term data on the results of new treatments.  There is a history treatments looking promising at first but having more disappointing long-term outcomes so there is always a need for caution.    It will likely be some time before we know for surewhich new treatments will give the best results for AVM but there is more hope at the moment than there has been for some time.

As noted above, if there is suspicion that there is an inherited underlying tendency to AVM, this can be an indication for testing the patient and clinically affected family members, to inform the need for screening tests for AVM.  For sporadic lesions, genetic testing does not currently make a significant difference to the standard management of AVM and is therefore mostly available only through research programs.  It is a prerequisite for entry into some trials of new targeted therapies for AVM, and so may provide access to new treatments.  As more targeted therapies become available it is likely that knowing the exact genetic variant which has caused the AVM may become of increasing importance.  There is also some emerging evidence that AVMs with some variants may progress more rapidly than others, which may feed into decision making about treatment.  Genetic testing for sporadic lesions requires testing of the tissue from the AVM itself, so requires a biopsy or surgery to have been performed, although there is increasing interest in ‘liquid biopsy’ testing of blood, especially blood taken at the time of sclerotherapy.  Testing may also be performed on surgical samples, even sometimes tissue from surgery performed some years ago.