In Reply: Stereotactic Radiosurgery Provides Long-Term Safety for Patients With Arteriovenous Malformations in the Diencephalon and Brainstem: The Optimal Dose Selection and Long-Term Outcomes

To the Editor: We sincerely appreciate the opportunity to respond to the Letter to the Editor from Dr Wang1 on our recent article titled “Stereotactic radiosurgery provides long-term safety for patients with arteriovenous malformations in the diencephalon and brainstem: the optimal dose selection and long-term outcomes.”2 We respectfully acknowledge Dr Wang for his thoughtful and encouraging comments on our study. The study analyzed the long-term outcomes of stereotactic radiosurgery (SRS) using the gamma knife for arteriovenous malformations (AVMs) in the diencephalon (DC) and brainstem (BS), which are difficult to access surgically; importantly, their rupture can lead to poor functional outcomes. One of the most important therapeutic factors affecting the outcome of SRS for AVMs is a prescribed radiosurgical dose.3,4 Our study highlighted that the prescription dose should neither be too low nor too high for DC-AVMs and BS-AVMs to balance sufficient therapeutic effects and good functional preservation. Owing to fear of radiation injury to the important neuroanatomic structures clustered within the DC and BS, lower radiosurgical doses tend to be prescribed. However, this may actually reduce the chance of AVM obliteration. In fact, AVM obliteration was significantly less likely in the low-dose group (<18 Gy) than in the medium-dose (18-20 Gy) and high-dose (>20 Gy) groups. More importantly, this lower obliteration rate resulted in a higher post-SRS hemorrhage rate in the low-dose group than in the medium-dose and high-dose groups, although this difference was not statistically significant. In SRS for DC-AVMs and BS-AVMs, complete AVM obliteration is the most important contributor to good neurological outcomes. It is noteworthy that although the high-dose and medium-dose groups showed similar rates of AVM obliteration, the former was associated with a higher risk of adverse radiation effects (AREs). Although this difference did not reach statistical significance, it can potentially guide the choice of radiosurgical doses for DC-AVMs and BS-AVMs. Overall, medium-dose SRS appeared to provide a better functional outcome than the low-dose SRS which was insufficient for hemorrhage prevention and the high-dose SRS which increased the risk of AREs because of excessive irradiation. Based on our results, the optimal dose for DC-AVMs and BS-AVMs should be 18 to 20 Gy. Although this may sound slightly too aggressive, our emerging techniques, such as the integration of three-dimensional rotational angiography and tractography, will enable high-precision conformal targeting, avoiding important nerve tracts and contributing to a safer administration of the definitive dose to DC-AVMs and BS-AVMs (a representative case is shown in Figure1).5-9 Nevertheless, as Dr Wang stated, further multicenter randomized controlled clinical trials are needed to validate our results because the differences were not statistically significant.FIGURE.: Radiosurgical plans after the integration of 3DRA and Q-ball tractography in a 37-year-old woman with an unruptured brainstem arteriovenous malformation. The nidus is 26 × 24 × 20 mm and difficult to identify because of the diffuse vascular structure. The integration of 3DRA and Q-ball tractography shows clear and detailed nidus and corticospinal tracts. The upper row shows a brainstem arteriovenous malformation before radiosurgical planning. The middle row shows the radiosurgical dose planning based on the 3DRA and Q-ball tractography findings. The yellow shapes indicate the target area for the prescription dose. The cyan and purple shapes indicate the right and left cortico-spinal tracts, respectively. E1, The 3-dimensional view of the target nidus (yellow) and the right (cyan) and left (purple) corticospinal tracts. E2, The lateral aspect of the preradiosurgery digital subtraction angiography image. E3, The image obtained after obliteration. A1 and A2, 3DRA images; B1 and B2, gadolinium T1-weighted images; C1 and C2, computed tomography angiography images; D1 and D2, time-of-flight images; and E2 and E3; digital subtraction angiography images. 3DRA, 3-dimensional rotational angiography.