Abstract
Objective
In this study, we aimed to evaluate flow characteristics in partially resected arteriovenous malformations (AVMs) and identify factors associated with complete obliteration following Gamma Knife radiosurgery (GKRS).
Methods
Among 587 patients with AVMs treated with GKRS between August 1992 and October 2024, 31 patients with residual AVMs after microsurgical resection were included. Clinical charts and imaging studies were retrospectively reviewed. AVM flow was categorized as low, moderate, or high based on the number of angiographic frames from nidus visualization to draining vein opacification. Kaplan-Meier survival analysis was used to estimate complete obliteration rates, while univariate and multivariate Cox proportional hazards regression analyses were applied to identify factors associated with complete obliteration. The log-rank test was used to compare actuarial complete obliteration rates among the three AVM flow groups.
Results
The median clinical follow-up period after GKRS was 44 months (interquartile range, 35–105). Complete obliteration was achieved in 24 patients (77.4%). The actuarial obliteration rates were 29.0%, 56.7%, and 69.7% at 3, 5, and 10 years, respectively. Complete obliteration occurred in three (37.5%) high-flow, 14 (87.5%) moderate-flow, and all low-flow AVMs. The mean time to obliteration was 131.7 months (95% confidence interval [CI], 92.9–170.6) for high-flow AVMs, 73.0 months (95% CI, 44.2–101.8) for moderate-flow AVMs, and 36.7 months (95% CI, 22.1–51.3) for low-flow AVMs (p=0.004, log-rank test). Prior embolization and lower AVM flow were independently associated with complete obliteration on both univariate and multivariate analyses.
Conclusion
GKRS produced favorable outcomes in patients with partially resected AVMs. Slower-flow residual AVMs and prior embolization were independently associated with higher rates of complete obliteration.
Keywords: Gamma Knife radiosurgery, Cerebral arteriovenous malformation, Clinical outcome
INTRODUCTION
Microsurgical resection of arteriovenous malformations (AVMs) does not always achieve complete removal in all patients. In cases of AVMs associated with hemorrhage, the ability of the surgeon to fully visualize the nidus may be limited by factors such as brain edema and hematoma formation [2,31]. Deeply located AVMs generally yield less favorable outcomes compared with superficial lesions [8,22]. Similarly, AVMs situated in eloquent brain regions may result in neurological deficits, leading surgeons to perform intentional partial resections [17,28]. In theory, prior microsurgical intervention can alter the hemodynamic characteristics of an AVM, as surgery aims to occlude feeding arteries and draining veins and to remove the nidus [15]. Consequently, the transit times of blood flow within these residual AVMs may also be affected. Importantly, residual AVMs continue to pose a risk of hemorrhage, as previous studies have demonstrated a persistent bleeding risk in partially obliterated lesions [13,14]. Although residual AVMs can be managed through repeat surgery, embolization, or Gamma Knife radiosurgery (GKRS), GKRS offers a minimally invasive treatment strategy [5]. Moreover, hemodynamic changes following partial resection may influence the likelihood of obliteration after subsequent GKRS.
While several studies have investigated factors associated with complete obliteration of AVMs treated with GKRS, few have examined these factors in partially resected AVMs [5]. The primary aim of this study was to assess the hemodynamic characteristics of residual AVMs and identify predictors of complete obliteration following GKRS.
MATERIALS AND METHODS
This retrospective study was approved by the Institutional Review Board of Kyung Hee University Hospital (IRB No. 2025-10-063). All patient identifiers were removed before data analysis.
Patient population
Among 587 patients with AVMs treated with GKRS between August 1992 and October 2024, 31 patients with residual AVMs following microsurgical resection were included in this study (Fig. 1). Surgical procedures that did not aim to remove the nidus, such as external ventricular drain insertion or hematoma evacuation, were excluded. Patients without imaging data or those lost to follow-up were also excluded. In total, 31 patients were included in this study.
Fig. 1.

Flowchart showing the inclusion criteria. AVM : arteriovenous malformation, GKRS : Gamma Knife radiosurgery, MRI : magnetic resonance imaging, DSA : digital subtraction angiography.
Data and variables
Data extracted from chart and imaging reviews included patient demographics, AVM angioarchitecture, and dosimetric parameters from GKRS. Variables analyzed for treatment outcomes included sex, age, prior hemorrhage, history of embolization, AVM location, GKRS target volume, margin dose, maximum dose, presence of venous varix or associated aneurysm, deep AVM location, number of feeding arteries, number of draining veins, Spetzler-Martin (SM) grade, and AVM flow characteristics.
GKRS
For GKRS, the Leksell G frame was affixed to the head of the patient under local anesthesia in adults, whereas children younger than 18 years underwent the procedure under general anesthesia. Pre-treatment planning included contrast-enhanced T1-weighted and T2-weighted magnetic resonance (MR) imaging, along with biplane cerebral angiography for all patients. Until January 2002, treatment planning was performed using the Leksell KULA system (Elekta Instruments, Stockholm, Sweden); thereafter, the Leksell GammaPlan software (Elekta Instruments) was used. GKRS was performed with the Leksell Gamma Knife Model B (Elekta AB, Stockholm, Sweden) until October 2011, and with the Leksell Gam-ma Knife Perfexion (Elekta AB) thereafter. The target volume corresponded to the AVM nidus, defined as the site of the arteriovenous shunt. All patients received 20 mg of intravenous methylprednisolone before and after GKRS, followed by 4 mg of oral methylprednisolone every 12 hours for 5 days, administered on a tapering schedule.
Flow assessment
All flow analyses were performed using digital subtraction angiography (DSA) acquired on the day of GKRS. Anteroposterior and lateral angiograms were obtained at a rate of two frames per second. AVM flow was evaluated according to the method described by Alzate et al. [1], which quantifies the number of frames from nidus opacification to visualization of the draining vein. High flow was defined as a nidus-to-draining vein transit time of less than one frame. Moderate flow was defined as a transit time of at least one frame but less than two frames. Low flow was defined as a transit time of two or more frames.
Follow-up assessment
Patients were advised to undergo clinical and imaging follow-up 6 months after GKRS, with subsequent imaging performed at 6–12-month intervals. Follow-up studies included T1-weighted, contrast-enhanced T1-weighted, and time-offlight (TOF) MR imaging to monitor AVM evolution. Complete obliteration on MR imaging was defined as the absence of nidus enhancement on contrast-enhanced T1-weighted images and the absence of venous drainage on TOF sequences. If neurologic symptoms occurred after repeat GKRS, patients underwent computed tomography or MR imaging to evaluate the presence of hemorrhage or adverse radiation effects (AREs). AREs were defined on MR imaging as areas of high signal intensity with mass effect around the AVM on T2-weighted images [6].
Statistical analysis
Kaplan-Meier survival analysis was used to estimate survival curves and calculate complete obliteration rates. Univariate Cox proportional hazards regression was performed to identify factors associated with complete obliteration. Variables with p<0.1 in the univariate analysis were subsequently included in a multivariate Cox proportional hazards model to determine independent predictors of complete obliteration. In the multivariate analysis, a p-value <0.05 was considered statistically significant. The log-rank test was applied to compare actuarial complete obliteration rates among the three AVM flow categories. The Mann-Whitney U test was used for continuous variables, and Fisher’s exact test for categorical variables, in the univariate analysis of potential factors associated with AREs. Variables with p-values <0.1 were entered into multivariate analysis. For multivariate analysis, binary logistic regression was performed to identify potential factors associated with AREs. A p-value <0.05 was considered statistically significant. All statistical analyses were conducted using SPSS software, version 22.0 (IBM Corp., Armonk, NY, USA).
RESULTS
Demographic and clinical characteristics of the 31 patients included in this study are presented in Table 1. Sixteen patients (51.6%) were female, and 15 (48.4%) were male. The median age at the time of GKRS was 25 years (interquartile range [IQR], 19–40). Thirty patients (96.8%) had experienced hemorrhage prior to their initial microsurgical resection, and eight patients (25.8%) underwent embolization after microsurgical resection but before GKRS. AVM locations were as follows : frontal lobe in seven patients (22.6%), parietal lobe in six (19.4%), temporal lobe in one (3.2%), occipital lobe in nine (29.0%), corpus callosum in one (3.2%), cerebellum in four (12.9%), and brainstem in three (9.7%). No AVMs were located in the thalamus or basal ganglia. Regarding GKRS dosimetric parameters, the median target volume was 0.8 cm3 (IQR, 0.5–2.2), the median margin dose was 24 Gy (IQR, 22.5–25.0), and the median maximum dose was 38.46 Gy (IQR, 36.7–40.0). Venous varices were identified in 13 patients (41.9%), and associated aneurysms were present in six (19.4%). Multiple feeding arteries were observed in 19 patients (61.3%; median, 2; IQR, 0–2), whereas multiple draining veins were seen in 13 patients (41.9%; median, 1; IQR, 1–2). The SM grade distribution was as follows : grade I in seven patients (22.6%), grade II in 10 (32.3%), grade III in nine (29.0%), grade IV in five (16.1%), and grade V in none. Based on DSA, AVM flow was categorized as high in eight patients (25.8%), moderate in 16 (51.6%), and low in seven (22.6%).
Table 1.
Demographic data of clinical characteristics in 31 patients
| Characteristic | Value |
|---|---|
| Gender | |
| Female | 16 (51.6) |
| Male | 15 (48.4) |
| Age at repeat treatment (years) | 25 (19–40) |
| Previous hemorrhage | 30 (96.8) |
| Previous embolization | 8 (25.8) |
| AVM location | |
| Frontal | 7 (22.6) |
| Parietal | 6 (19.4) |
| Temporal | 1 (3.2) |
| Occipital | 9 (29.0) |
| Corpus callosum | 1 (3.2) |
| Thalamus | 0 (0.0) |
| Basal ganglia | 0 (0.0) |
| Cerebellum | 4 (12.9) |
| Brainstem | 3 (9.7) |
| GKRS parameters | |
| Target volume (cm3) | 0.8 (0.5–2.2) |
| Margin dose (Gy) | 24 (22.5–25.0) |
| Maximum dose (Gy) | 38.46 (36.7–40.0) |
| Venous varix | 13 (41.9) |
| Associated aneurysm | 6 (19.4) |
| Deep location | 11 (35.5) |
| Number of feeding arteries | 2 (0–2) |
| Multiple feeding arteries | 19 (61.3) |
| Number of draining veins | 1 (1–2) |
| Multiple draining veins | 13 (41.9) |
| SM grade | |
| I | 7 (22.6) |
| II | 10 (32.2) |
| III | 9 (29.0) |
| IV | 5 (16.1) |
| V | 0 (0.0) |
| AVM flow after surgical resection | |
| High | 8 (25.8) |
| Moderate | 16 (51.6) |
| Low | 7 (22.6) |
Values are presented as median (interquartile range) or number (%). AVM : arteriovenous malformation, GKRS : Gamma Knife radiosurgery, SM : Spetzler-Martin
The median clinical follow-up period after GKRS was 44 months (IQR, 35–105). Twenty-four patients (77.4%) achieved complete obliteration at their last follow-up. Based on KaplanMeier analysis, the actuarial obliteration rates were 29.0%, 56.7%, and 69.7% at 3, 5, and 10 years, respectively (Fig. 2). The mean time to complete obliteration was 78.3 months (95% confidence interval [CI], 57.2–99.6). By AVM flow category, three (37.5%) high-flow, 14 (87.5%) moderate-flow, and all low-flow AVMs achieved complete obliteration. The mean times to complete obliteration were 131.7 months (95% CI, 92.9–170.6) for high-flow AVMs, 73.0 months (95% CI, 44.2–101.8) for moderate-flow AVMs, and 36.7 months (95% CI, 22.1–51.3) for lowflow AVMs (p=0.004, log-rank test; Fig. 3). The corresponding actuarial obliteration rates at 3, 5, and 10 years were 0%, 14.7%, and 31.4% for high-flow AVMs; 37.5%, 63.5%, and 72.7% for moderate-flow AVMs; and 42.9%, 85.7%, and 100% for low-flow AVMs, respectively. Among patients who underwent embolization prior to GKRS, seven (87.5%) achieved complete obliteration, whereas 17 (73.9%) of those without prior embolization achieved complete obliteration. The mean time to complete obliteration was 35.6 months (95% CI, 15.7–55.5) in the embolization group and 91.1 months (95% CI, 66.9–115.4) in the nonembolization group (p=0.007, log-rank test). The actuarial obliteration rates were 67.5% and 100% at 3 and 5 years, respectively, for patients who had undergone embolization, compared with 17.4%, 45.9%, and 56.7% at 3, 5, and 10 years, respectively, for those who had not.
Fig. 2.

Kaplan-Meier estimates of complete obliteration after GKRS in 31 patients with residual AVMs following nidus removal surgery. GKRS : Gamma Knife radiosurgery, AVM : arteriovenous malformation.
Fig. 3.

Kaplan-Meier estimates of complete obliteration stratified by AVM flow (low-, moderate-, and high-flow) in patients with residual AVMs after nidus removal surgery. GKRS : Gamma Knife radiosurgery, AVM : arteriovenous malformation.
Factors associated with complete obliteration based on univariate and multivariate cox regression analyses are presented in Table 2. In the univariate analysis, target volume (hazard ratio [HR], 0.453; 95% CI, 0.247–0.828; p=0.01) and prior embolization (HR, 3.362; 95% CI, 1.295–8.728; p=0.013) were significantly associated with complete obliteration. Using high-flow AVMs as the reference, both moderate-flow (HR, 4.418; 95% CI, 1.24–15.74; p=0.021) and low-flow AVMs (HR, 8.664; 95% CI, 2.048–36.654; p=0.003) showed significant associations with complete obliteration. When flow was analyzed as a continuous variable, it remained significantly associated with complete obliteration (HR, 2.729; 95% CI, 1.446–5.152; p=0.002). In the multivariate analysis, prior embolization (HR, 3.659; 95% CI, 1.244–10.765; p=0.018), moderate-flow AVMs (HR, 4.820; 95% CI, 1.125–20.641; p=0.034), and low-flow AVMs (HR, 4.343; 95% CI, 1.144–16.481; p=0.031), using high-flow AVMs as the reference, were independently associated with complete obliteration. In an alternative multivariate model, flow analyzed as a continuous variable also remained an independent predictor of complete obliteration (HR, 2.012; 95% CI, 1.067–3.794; p=0.031).
Table 2.
Univariate and multivariate cox proportional hazards regression analyses on factors associated with complete obliteration
| Variable | Univariate |
Multivariate |
||||
|---|---|---|---|---|---|---|
| p-value | HR | 95% CI | p-value | HR | 95% CI | |
| Age | 0.877 | 1.002 | 0.973–1.033 | - | - | - |
| Female | 0.500 | 0.756 | 0.335–1.705 | - | - | - |
| Target volume | 0.010* | 0.453 | 0.247–0.828 | 0.085 | 0.625 | 0.366–1.067 |
| Margin dose | 0.332 | 1.148 | 0.869–1.518 | - | - | - |
| Maximum dose | 0.491 | 1.035 | 0.938–1.142 | - | - | - |
| Prior embolization | 0.013* | 3.362 | 1.295–8.728 | 0.018 | 3.659 | 1.244–10.765 |
| Previous hemorrhage | 0.213 | 0.264 | 0.033–2.150 | - | - | - |
| SM grade | 0.445 | 0.849 | 0.559–1.292 | - | - | - |
| Deep location | 0.403 | 0.694 | 0.295–1.633 | - | - | - |
| Number of feeding arteries | 0.511 | 0.838 | 0.496–1.417 | - | - | - |
| Number of draining veins | 0.802 | 0.947 | 0.619–1.45 | - | - | - |
| Venous varix | 0.314 | 0.653 | 0.285–1.496 | - | - | - |
| Associated aneurysm | 0.167 | 0.463 | 0.155–1.381 | |||
| Slower flow | 0.002* | 2.729 | 1.446–5.152 | 0.031 | 2.012 | 1.067–3.794 |
| High flow | 0.004* | 0.030 | ||||
| Moderate flow | 0.021* | 4.418 | 1.24–15.74 | 0.034 | 4.820 | 1.125–20.641 |
| Low flow | 0.003* | 8.664 | 2.048–36.654 | 0.031 | 4.343 | 1.144–16.481 |
Indicates statistical significance at p<0.10.
HR : hazard ratio, CI : confidence interval, SM : Spetzler-Martin
Two major adverse outcomes, post-GKRS hemorrhage and AREs, were assessed. No patients experienced post-GKRS hemorrhage, whereas four patients (12.9%) developed AREs after GKRS. The initial clinical presentation of AREs was dizziness in one patient, while the remaining three patients were asymptomatic. No patient developed permanent symptoms. In univariate analysis, margin dose (p=0.039) and SM grade (p=0.078) were significantly associated with AREs after GKRS (Supplementary Table 1). However, in multivariate analysis, neither variable remained independently associated with AREs after GKRS.
Representative case of low-flow AVM
A 20-year-old male patient presented with a generalized tonic-clonic seizure and was admitted to our hospital (Fig. 4). The patient had no relevant medical history. MR imaging, computed tomography, and DSA revealed an AVM located in the occipital region, classified as SM grade III. Microsurgical resection was performed, but postoperative DSA demonstrated a residual nidus. Consequently, GKRS was planned as adjuvant treatment. In the vertebral artery angiogram obtained for GKRS planning 2 months after surgery, the residual AVM demonstrated low-flow characteristics, as the draining vein appeared five frames after visualization of the nidus. Follow-up TOF MR imaging performed 12 months after GKRS confirmed complete obliteration of the residual AVM.
Fig. 4.

Representative case of a low-flow residual AVM. A 20-year-old male patient underwent GKRS after nidus removal surgery and achieved complete obliteration. A : Vertebral artery angiogram obtained before nidus removal surgery, showing the occipital AVM (white arrow). B : Postoperative CT image after nidus removal surgery. C and D : Vertebral artery angiogram obtained 2 months after nidus removal surgery, showing the residual AVM (C, white arrow). The AVM was classified as low-flow because the draining vein appeared five frames after nidus visualization on DSA (D, white arrow). E : TOF MR image obtained during GKRS planning, showing the residual nidus (white arrow). F : TOF MR image obtained 12 months after GKRS, showing complete disappearance of the nidus (white arrow). AVM : arteriovenous malformation, GKRS : Gamma Knife radiosurgery, CT : computed tomography, DSA : digital subtraction angiography, TOF : time-of-flight, MR : magnetic resonance.
Representative case of high-flow AVM
A 17-year-old male patient presented with mental deterioration and was admitted to our hospital (Fig. 5). The patient had no relevant medical history. MR imaging, computed tomography, and DSA revealed an AVM located in the corpus callosum, classified as SM grade IV. Microsurgical resection was performed; however, postoperative computed tomography demonstrated a residual nidus in the posterior portion of the AVM. Consequently, GKRS was planned as adjuvant treatment. In the internal carotid artery angiogram obtained for GKRS planning 4 months after surgery, the residual AVM demonstrated high-flow characteristics, as the draining vein appeared in the same frame as the nidus visualization. Follow-up TOF MR imaging performed 8 years after GKRS revealed a persistent resid-ual nidus, which, although markedly reduced, remained detectable on MR imaging 16 years post-treatment.
Fig. 5.

Representative case of a high-flow residual AVM. A 17-year-old male patient underwent GKRS after nidus removal surgery but did not achieve complete obliteration after 16 years. A : Right internal carotid artery angiogram obtained before nidus removal surgery, showing the corpus callosal AVM (white arrow). B : Postoperative CT image showing a residual nidus (white arrow). C : Right internal carotid artery angiogram obtained 4 months after nidus removal surgery, showing the residual AVM (white arrow). The AVM was classified as high-flow because the draining veins appeared in the same frame as the nidus visualization on DSA (white arrowheads). D : TOF MR image obtained during GKRS planning, showing the residual nidus (white arrow). E : TOF MR image obtained 8 years after GKRS, showing a persistent nidus (white arrow). F : TOF MR image obtained 16 years after GKRS showing a residual but markedly reduced nidus (white arrow). AVM : arteriovenous malformation, GKRS : Gamma Knife radiosurgery, CT : computed tomography, DSA : digital subtraction angiography, TOF : time-of-flight, MR : magnetic resonance.
DISCUSSION
GKRS as a secondary treatment for AVMs has not been as extensively studied as GKRS used as a primary treatment, and reported obliteration rates vary depending on the initial intervention. In patients who undergo embolization prior to GKRS, no consensus exists on whether embolization has a positive or negative effect on complete obliteration rates, as studies have reported conflicting results [3,4,7,9,10]. For patients who undergo repeat GKRS for residual AVMs after a prior GKRS, the complete obliteration rates reportedly range 55–70% [12,18,26,30]. Studies specifically evaluating outcomes after partial microsurgical resection of AVMs remain limited. Ding et al. [5] reported a 72.7% obliteration rate in a retrospective cohort of 88 patients, whereas Kawashima et al. [16] observed a 97% 5-year obliteration rate among 60 patients. In the present study, we report our institutional experience with 31 patients who underwent GKRS following identification of a residual nidus after microsurgical resection. Our aim was to evaluate the complete obliteration rate and identify predictors associated with higher obliteration probability in this subgroup. Overall, the outcomes suggest that partially resected AVMs treated with GKRS achieve favorable obliteration rates, with prior embolization and lower AVM flow identified as independent predictors of complete obliteration.
Because few studies have reported outcomes of GKRS following partial AVM resection, comparison with our findings is limited. Ding et al. [5] compared Kaplan-Meier curves of partially resected and unresected AVMs and found no significant difference in actuarial obliteration rates. This suggests that the obliteration rates of partially resected AVMs may be comparable to those of AVMs treated with GKRS as a primary modality. Previous studies have reported complete obliteration rates ranging from 60% to 80% for AVMs treated with a single GKRS session [21,24,27]. Accordingly, the complete obliteration rates observed in our cohort of partially resected AVMs are consistent with these prior results. Furthermore, with respect to adverse outcomes, no post-GKRS hemorrhage was observed, whereas four patients developed AREs. Although the small number of events may be attributable to the limited sample size, the observed incidence was nevertheless consistent with previous reports [11,27]. These findings support the notion that GKRS may represent a reasonable and effective treatment option for residual AVMs following partial surgical resection.
To our knowledge, this is the first study to assess flow characteristics in partially resected AVMs. Several previous studies have investigated AVM flow velocity prior to treatment and its correlation with post-treatment outcomes. Alzate et al. [1], Taeshineetanakul et al. [29], and Panni et al. [20] evaluated AVM flow before GKRS by quantifying the number of frames between nidus and draining vein opacification on DSA, categorizing AVMs into low-, moderate-, and high-flow groups. These studies suggested that lower AVM flow is associated with higher rates of complete obliteration, implying that hemodynamic factors play a crucial role in determining treatment response after GKRS. Microsurgical resection aims to occlude feeding arteries and draining veins and to remove the nidus [25]. As a result, the hemodynamic properties of residual AVMs are expected to differ substantially from those of the original lesion [15]. We therefore hypothesized that post-resection AVM flow may be associated with complete obliteration. To evaluate this, we applied the frame-counting method described by Alzate et al. [1], which has been validated against computer-based quantitative analyses comparing the peak time of the nidus and draining vein. In our study, slower flow in partially resected AVMs was independently associated with complete obliteration. Notably, all residual AVMs classified as low-flow achieved complete obliteration. These findings suggest that residual AVMs with slower flow following microsurgical resection may have a greater likelihood of complete obliteration, whereas those that remain high-flow may have less favorable outcomes.
In our cohort, prior embolization was independently associated with complete obliteration. For AVMs treated with a single session of GKRS, the effect of prior embolization on obliteration outcomes remains controversial. Some studies have reported that embolization before GKRS negatively influences obliteration rates, whereas others have found no significant effect [3,4,10,23]. In the present study, seven of eight patients (87.5%) with a history of embolization achieved complete obliteration. The single case that did not achieve obliteration was classified as high-flow, whereas all seven cases with complete obliteration exhibited moderate or low flow. This finding suggests that among patients with prior embolization, those with moderateor low-flow AVMs may still have a high probability of achieving complete obliteration. However, given the small number of patients with prior embolization in this cohort, further studies with larger sample sizes are needed to clarify the relationship between AVM flow dynamics and obliteration outcomes in this subgroup.
This study has some limitations. First, its retrospective design and small sample size may have introduced selection bias. Second, complete obliteration was assessed using MR imaging, although DSA remains the gold standard for confirmation. However, previous reports have shown that the accuracy of MR imaging in confirming complete obliteration can reach up to 90%, suggesting that it remains an acceptable modality for this purpose [19]. Third, the reported obliteration rates may have been influenced by the imaging interval, as MR imaging was performed every 6–12 months. This delay may have separated the actual time of obliteration from the imaging confirmation. Fourth, we did not compare AVM flow characteristics on DSA before and after microsurgical resection. Future studies should investigate the degree of flow alteration following partial resection and determine whether the extent of flow reduction influences subsequent obliteration rates.
CONCLUSION
This study suggests favorable outcomes of GKRS for partially resected AVMs. Slower-flow residual AVMs and a history of prior embolization independently correlated with complete obliteration.
Footnotes
Conflicts of interest
No potential conflict of interest relevant to this article was reported.
Informed consent
This type of study does not require informed consent.
Author contributions
Conceptualization : JIP, JR, SKC; Data curation : JIP; Formal analysis : JIP; Methodology : JIP, SKC; Project administration : JIP; Visualization : JIP; Writing - original draft : JIP; Writing - review & editing : JIP, JR, SKC
Data sharing
All data generated or analyzed during this study are included in this article. Further enquiries can be directed to the corresponding author.
Preprint
None
Supplementary materials
The online-only data supplement is available with this article at https://doi.org/10.3340/jkns.2026.0004.
Univariate and multivariate analysis for ARE after GKRS
References
- 1.Alzate JD, Berger A, Bernstein K, Mullen R, Qu T, Silverman JS, et al. Preoperative flow analysis of arteriovenous malformations and obliteration response after stereotactic radiosurgery. J Neurosurg. 2022;138:944–954. doi: 10.3171/2022.7.JNS221008. [DOI] [PubMed] [Google Scholar]
- 2.Aoun SG, Bendok BR, Batjer HH. Acute management of ruptured arteriovenous malformations and dural arteriovenous fistulas. Neurosurg Clin N Am. 2012;23:87–103. doi: 10.1016/j.nec.2011.09.013. [DOI] [PubMed] [Google Scholar]
- 3.Chen CJ, Ding D, Lee CC, Kearns KN, Pomeraniec IJ, Cifarelli CP, et al. Stereotactic radiosurgery with versus without embolization for brain arteriovenous malformations. Neurosurgery. 2021;88:313–321. doi: 10.1093/neuros/nyaa418. [DOI] [PubMed] [Google Scholar]
- 4.De Leacy R, Ansari SA, Schirmer CM, Cooke DL, Prestigiacomo CJ, Bulsara KR, et al. Endovascular treatment in the multimodality management of brain arteriovenous malformations: report of the Society of NeuroInterventional Surgery Standards and Guidelines Committee. J Neurointerv Surg. 2022;14:1118–1124. doi: 10.1136/neurintsurg-2021-018632. [DOI] [PubMed] [Google Scholar]
- 5.Ding D, Xu Z, Shih HH, Starke RM, Yen CP, Sheehan JP. Stereotactic radiosurgery for partially resected cerebral arteriovenous malformations. World Neurosurg. 2016;85:263–272. doi: 10.1016/j.wneu.2015.10.001. [DOI] [PubMed] [Google Scholar]
- 6.Ding D, Yen CP, Xu Z, Starke RM, Sheehan JP. Radiosurgery for patients with unruptured intracranial arteriovenous malformations. J Neurosurg. 2013;118:958–966. doi: 10.3171/2013.2.JNS121239. [DOI] [PubMed] [Google Scholar]
- 7.Erickson N, Mooney J, Salehani A, Thomas E, Ilyas A, Rahm S, et al. Predictive factors for arteriovenous malformation obliteration after stereotactic radiosurgery: a single-center study. World Neurosurg. 2022;160:e529–e536. doi: 10.1016/j.wneu.2022.01.060. [DOI] [PubMed] [Google Scholar]
- 8.Gross BA, Duckworth EA, Getch CC, Bendok BR, Batjer HH. Challenging traditional beliefs: microsurgery for arteriovenous malformations of the basal ganglia and thalamus. Neurosurgery. 2008;63:393–410. doi: 10.1227/01.NEU.0000316424.47673.03. discussion 410-411. [DOI] [PubMed] [Google Scholar]
- 9.Jiang X, Zhao Z, Zhang Y, Wang Y, Lai L. Preradiosurgery embolization in reducing the postoperative hemorrhage rate for patients with cerebral arteriovenous malformations: a systematic review and meta-analysis. Neurosurg Rev. 2021;44:3197–3207. doi: 10.1007/s10143-021-01530-4. [DOI] [PubMed] [Google Scholar]
- 10.Jiang Z, Zhang X, Wan X, Wei M, Liu Y, Ding C, et al. Efficacy and safety of combined endovascular embolization and stereotactic radiosurgery for patients with intracranial arteriovenous malformations: a systematic review and meta-analysis. Biomed Res Int. 2021;2021:6686167. doi: 10.1155/2021/6686167. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Kano H, Flickinger JC, Tonetti D, Hsu A, Yang HC, Flannery TJ, et al. Estimating the risks of adverse radiation effects after gamma knife radiosurgery for arteriovenous malformations. Stroke. 2017;48:84–90. doi: 10.1161/STROKEAHA.116.014825. [DOI] [PubMed] [Google Scholar]
- 12.Kano H, Kondziolka D, Flickinger JC, Yang HC, Flannery TJ, Awan NR, et al. Stereotactic radiosurgery for arteriovenous malformations, part 3: outcome predictors and risks after repeat radiosurgery. J Neurosurg. 2012;116:21–32. doi: 10.3171/2011.9.JNS101741. [DOI] [PubMed] [Google Scholar]
- 13.Karlsson B, Jokura H, Yang HC, Yamamoto M, Martinez-Alvarez R, Kawagishi J, et al. Risk for hemorrhage the first 2 years after Gamma Knife surgery for arteriovenous malformations: an update. Neurosurgery. 2022;91:920–927. doi: 10.1227/neu.0000000000002130. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Karlsson B, Lax I, Söderman M. Risk for hemorrhage during the 2-year latency period following gamma knife radiosurgery for arteriovenous malformations. Int J Radiat Oncol Biol Phys. 2001;49:1045–1051. doi: 10.1016/s0360-3016(00)01432-2. [DOI] [PubMed] [Google Scholar]
- 15.Kato Y, Yamada Y, Sadato A, Nouri M, Cherian I, Tanaka T, et al. Intraoperative anatomical and hemodynamic analysis of intracerebral arteriovenous malformations by semi-quantitative color-coded indocyanine green videoangiography. Asian J Neurosurg. 2017;12:638–643. doi: 10.4103/ajns.AJNS_62_14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Kawashima M, Hasegawa H, Shin M, Shinya Y, Ishikawa O, Koizumi S, et al. Outcomes of stereotactic radiosurgery for hemorrhagic arteriovenous malformations with or without prior resection or embolization. J Neurosurg. 2020;135:733–741. doi: 10.3171/2020.7.JNS201502. [DOI] [PubMed] [Google Scholar]
- 17.Lepski G, Honegger J, Liebsch M, Sória MG, Narischat P, Ramina KF, et al. Safe resection of arteriovenous malformations in eloquent motor areas aided by functional imaging and intraoperative monitoring. Neurosurgery. 2012;70(2 Suppl Operative):276–288. doi: 10.1227/NEU.0b013e318237aac5. discussion 288-289. [DOI] [PubMed] [Google Scholar]
- 18.Maroufi SF, Habibi MA, Mirjani MS, Molla A, Pabarja N, Mehmandoost M, et al. Repeat single-session stereotactic radiosurgery for arteriovenous malformation: a systematic review and meta-analysis. Neurosurg Rev. 2024;47:203. doi: 10.1007/s10143-024-02438-5. [DOI] [PubMed] [Google Scholar]
- 19.OʼConnor TE, Friedman WA. Magnetic resonance imaging assessment of cerebral arteriovenous malformation obliteration after stereotactic radiosurgery. Neurosurgery. 2013;73:761–766. doi: 10.1227/NEU.0000000000000086. [DOI] [PubMed] [Google Scholar]
- 20.Panni P, Gallotti AL, Gigliotti CR, Shehata AB, Albano L, da Passano CF, et al. Impact of flow and angioarchitecture on brain arteriovenous malformation outcome after gamma knife radiosurgery: the role of hemodynamics and morphology in obliteration. Acta Neurochir (Wien) 2020;162:1749–1757. doi: 10.1007/s00701-020-04351-4. [DOI] [PubMed] [Google Scholar]
- 21.Pollock BE, Gorman DA, Coffey RJ. Patient outcomes after arteriovenous malformation radiosurgical management: results based on a 5- to 14-year follow-up study. Neurosurgery. 2003;52:1291–1296. doi: 10.1227/01.neu.0000064800.26214.fe. discussion 1296-1297. [DOI] [PubMed] [Google Scholar]
- 22.Potts MB, Jahangiri A, Jen M, Sneed PK, McDermott MW, Gupta N, et al. Deep arteriovenous malformations in the basal ganglia, thalamus, and insula: multimodality management, patient selection, and results. World Neurosurg. 2014;82:386–394. doi: 10.1016/j.wneu.2014.03.033. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Russell D, Peck T, Ding D, Chen CJ, Taylor DG, Starke RM, et al. Stereotactic radiosurgery alone or combined with embolization for brain arteriovenous malformations: a systematic review and meta-analysis. J Neurosurg. 2018;128:1338–1348. doi: 10.3171/2016.11.JNS162382. [DOI] [PubMed] [Google Scholar]
- 24.Sasagasako T, Mori H, Hattori EY, Ikedo T, Hamano E, Shimonaga K, et al. Radiation-induced changes associated with obliteration of brain AVMs after repeat radiosurgery. AJNR Am J Neuroradiol. 2023;44:143–149. doi: 10.3174/ajnr.A7772. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Smrcka M, Navratil O, Hovorka E, Duris K. The efficacy of surgical treatment of cerebral arteriovenous malformations in a single academic institution: a case series. Croat Med J. 2021;62:353–359. doi: 10.3325/cmj.2021.62.353. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Stahl JM, Chi YY, Friedman WA. Repeat radiosurgery for intracranial arteriovenous malformations. Neurosurgery. 2012;70:150–154. doi: 10.1227/NEU.0b013e31822c5740. discussion 154. [DOI] [PubMed] [Google Scholar]
- 27.Starke RM, Kano H, Ding D, Lee JY, Mathieu D, Whitesell J, et al. Stereotactic radiosurgery for cerebral arteriovenous malformations: evaluation of longterm outcomes in a multicenter cohort. J Neurosurg. 2017;126:36–44. doi: 10.3171/2015.9.JNS151311. [DOI] [PubMed] [Google Scholar]
- 28.Sussman ES, Gummidipundi SE, Pendharkar AV, Church EW, Ho AL, Han SS, et al. Staged surgical resection of brain arteriovenous malformations. World Neurosurg. 2021;146:e925–e930. doi: 10.1016/j.wneu.2020.11.036. [DOI] [PubMed] [Google Scholar]
- 29.Taeshineetanakul P, Krings T, Geibprasert S, Menezes R, Agid R, Terbrugge KG, et al. Angioarchitecture determines obliteration rate after radiosurgery in brain arteriovenous malformations. Neurosurgery. 2012;71:1071–1078. doi: 10.1227/NEU.0b013e31826f79ec. discussion 1079. [DOI] [PubMed] [Google Scholar]
- 30.Yen CP, Jain S, Haq IU, Jagannathan J, Schlesinger D, Sheehan J, et al. Repeat γ knife surgery for incompletely obliterated cerebral arteriovenous malformations. Neurosurgery. 2010;67:55–64. doi: 10.1227/01.neu.0000370204.68711.ac. discussion 64. [DOI] [PubMed] [Google Scholar]
- 31.Zhang Y, Chen Y, Han H, Ma L, Li R, Li Z, et al. Timing of microsurgical resection for ruptured brain arteriovenous malformations: a propensity scorematched analysis using prospective single-center registry data. J Neurosurg. 2023;140:164–171. doi: 10.3171/2023.5.JNS222666. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Univariate and multivariate analysis for ARE after GKRS
