Risks of Radiosurgery

Introduction

Intracranial arteriovenous malformations (iAVMs) are congenital vascular anomalies composed of blood vessels that directly shunt blood from arteries to the venous system without an intervening capillary network. With a reported incidence of 1.12–1.34 per 100,000 person-years, AVMs account for approximately 1.5%–4% of intracranial masses and are thought to occur one-tenth as often as intracranial aneurysms.

Patients with iAVMs can present with headaches, seizures, progressive neurologic deficits, and mental deterioration. However, hemorrhagic stroke is the most frequent and severe life-threatening complication of these lesions. Intracranial AVMs account for 9% of subarachnoid hemorrhages and 1%–2% of all strokes. Various studies have reported the annual risk of iAVM hemorrhage to be approximately 2%–5%, with 20%–30% morbidity and 10%–15% mortality per event. Once a patient has experienced iAVM hemorrhage, the risk of recurrent hemorrhage is increased—especially in the first year after the event. In light of the significant hemorrhage risks, the main goal of treatment is completely resecting or obliterating the AVM and excluding it from circulation, thus preventing any future hemorrhage.

The management options for iAVMs include observation, microsurgery, endovascular embolization, and stereotactic radiosurgery (SRS). Multimodality management has been shown to increase therapeutic efficacy and improve patient outcomes and is currently the standard of care for patients with iAVMs.

Stereotactic radiosurgery was initially described by Lars Leksell in 1951 ; the concept was a breakthrough in neurosurgery and changed the management of many neurological pathologies, including AVMs. SRS can be performed with various modalities that achieve the desired characteristics of small fields, steep dose fall-off, and highly accurate targeting through the use of two fundamental principles: superposition of beams and stereotactic targeting.

As an alternative to surgical removal (resection), SRS is a safe and effective method of managing iAVMs. The obliteration rate has been reported at 70%–90% at 3–5 years after treatment. In general, SRS is indicated for small- to moderate-sized AVMs located in deep or eloquent brain areas and for patients with a high risk of perioperative complications (e.g., elderly patients, patients with medical comorbidities). Given its less invasive nature and favorable benefit-risk ratio, superficial iAVMs amenable to surgery have been treated with SRS. Recent reports describe staging strategies for effectively treating large iAVMs as well.

Although SRS is a minimally invasive technique, it is not risk-free. Because it induces AVM obliteration progressively, both the beneficial and adverse effects manifest in a delayed manner, typically months or even years after treatment. The primary disadvantage of SRS lies in the risk of hemorrhage during the latency interval between SRS and total obliteration of the AVM. Additionally, adverse radiation effects (ARE) may occur. These reactions have been stratified into acute effects; early delayed effects (occurring a few weeks to several months after SRS), mainly consisting of radiation-induced imaging changes (RIC); and late delayed effects (occurring several months to several years after SRS), including radiation necrosis, cyst formation, encapsulated hematoma, stenosis of major vessels, and radiation-induced neoplasm.

Ilyas et al. performed a meta-analysis of 51 studies of iAVMs treated with SRS and found overall rates of RIC, radiation necrosis, and cyst formation of 35.5% (1143/3222 patients), 2.9% (40/1360 patients), and 3.2% (35/1081 patients), respectively. One should note that there is significant heterogeneity in the literature concerning the definition of different radiation-induced complications, the indications treated, and the SRS modality used, which confounds the generalizability of these studies.

Several factors, including radiation dose and target volume (> 3 cm ³ ), are reported as important predictors for ARE. Other risk factors, such as prior hemorrhage, AVM location, and repeated radiosurgery, have been described in some series. The relationship between AVM embolization and post-SRS outcomes is multifactorial and complex.

Risk of Hemorrhage Post-SRS

Hemorrhagic stroke is both the most severe and the most frequent life-threatening complication of harboring an iAVM. It is hence understandable that significant efforts have been undertaken to identify risk factors predicting AVM rupture. However, considerable disagreement exists about the risk of AVM bleeding after SRS. While some investigators have reported an increased risk, more extensive and more detailed analysis of this question has confirmed that the risk is either unchanged or decreased.

Colombo et al. reported on 180 iAVM patients treated with SRS with a mean follow-up of 43 months. In cases with complete irradiation, the bleeding risk decreased from 4.8% in the first 6 months after SRS to 0% starting at 1 year following SRS. However, in patients with partially irradiated iAVMs, the bleeding risk increased from 4% to 10% over the first 2 years, then decreased.

Additionally, Ding et al. analyzed 2320 iAVM cases. The authors reported a decline in hemorrhage rate from 15.4 per 1000 person-years to 11.9 with SRS ( P = .001). However, SRS only improved the hemorrhage risk of Spetzler-Martin grade I–III AVMs, while the risk increased for grade IV–V AVMs. Also, deep-seated iAVMs were prone to hemorrhage during the latency period, whereas the use of a higher margin dose decreased the risk of post-SRS hemorrhage. The presence of an untreated AVM-associated aneurysm was the strongest predictor of post-SRS hemorrhage. Thus targeted occlusion of AVM-associated aneurysms via embolization or surgical clipping should be considered in most cases to reduce the risk of hemorrhage during the latency period following SRS.

Similarly, Chye et al. reviewed data from 1515 iAVM patients with long-term follow-up and found that the annual hemorrhage rate was lower in the SRS group (1.59 per 100 patient-years) than in the non-SRS group (1.99 per 100 patient-years). Among the patients with ruptured iAVMs, the annual risk of bleeding in the Gamma Knife radiosurgery (GKS) group was significantly lower than that in the non-GKS group. Furthermore, the authors found that SRS treatment of unruptured AVMs in patients who are more than 40 years old might increase the risk of hemorrhage. Tonetti et al. reported in their study, which included 233 ARUBA-eligible patients with unruptured iAVMs, that SRS was associated with low annual rates of hemorrhage (0.4%) and stroke or death (0.8%) among the 218 patients with more than 3 years of follow-up.

Subtotal obliteration, defined by Steiner et al. as the disappearance of the nidus on angiogram but with the persistence of an early draining vein, appears to be associated with a decreased risk of hemorrhage as well. Pollock et al. reported 19 such cases with no instances of hemorrhage. Nevertheless, as long as an early draining vein is present, evidence for continued shunting, there is some residual degree of hemorrhage risk.

Acute Effects

Immediate complications are rare after SRS, in contrast to resective surgery, and patients can expect to be able to promptly return to their regular activities after radiosurgical treatment. It is for these reasons that many patients and physicians choose this modality. Most acute reactions are likely due to edema and appear to be self-limited. Headache is treated with mild analgesics. Ice packs are applied as necessary to pin sites. It is essential to confirm compliance with medication regimens before SRS in patients with a known seizure history or patients at increased risk of seizures (e.g., those with subcortical lobar AVMs).

Reporting on 247 iAVM patients, Steiner et al. noted 29 patients (12%) with uncontrolled seizures following SRS. Of the 188 patients without a history of seizures before treatment, 11 (6%) developed new-onset seizures. In contrast, Liščák et al. reported that termination of seizures or improvement in seizure frequency was observed in 42% of patients after SRS, while a worsening of epilepsy was detected in only 4% of patients. Similarly, Peciu-Florianu et al. reported seizure-free rates of approximately 80%, with seizure being the most common symptom in their study’s population.