Treatment options include observation alone, microsurgery, focused irradiation and embolisation. Management by observation alone may be appropriate. The decision to refrain from intervention depends on an understanding of the risks of treatment and the risks of non-treatment. The risk of hemorrhage (a bleed into the brain known as a stroke) averages 2 to 4 % per year. This means that over the next 10 years the risk of hemorrhage approximates 30%, over the next 20 years approximates 50% and over the next 30 years approximates 66%. If a hemorrhage occurs the chance of death or permanent deficits affecting brain function is close to 50%.
Hemorrhage occurs because of wear-and-tear changes. Therefore, the chance of hemorrhage varies according to the presence or absence of these wear-and-tear changes. Evidence that wear-and-tear has reached a dangerous point is clearly evident if a hemorrhage has occurred. If this has occurred the risk of another hemorrhage during the next year may be higher than 10%. Another sign of wear-and-tear is the presence of small "blow-outs" (known as aneurysms). In the presence of these aneurysms (seen by angiography) the risk of hemorrhage each year is 7%. This means that over the next 10 years the risk of hemorrhage approximates 50%, over the next 20 years approximates 77% and over the next 30 years approximates 90%. In the absence of hemorrhage or signs of wear-and-tear on angiography the risk of hemorrhage is about 1.7% per year. This means that over the next 10 years the risk of hemorrhage approximates 16%, over the next 20 years approximates 30% and over the next 30 years approximates 40%.
Given the variable risk for hemorrhage, assessment of individual risks may not be possible without a number of investigations, including MRI scan and cerebral angiography. Furthermore, the wear-and-tear changes may take time to develop and repeat investigations may be needed from time to time in order to see if the risks of hemorrhage have changed.
Apart from hemorrhage, other problems can occur. The most common problems that can arise other than hemorrhage are seizures (or epilepsy) and neurological deficits (loss of some form of body function). These problems account for a decline in function in about one quarter of all people with AVMs over a 20 year period from diagnosis. For an AVM to cause epilepsy it must be in a part of the brain that when neurons (nerve cells) are irritated a chain reaction results in many neurons firing to produce a seizure. Not all AVMs are capable of producing seizures. A loss of function in the absence of hemorrhage is usually associated with large AVMs.
Treatment by microsurgery may be recommended as the preferred method of management. The aim of surgery is the complete removal of the AVM without damaging critical brain. A number of considerations need to be made in order to estimate risk of surgery with the following having a large impact on these risks: location; size; blood vessel supply; and drainage.
A grading system is used that allows a rough guide to risks. This grading system (Spetzler-Martin grading) assigns 1 point to AVMs smaller than 3 cm in largest diameter, 2 points to AVMs between 3 and 6 cm in largest diameter, and 3 points for AVMs larger than 6 cm. In addition, a point is added if the AVM is located in critical brain, and a point if the veins drain deeply (into the veins in the centre of the brain).
Therefore, AVMs can be graded from 1 (the simplest surgical cases) to grade 5 (the most difficult AVM). Beside these features, AVMs that are grade 3 with artery supply from small deep arteries (ie. lenticulostriate arteries) are more difficult than grade 3 AVMs that do not have such an arterial supply.
In surgery on more than 700 cases of AVMs performed by Professor Morgan the risk of death and new deficit varies depending upon the grade. For AVMs that were grade 1 or 2 (more than 300 cases) death or new deficit has occurred in 0.7% of patients (95% confidence interval 0-3%). For Grade 3 and 4 AVMs occurring in non-eloquet (or non-critical) brain the risk for some new problem (including death) is 17% (95% confidence interval 10-28%). For patients with grade 3,4 and 5 AVM located in critical brain the risk of a new problem is 21%. Obliteration of AVMs occurred in almost all cases.
Being able to classify AVMs into their grade provides a comparison to treatments by observation, focused irradiation and embolisation. This allows recommendation of the most appropriate treatment.
Results of microsurgery for brain AVM by Professor Michael Morgan
Outcome for surgery of brain AVM
For those selected for surgery, the outcomes and 95% confidence intervals for each of the three Spetzler-Martin Grading system for 630 AVM of the brain undergoing surgery (1989-2013) is illustrated in this figure. Of importance, when trying to understand what the risks of surgery might be, the risks need to be placed in context for experience and generalizability. From our experience we can say that the results for surgery for Spetzler-Martin Grades 1 and 2 AVM can be generalized to all such grades of AVM and we are 95% sure that the risks of producing a permanent new neurological deficit from surgery is less than 3% for these lesions. For Spetzler-Martin Grades 3, for those cases that we have considered suitable for surgery, and not all such cases are suitable for surgery, we are 95% sure that the risk for you developing a new permanent neurological deficit is less than 25% but is greater than 15%. Of these deficits, half are very bad (e.g. this may include paralysis, significnat loss of vision, loss of memory, death). For Spetzler-Martin Grades 4 and 5, very few of these can be operated upon. However, for those few cases that we have considered appropriate to undergo surgery, we are 95% sure that the risks for these cases is at least a 30% of a new permanent neurological deficit and may even be 50%.
Focused irradiation (sometimes called gamma knife or x-knife treatment) is where irradiation is precisely delivered to the AVM without the need to expose the brain in surgery. It works by slowly blocking off the blood vessels within the AVM. The higher the dose of radiation delivered to the margins of the AVM the quicker obliteration is achieved. However, the dose of irradiation that is absorbed by normal brain must be kept below critical levels in order to minimize the risks of radiation damage. AVMs that are best suited to focused irradiation that allows the maximum rate of cure with the least damage to critical brain are small AVMs. Cure takes on average 2 to 3 years to occur. Therefore, the risks of focused irradiation are a combination of the risk of radiation damage to the brain and the risk of hemorrhage occurring before complete obliteration of the lesion.
For AVMs that are small (<3 cm maximum diameter) obliteration can be achieved in approximately 75% of cases in 2 to 3 years. The average risk for these lesions is 8% of some permanent complication (from death to mild deficits). For larger lesions the obliteration rate becomes considerably less.
Microsurgery is generally recommended for grade 1 and 2 AVMs. Focused irradiation may be best suited to small deeply placed grade 3 AVMs fed by deep arteries.
Embolisation involves occluding all or part of the AVM by injecting glue or particles. The embolisation is performed during angiography by the interventional radiologist. Angiography is a procedure that involves inserting a catheter (a fine tube) into the femoral artery (an artery by the hip) and threading the catheter through the arteries of the body into the arteries supplying the AVM. Angiography then is performed by injecting an iodine based fluid whilst X-ray pictures are taken. This shows the AVM.
Embolisation is infrequently used in attempting to cure AVMs but is often used by neurosurgeons to reduce bleeding at the time of surgery. Because the risk of embolisation is significant (as reported in many publications) it is reserved for more difficult surgical cases by many surgeons. At one stage Professor Morgan embolised many of his cases of AVM before surgery to reduce the threat of bleeding. However, he has not embolised a case since the middle of 2008 for this purpose because the additional benefit from embolisation has not reduced the risks of surgery.