Radiological aspects of gamma knife radiosurgery for arteriovenous malformations and other non-tumoural disorders of the brain

Acta Radiol Suppl. 1993;388:1-34.


The aims of the thesis were to investigate stereotaxic procedures in radiosurgery for cerebral arteriovenous malformations (AVMs) and radiation effects of single session high-dose irradiation delivered by gamma knife on the human brain. Investigation of gamma knife radiosurgery in 1,464 patients constitutes the data base of this thesis. High quality stereotaxic angiography is the gold standard targeting imaging in radiosurgery for cerebral AVMs, particularly for small AVMs or residual AVMs after other treatments. For medium and large size AVMs, stereotaxic MR techniques can improve targeting precision and decrease irradiation volume as compared to stereotaxic angiography in selected cases provided that proper pulse sequences are used. Combined treatments, where embolization precedes radiosurgery, can improve amenability of the treatment for large AVMs. This is on condition that the partially embolized nidi are well delineated and the volume of the residual nidi has been decreased to a level where an optimum irradiation can be safely prescribed. Radiologically, adverse radiation effects (ARE) of gamma knife radiosurgery for cerebral AVMs are observed in 16% (131/816) of the patients. The ARE are observed as a focal low attenuation on CT or as a focal high signal on MR image without enhancement in 47% (61/131), and as a peripheral or homogeneous enhancing lesion in 48% (63/131). MR imaging is more sensitive than CT in detecting the ARE. 91% of the ARE are observed within 18 months after radiosurgery and 89% are seen to regress within 18 months. Clinically, symptomatic ARE are only observed in 6% (51/816) and only in half of them, i.e. 3%, are the symptoms permanent. The risk of ARE in radiosurgery for venous angiomas is higher as compared to AVMs. Other mechanisms have probably been employed. In gamma capsulotomy, the necrotic lesions and reaction volumes created by using multiple isocentres of 4 mm collimators are less predictable as compared to that by single isocentre. Volume effects and depreciation of the steep isodose gradient are hypothesised as the leading factors of the inconsistency. Based on the in vivo assessment of the radiation effects observed on the basically normal human brain it is concluded that irradiation volume is strongly related to the radiation effects and is one of the important considerations in decision making for radiosurgery. Volume of brain tissue exposed to irradiation could be minimised and precision of targeting could be maximised provided that a proper stereotaxic imaging is used.(ABSTRACT TRUNCATED AT 400 WORDS)

Publication types

  • Multicenter Study

MeSH terms

  • Anxiety Disorders / surgery
  • Brain / diagnostic imaging
  • Brain / pathology
  • Brain / surgery*
  • Brain Neoplasms / surgery
  • Cerebral Angiography
  • Cerebrovascular Circulation
  • Combined Modality Therapy
  • Embolization, Therapeutic
  • Follow-Up Studies
  • Hemangioma / surgery
  • Humans
  • Intracranial Arteriovenous Malformations / diagnostic imaging
  • Intracranial Arteriovenous Malformations / pathology
  • Intracranial Arteriovenous Malformations / surgery*
  • Intracranial Arteriovenous Malformations / therapy
  • Magnetic Resonance Imaging
  • Obsessive-Compulsive Disorder / surgery
  • Radiosurgery* / adverse effects
  • Radiosurgery* / methods
  • Radiotherapy Dosage
  • Stereotaxic Techniques
  • Tomography, X-Ray Computed