News

Radiosurgery treatment of brainstem cavernous hemangioma

Radiosurgery treatment of brainstem cavernous hemangioma. Cavernous hemangioma (Cavernoma, CA) is a common vascular malformation in the brain, accounting for about 10% to 15% of all cerebral vascular malformations, so it is also called cavernous malformation (CM).

1 Introduction

Cavernous hemangioma (Cavernoma, CA) is a common vascular malformation in the brain, accounting for about 10% to 15% of all cerebral vascular malformations, so it is also called cavernous malformation (CM).

Anatomically, these lesions consist of abnormal cystic vascular lumen covered by a single layer of endothelial cells, and the tight junctions between these endothelial cells are not perfect. With the increase in the use of magnetic resonance imaging (MRI) technology in the brain, more and more asymptomatic CAs have been accidentally discovered, and their actual incidence is on the increase. It is estimated that almost 40% of CAs are accidentally discovered.

The age of onset of CA patients is usually 40 to 50 years old, and about 25% of the cases develop in childhood. Depending on the location of the lesion, the clinical symptoms can be epilepsy, headache, progressive neurological deficit or intracranial hemorrhage. The most common symptoms of supratentorial CA are new-onset epilepsy (40% to 70%), symptomatic hemorrhage (25%), and focal neurological deficits without acute hemorrhage imaging.

There is a hypothesis that the formation of sporadic CA may be related to Developmental Venous Anomaly (DVA). Ciricillo et al. hypothesized that the venous hypertension associated with DVA can allow red blood cells to enter the extracellular space and release vascular growth factors to form CA.

Others hypothesize that increased pressure in the DVA lumen can cause decreased tissue perfusion and hypoxia, which can stimulate local angiogenesis and form CA. Approximately 20% to 30% of clinical cases have associated DVA. However, 7T MRI found that almost all sporadic CAs have small venous malformations, which indicates that DVA may play a role in the formation of sporadic CA. In addition, radiotherapy is one of the reasons for the formation of CA, which mostly occurs in young patients, and the lesions can be multiple.

Up to 20%~50% of CA are familial cases, following somatic dominant inheritance, and the related genes are CCA1 (KRIT1), CCA2 (MGC4607) and CCA3 (PDCD10) located on chromosomes 7q, 7P and 3q, respectively. The corresponding mutation rates are 53%-65%, 15%-19%, and 10%-16%. CCA3 variants are associated with more severe clinical manifestations, manifested as earlier ICH, and in some cases multiple meningiomas.

Familial CA cases usually have multiple lesions, while sporadic cases are usually single lesions.

2. Diagnosis of cavernous hemangioma (imaging features)

CA can be accurately diagnosed entirely by imaging examinations (especially MRI). On CT, CA shows non-specific calcification, while on MRI, CA can show typical “popcorn-like” signs, indicating that the lesion contains large sinusoidal spaces, bleeding signals of different periods, and calcification; On the T2-weighted image, a typical “hemosiderin ring” can be seen; multiple lesions and related venous malformations can be revealed.

But acute appearance can mask the above-mentioned typical characteristics. The traditional T2-weighted gradient echo sequence (Gradient Recalled Echo, GRE) can reveal the “blooming” effect of hemosiderin, increasing the sensitivity of CA detection. In particular, the sensitivity of Susceptibility Weighted Imaging (SWI) is higher, and a large number of lesions that are not displayed on traditional imaging can be found in familial cases.

Zabramski classified CA into 4 types according to its NMR performance: Type Ⅰ lesions, which are high signal on T1 and T2 sequences, often combined with mild space-occupying effects and varying degrees of edema; Type Ⅱ lesions are manifested on T2 sequence It is a grid-like structure surrounded by a hemosiderin ring; Type Ⅲ lesions only have hemosiderin deposits on the standard T2 sequence without a network-like central structure; Ⅳ lesions can only be visualized on GRE or SWI sequences.

Because CA is a low-pressure system, these vascular lesions are not developed in diagnostic angiography, so they are called contrast-negative vascular malformations. However, the possibility of other vascular malformations can be ruled out through angiography, and coexisting DVA can be found at the same time.

Quantitative Susceptibility Mapping (QSM) and Dynamic Contrast Enhanced Quantitative Perfusion (DCEQP) can be used to measure iron ion deposition and vascular permeability in CA, respectively, and may be used to evaluate the clinical characteristics of CA .

3. Treatment of cavernous hemangioma

Some scholars believe that CA that includes the following conditions should be actively treated: refractory epilepsy, obvious progressive neurological deficits, first symptomatic bleeding in non-functional area lesions, and second symptomatic bleeding in functional areas (including brain stem) .

In 2017, the Angioma Alliance Scientific Advisory Board published CA clinical treatment expert guidelines, which mainly include:

At the time of diagnosis, 3 generations of family history should be collected;

Carry out genetic testing including CCA1, CCA2 and CCA3 for patients with multiple lesions and who have not received radiotherapy, and give genetic advice to abnormal cases;

Conservative imaging follow-up for asymptomatic lesions;

Anti-epileptic drugs are given for CA-related epilepsy, but when the epilepsy progresses to intractable epilepsy, surgery can be considered (early resection of a single lesion can achieve better epilepsy control);

Generally speaking, surgical resection is only for symptomatic lesions. For resectable lesions, surgical resection can be performed after the first symptomatic bleeding (because the 5-year rebleeding rate has risen to 29%);

For deep lesions, surgery is usually performed after the second symptomatic bleeding;

For lesions with high surgical risk and difficult to remove, radiosurgery can be considered;

It is not recommended to treat asymptomatic lesions in familial cases.

In addition, some pathological characteristics of CA are similar to cardiovascular diseases or tumor diseases, such as autophagy, angiogenesis, inflammation, and oxidative stress. These commonalities make it possible to discover new effects of some drugs.

Currently, clinical trials are evaluating the role of statins in controlling the progression of CA. McDonald et al. found that the Rho enzyme inhibitor, fasudil, can reduce the incidence, size and bleeding probability of CCA in CCA1 mice.

4. Characteristics of brainstem cavernous hemangioma

CA is usually located on the supratentorial. Brainstem CA accounts for about 8%-22% of all cases. In brainstem lesions, the pontine is the most common location. It can be used as an independent lesion of sporadic CA, or as a local manifestation of familial CA.

4.1 Bleeding rate

In population studies and case series studies without inclusion criteria, the risk of first symptomatic bleeding in CAs discovered by accident is very low, about 0.08%/person.year. However, once bleeding symptoms appear, the annual risk of re-bleeding increases significantly, and most of them are estimated to be at least 10 times higher than before.

The bleeding rate of brain stem disease and the recurrence rate of symptomatic bleeding are both high. For non-familial brainstem CA, the risk of symptomatic bleeding is about 0.25%~6.5%/person.year. Because these lesions are more in superficial areas, the annual bleeding rate can rise from 3.8% to 35% after the first bleeding. The risk of recurrence within 2 years after the first bleeding is significantly increased, but it seems that this risk is significantly reduced again after 2 years.

This phenomenon is called “time clustering” and has been confirmed by many studies. Because brainstem CA is adjacent to the nerve nucleus and motor and sensory fiber bundles, about 60% of brainstem CA bleeding are symptomatic bleeding. For brainstem CA hemorrhage, Abe et al. reported that the incidence of severe neurological symptoms after hemorrhage was 21%, and the incidence of persistent neurological deficits after 1 month was 6.8%.

4.2 Risk factors for bleeding

Women, epilepsy, and familialism are considered risk factors for bleeding, but there is still a lack of strong evidence. Some early prospective studies on the natural history of CA indicate that the location of the lesion is not a significant risk factor for bleeding. This conclusion is still controversial. Porter et al. reported that the bleeding risk of subtentorial CA was 3.8%, while that of supratentorial CA was 0.4%. They believed that deep CA, especially lesions located in the brain stem, had a lower prognosis than supratentorial lesions. Two recent prospective studies have also shown that the risk of brainstem CA bleeding has an increasing trend, but the results have not reached statistical significance.

4.3 Surgical treatment

The choice of surgery for brainstem CA lesions should be cautious, and the risk of surgery, the natural history of the lesion, and the location of the lesion should be weighed. In case reports, the disability rate and fatality rate of supratentorial CA surgery are low, but the risk of brain stem disease is clearly increased.

Therefore, for asymptomatic lesions, especially those located deep in the brain tissue or the brain stem, surgical resection has not been universally approved. Therefore, surgery is usually used for symptomatic lesions that are easy to remove, because these lesions have an increased risk of rebleeding and a lower rate of surgery-related disability.

It has been reported in the literature that nearly half of the cases of surgical removal of brainstem CA have a higher disability rate in the early stage, but most patients can get better over time. In a group of 137 cases of surgically treated brainstem CA, 77% of patients had cranial nerve injury, and 53% of patients had hemiplegia of the limbs. The most common symptoms of lesions located in the midbrain were diplopia (69%), hemiparesis (48%), hydrocephalus (38%), and ataxia (38%).

Individual patients have red nucleus tremor, involuntary laughter, paroxysmal coma, and vertical gaze disturbance. A total of 90 patients in this group had lesions in the pons, and their main symptoms were cranial nerve dysfunction (76%), hemiparesis (57%), hemisensory disorder (49%) and dizziness (44%). All patients with lesions located in the medulla oblongata have dysphagia, while 44% of patients have hemiparesis, ataxia, and hemisensory disorders.

And 28% of patients have specific refractory hiccups. In another group of 300 cases of brainstem CA treated by surgery, Abla et al. reported that surgery can significantly improve the risk of re-bleeding and related symptoms. They believe that surgical resection should be considered for patients who are resectable.

However, about 53% of patients experienced aggravation of neurological symptoms after surgery, 36% of patients developed new neurological dysfunction and long-term existence, and 28% of patients experienced perioperative complications. A meta-analysis by Gross A et al. included 68 studies involving 1390 cases of surgical treatment of brainstem CA and found that the persistent disability rate was 14% and the fatality rate was 1.5%. The total resection rate was 91%, 62% of patients with partial resection had rebleeding, and the fatality rate was 6%.

Therefore, the author recommends proper evaluation of brainstem CA patients before surgery, and recommends that only superficial (close to the pia) bleeding lesions be surgically removed. The skull base approach can reduce the difficulty of surgical resection of brainstem lesions. The combined application of multiple technologies including image navigation, neuroelectrophysiological detection, and laser-assisted technology can improve the effect of surgical resection. However, there is no controlled study to prove the special effect of a certain way.

Tanaris et al. conservatively treated 15 patients with brainstem CA, and performed surgical treatment on 6 patients. They found that during the 79.7-month follow-up, there was no significant difference in the neurological function assessed by the Rankin score between the two groups.

Therefore, when there is no prospective randomized study, it is recommended to adopt the principle of individualization for the treatment of patients. Issam A. Awad et al. proposed some consensus on the surgical treatment of CA, some of which involve brainstem CA and radiosurgery treatment: surgical resection of asymptomatic CA is not recommended, especially for lesions located in functional areas, deep brain tissue, brain stem or spinal cord For multiple asymptomatic lesions, surgical resection is also not recommended.

(Class III, Level B evidence); For deep CA with symptoms or previous bleeding, but the risk of surgical disability and death is equivalent to the risk of survival with disease for 5-6 years, surgical resection may be considered. (Class Ⅱb, Level B evidence); For the brainstem CA of the second symptomatic hemorrhage, after evaluating the risk of early postoperative disability and mortality, and the impact on the quality of life, a complete resection can be considered Because the course of these lesions is more aggressive (Class IIb, Level B evidence); for a single disabling hemorrhage of the brain stem or spinal cord CA, it is not a strong indication for surgery. (Class IIb, Level C evidence); For CA lesions located in important functional areas, radiosurgery can be considered when symptomatic bleeding occurs and the surgical risk is unacceptable. (Class IIb, Level B evidence); Radiosurgery is not recommended for asymptomatic CA, resectable CA, and familial CA (because it may produce new lesions) (Class III, Level C evidence);

For brainstem CA, the operation or not depends more on whether the lesion is located near the surface of the pia mater. Although surgery is an effective treatment, it is not the first treatment for deep asymptomatic or mildly symptomatic CA.

4.4 Radiosurgery treatment

4.4.1 The influence of radiosurgery on bleeding rate

Initially, SRS was proposed as a treatment option for aggressive brainstem CA lesions located deep in the parenchyma and difficult to resected by surgery. After CA treatment, neither angiography nor imaging examination can confirm the occlusion of the vascular structure. Therefore, the current research focuses on the reduction of bleeding after SRS treatment. Most cohort studies found that the bleeding rate decreased after 2 years of SRS treatment. However, the causal relationship between the two is difficult to determine, because some scholars believe that in the natural history of CA, after 2 years of bleeding, the bleeding rate also has a downward trend.

A study of gamma knife treatment of brain stem CA included 68 patients. Monaco et al. found that after 2 years of follow-up, radiosurgery can reduce the annual bleeding rate from 32% to 1.3%. Lunsford et al. also reported that radiosurgery can reduce the bleeding rate from 32% to 1% after 2 years. As mentioned earlier, the risk of bleeding in CA may be significantly reduced by the invention 2 years after the first bleeding. Therefore, the positive results after radiosurgery may only reflect the natural course of these lesions. The use of radiosurgery to treat CA remains controversial.

In order to study the specific changes of bleeding rate after SRS treatment, scholars have carried out further research. Lunsford et al. reported the therapeutic effect of using SRS to treat 103 cases of single symptomatic CA. During the 17-year follow-up, the annual bleeding rate was 10.8% in the first 2 years after treatment, and then decreased to 1.1%.

A meta-analysis of 178 cases of SRS-treated brainstem CA included in 4 studies showed that after SRS treatment, the reduction in the risk of CA bleeding was statistically significant. Sheffield National Gamma Knife Center et al. found that the annual bleeding rate in the high-risk group dropped from 30% to 15% within the first 2 years after SRS treatment, and then to 2.4% thereafter.

Another study of brainstem CA in most cases. During a 16-year follow-up period, within the first 2 years after SRS treatment, the bleeding rate dropped from 39% before treatment to 9.7%. In the subsequent follow-up period, only 2 patients had new bleeding, indicating that the rebleeding rate after the biological effect of SRS was about 0.56%. In another meta-analysis of gamma knife radiosurgery for CA, most of the lesions are located in the deep brain tissue, especially the brain stem.

The study found that the total RR of the bleeding rate before and after GKRS treatment was 6.08, indicating that the bleeding rate of CA was reduced after GKRS treatment. The bleeding rate before treatment and the bleeding rate within the first 2 years after treatment have a total RR of 3.30. Comparing the bleeding rate before treatment with the bleeding rate after 2 years of treatment, the total RR was 12.13. This shows that the bleeding rate has decreased within the first 2 years after treatment and after 2 years. And there was no significant difference in bleeding rate within 2 years and after 2 years (RR: 2.81). Therefore, we have reason to believe that after SRS treatment, the probability of brainstem CA hemorrhage is continuously decreasing.

These studies are retrospective studies, so there is a natural selection bias: these patients entered the study cohort after bleeding. The observation period before treatment must be very short, which makes the bleeding rate before treatment tend to be higher. The uncertainty of bleeding rate before treatment makes the measurement and interpretation of RR lack credibility.

Because the calculation of bleeding rate before SRS treatment is a key parameter to study whether SRS really reduces the bleeding rate. Currently, the limitations of radiosurgery treatment of CA are the small number of cases, short follow-up period, and lack of comparison with untreated brainstem CA. However, a systematic review and meta-analysis of single-cohort studies have given us a better understanding of the role of SRS in the treatment of brainstem CA. SRS is still one of the good treatment options for inoperable or high-risk brainstem CA.

4.4.2 Radiosurgery complications

The incidence of long-term neurological deficits after SRS treatment varies greatly with the location of the disease. The incidence of CA located deep in the brain stem or brain tissue is 0%~59%. Lunsford et al. reported that the disability rate was 13.5% in the cohort. Therefore, the high risk of radiosurgery for brainstem CA is questioned. The most common related complications in radiosurgery are neurological deficits, including headache, dizziness, facial paralysis, dizziness, limb weakness, facial numbness, diplopia, dysarthria, etc. The hemosiderin in the parenchyma around CA is a potential radiosensitizer and may be the reason for the increase in complications. The complication rate of SRS is related to the radiation dose. In most studies, the average marginal dose is less than 16Gy. However, the optimal dose that can produce therapeutic effects and reduce the incidence of radiation damage is currently unclear. Studies have shown that low-dose (13Gy) SRS irradiation can determine a good treatment effect.

4.4.3 Timing of radiosurgery treatment

Recently, a study conducted a retrospective study on the SRS treatment of the first hemorrhage and reoccurring brainstem CA, and found that there was no significant difference in the annual bleeding rate between the two groups. Natural history shows that patients who are treated after more than one bleeding are in worse condition. Therefore, some scholars recommend radiosurgery treatment as soon as possible after the first bleeding. Because the disability rate caused by repeated bleeding is cumulative, and the current adverse reaction rate caused by radiotherapy is very low, some scholars advocate early treatment.

4.4.4 Pathological changes after radiosurgery treatment

It is documented that radiotherapy can cause arteriovenous malformations and vascular wall thickening and hyaline degeneration. The same pathological changes of radioactive vessels can also be seen in CA resected after radiotherapy. Gewirtz performed pathological examination on 8 cases of CA that were resected due to rebleeding 1-10 years after radiotherapy, and found that the irradiated lesions had specific fibrinoid necrosis, but none of the lesions completely formed thrombus. The study found that 75% of the lesions resected after 5 years of SRS treatment were occluded without any change in volume. These specimens are from lesions that still have symptoms after radiotherapy. Therefore, it is reasonable to think that the silent lesions may have complete occlusion after treatment. Another hypothesis is that even if there is no complete occlusion, the scar-like thickening of the outer wall of the lesion and the stable foot of the low-pressure lesion can prevent the lesion from bleeding again. Due to the lack of effective methods to confirm the pathological changes of CA after GKRS treatment, the onset time and pathological changes after treatment have not been confirmed.

5. Sum up

with the advancement of brain imaging technology, the improvement of the dose planning software system, and the optimization of dose delivery, SRS is becoming more effective and safer for the treatment of brainstem CA; for asymptomatic lesions, in order to further reduce lesion bleeding Opportunity to avoid serious neurological dysfunction caused by this, unless the lesion is larger and located in a more important part (medulla oblongata), early active SRS treatment can be used; for symptomatic lesions, active treatment should be taken, and a lower dose can achieve better Therapeutic effect: During the dosage planning, hemosiderin ring should be excluded from the target area as much as possible, which can reduce the risk of complications; and for the lesions with superficial location and repeated bleeding, surgical treatment should be actively performed.

The focus of future research is to conduct prospective studies of large numbers of patients and long-term follow-up for patients whose SRS is the first choice for brainstem CA treatment.