Rupture of brain arteriovenous malformations (AVMs) causes the majority of spontaneous intracerebral hemorrhages in the postinfancy pediatric population, and its occurrence can cause considerable morbidity and mortality.1 A lack of consensus regarding the management of pediatric AVMs is reflected by the relatively sparse literature on this topic.2 Resection affords definitive treatment of low-grade pediatric AVMs (i.e., those that are Spetzler-Martin [SM] grades I and II) with a reasonable safety profile, but management of intermediate (SM grade III) and high-grade (SM grades IV and V) AVMs commonly incorporates adjunctive or alternate interventions, such as endovascular embolization and stereotactic radiosurgery (SRS), to ameliorate the overall therapeutic risk.3,4 Prior studies suggest that AVMs in children have distinct characteristics compared with those in adults.5–8 Pediatric AVMs are more often located in deep brain regions, more likely to present with hemorrhage, and more likely to recur after obliteration.5,7,9–13 The longer life expectancy of children subjects them to a greater cumulative lifetime risk of AVM hemorrhage as compared with adults. Taken together, these factors provide the impetus for an aggressive stance on the management of pediatric AVMs.2,7,14,15 SRS has been established as an effective intervention for pediatric AVMs.5,9,10,16–23
AVM obliteration after SRS virtually eliminates the risk of hemorrhage. However, patients remain at risk for AVM hemorrhage during the latency period between treatment and obliteration, which can span a varying time interval that ranges from months to years after SRS.24,25 A previous study identified predictors of early obliteration in a large, multicenter study of AVM patients treated with SRS.26 As AVMs in children are different from those in adults, factors contributing to early obliteration of pediatric AVMs may not overlap with those in adult AVMs. Early obliteration has not been evaluated in a cohort of exclusively pediatric AVMs. The aims of this multicenter, retrospective cohort study of pediatric patients with SRS-treated AVMs were as follows: 1) to determine predictors of early obliteration and 2) to compare the outcomes in patients in whom early versus late obliteration was achieved.
Methods
Patient Identification, Ethical Approval of Study, and Informed Consent
This study follows the guidelines set forth by the STROBE statement. We obtained approval for this study from the institutional review board of each contributing institution, and patient consent was waived. We retrospectively surveyed a multicenter database of pediatric patients (age < 18 years) with AVMs who underwent SRS at 8 institutions participating in the International Radiosurgery Research Foundation (IRRF). Pediatric AVMs from the former International Gamma Knife Research Foundation (now IRRF) database (1987–2014) were updated.27 Data from additional participating institutions and pediatric AVMs treated with SRS between 2014 and 2018 were also included. Each respective institution verified and attested to data accuracy. Individual patient data from each contributing institution were then de-identified and pooled by an independent third party.
Only patients with complete AVM obliteration were included in the study cohort. AVM obliteration was defined on MRI as a lack of abnormal flow voids, on CTA as a lack of abnormal vasculature, or on catheter-based DSA as an absence of anomalous arteriovenous shunting. Neuroimaging follow-up was performed at approximately 6-month intervals for the first 2 years after SRS and then yearly thereafter. Confirmatory DSA was recommended to patients with AVM obliteration shown on follow-up MRI or CTA. Patients were then categorized into two cohorts based on the timing of AVM obliteration, as follows: 1) early obliteration, defined as AVM obliteration ≤ 24 months after SRS, or 2) late obliteration, defined as AVM obliteration > 24 months after SRS.
Baseline Data and Variables
Baseline data included patient, AVM, and SRS variables. The following data were extracted from directed chart review: 1) patient variables comprising sex and age; 2) AVM variables comprising maximum nidus diameter, nidus volume, deep venous drainage (categorized as exclusively superficial vs deep component), location (categorized as eloquent vs noneloquent), and presence of AVM-associated intranidal or perinidal arterial aneurysms; and 3) SRS variables comprising number of SRS treatments, maximum dose, margin dose, isodose line, and number of isocenters. The SM grade, Virginia Radiosurgery AVM Scale (VRAS) score, and modified radiosurgery-based AVM score (RBAS) were determined for each case.28–30 Sensorimotor, language, and visual cortices, and hypothalamus and thalamus, internal capsule, brainstem, cerebellar peduncles, and deep cerebellar nuclei were designated as eloquent locations.29 The thalamus, basal ganglia, and brainstem were designated as deep locations.30
The SRS technique for AVMs has been previously described.31 SRS was performed using the Gamma Knife, although the specific model differed by year and availability at each institution. In brief, the patient’s calvaria was affixed within a Leksell model G frame (Elekta AB) after induction of anesthesia. DSA and thin-slice (slice thickness 1–2 mm) MRI, or CTA when MRI was contraindicated, were used to delineate the angioarchitecture and spatial anatomy of the AVM nidus. A multidisciplinary team from each respective institution, comprising a neurosurgeon, radiation oncologist, and medical physicist, performed dose planning and radiosurgical delivery.
Outcomes
Post-SRS hemorrhage was defined as any AVM-related intracranial hemorrhage following the SRS procedure. Radiation-induced changes (RICs) were defined as perinidal hyperintensities on T2-weighted or FLAIR MRI sequences. Transient and permanent symptomatic RICs were defined as an RIC associated with any deterioration of neurological status with and without neurological recovery, respectively. SRS-associated cyst and tumor formation were also recorded. Clinical and neuroimaging follow-up examinations were obtained concurrently, when feasible. When in-person follow-up could not be obtained, clinical and neuroimaging data were acquired by the respective institution from referring hospitals or physicians for review. Time to obliteration was used as the follow-up for incidence.
Statistical Analyses
All statistical analyses were performed using Stata (version 15.1; StataCorp). We compared baseline characteristics between patients with early versus late AVM obliteration using Pearson’s chi-square or Fisher’s exact test for categorical variables, as appropriate, and the Student t-test or Mann-Whitney U-test for continuous variables, as appropriate. Follow-up duration was determined by calculating the time interval from initial SRS to the time of death or last follow-up. We compared the post-SRS outcomes of patients with early versus late obliteration using univariable logistic regression or Fisher’s exact test, as appropriate.
We then identified predictors of early obliteration. To overcome the potential issues of overfitting and good in-sample fit but poor out-of-sample prediction, the baseline characteristics were entered as independent variables into a logistic Lasso regression model. A 10-fold cross-validation was performed to identify a lambda that minimized the mean-square prediction error to achieve the best out-of-sample performance. Statistical significance was defined as p < 0.05, and all tests were two-tailed. Missing data were not imputed.
Results
Patient Characteristics, AVM Features, and Radiosurgery Parameters
Among the 539 pediatric patients in the IRRF database, 345 (64.0%) who achieved AVM obliteration after SRS were included in the study cohort. Of those with obliterated AVMs, 95 patients (27.5%) had early obliteration (≤ 24 months), and 250 (72.5%) had later obliteration (Fig. 1). Obliteration was confirmed by DSA in 85.6% and 91.6% of the early versus late obliteration cohorts, respectively (p = 0.137). There were no AVM recurrences after angiographic confirmation of obliteration. Table 1 compares the patient demographics, AVM characteristics, and SRS parameters of the early versus late obliteration cohorts. The mean ages in the overall cohort, early obliteration cohort, and late obliteration cohort were 12.8 ± 3.1 years, 12.6 ± 3.1 years, and 12.9 ± 3.8 years, respectively. Females comprised 47.3%, 44.2%, and 48.4% of the overall cohort, early obliteration cohort, and late obliteration cohort, respectively. The overall follow-up durations were 91.1 ± 67.6 months, 83.1 ± 65.6 months, and 94.1 ± 68.2 months for the overall cohort, the early obliteration cohort, and the late obliteration cohort, respectively. Patients in the early obliteration cohort had smaller AVMs, based on the mean maximum diameter (2.1 vs 2.4 cm; p = 0.010) and volume (2.6 vs 5.0 cm3; p < 0.001), and they had a lower mean RBAS (0.7 vs 0.9; p = 0.001). The VRAS score was not predictive of early obliteration (p = 0.200). AVMs with early obliteration were treated with a higher mean margin dose (21.6 vs 20.4 Gy; p = 0.002), a higher median isodose line (53% vs 50%; p = 0.039), and fewer isocenters (median 2 vs 4; p < 0.001), and they less frequently underwent ≥ 2 SRS treatments (6.4% vs 19.7%; p = 0.003).
Flow diagram detailing the inclusion criteria. The IRRF pediatric AVM database had a total of 539 pediatric AVMs treated with SRS. The 345 patients in whom complete obliteration was achieved were included in the overall study cohort, and they were allocated to the early (≤ 2 years, n = 95) or late (> 2 years, n = 250) obliteration cohorts. The 194 patients who did not achieve obliteration were excluded.
Comparison of patient demographics, AVM characteristics, and SRS parameters of the early versus late obliteration cohorts
| Characteristic | Overall Cohort (n = 345) | Early Obliteration Cohort (n = 95) | Late Obliteration Cohort (n = 250) | p Value |
|---|---|---|---|---|
| Age, mean yrs (SD) | 12.8 (3.1) | 12.6 (3.1) | 12.9 (3.8) | 0.406 |
| Female, no. (%) | 163/345 (47.3) | 42/95 (44.2) | 121/250 (48.4) | 0.486 |
| Prior EBRT, no. (%) | 53/345 (15.4) | 12/95 (12.6) | 41/250 (16.4) | 0.386 |
| Prior surgery, no. (%) | 17/345 (4.9) | 4/95 (4.2) | 13/250 (5.2) | 1.000 |
| Prior embolization, no. (%) | 39/345 (11.3) | 8/95 (8.4) | 31/250 (12.4) | 0.297 |
| AVM max diameter, mean cm (SD) | 2.3 (1.2) | 2.1 (1.1) | 2.4 (1.3) | 0.010 |
| AVM vol, mean cm3 (SD) | 4.3 (6.6) | 2.6 (2.8) | 5.0 (7.3) | <0.001 |
| ≥2 SRS procedures, no. (%) | 54/338 (16) | 6/94 (6.4) | 48/196 (19.7) | 0.003 |
| SRS max dose, mean Gy (SD)* | 37.5 (7.7) | 37.7 (7.7) | 37.4 (7.7) | 0.733 |
| SRS margin dose, mean Gy (SD)* | 20.8 (3.3) | 21.6 (3.1) | 20.4 (3.3) | 0.002 |
| Isodose line, median % (IQR)* | 50 (50–60) | 53 (50–70) | 50 (50–58) | 0.039 |
| Isocenters, median (IQR)* | 3 (2–9) | 2 (1–5) | 4 (2–9) | <0.001 |
| Eloquent location, no. (%)† | 82/345 (23.8) | 18/95 (19) | 64/250 (25.6) | 0.195 |
| Deep location, no. (%)‡ | 114/341 (33.4) | 36/93 (38.7) | 78/248 (31.5) | 0.206 |
| Arterial aneurysm, no. (%) | 17/345 (4.9) | 5/95 (5.3) | 12/250 (4.8) | 0.788 |
| Deep venous drainage, no. (%) | 222/345 (64.4) | 65/95 (68.4) | 157/250 (62.8) | 0.330 |
| SM grade, no. (%) | 0.055 | |||
| I | 33/345 (9.6) | 7/95 (7.4) | 26/250 (10.4) | |
| II | 114/345 (33) | 30/95 (31.6) | 84/250 (33.6) | |
| III | 151/345 (43.8) | 51/95 (53.7) | 100/250 (40) | |
| IV | 41/345 (11.9) | 5/95 (5.3) | 36/250 (14.4) | |
| V | 6/345 (1.7) | 2/95 (2.1) | 4/250 (1.6) | |
| VRAS score, no. (%) | 0.200 | |||
| 0 | 9/345 (2.6) | 3/95 (3.2) | 6/250 (2.4) | |
| 1 | 51/345 (14.8) | 13/95 (13.7) | 38/250 (15.2) | |
| 2 | 127/345 (36.8) | 42/95 (44.2) | 85/250 (34) | |
| 3 | 94/345 (27.3) | 18/95 (19) | 76/250 (30.4) | |
| 4 | 64/345 (18.6) | 19/95 (20) | 45/250 (18) | |
| RBAS, mean (SD) | 0.9 (0.7) | 0.7 (0.4) | 0.9 (0.8) | 0.001 |
| FU, mos (SD) | 91.1 (67.6) | 83.1 (65.6) | 94.1 (68.2) | 0.168 |
EBRT = external-beam radiation therapy; FU = follow-up.
Patients without AVM obliteration and those without timing of AVM obliteration were excluded. Early obliteration was defined as AVM obliteration ≤ 24 months after SRS, and late obliteration was defined as AVM obliteration > 24 months after SRS. Boldface type indicates statistical significance.
* Initial SRS parameter.
Sensorimotor, language, and visual cortex, and hypothalamus and thalamus, internal capsule, brainstem, cerebellar peduncles, and deep cerebellar nuclei.
Thalamus, basal ganglia, and brainstem.
Post-SRS Outcomes of the Early Versus Late Obliteration Cohorts
Table 2 compares the outcomes of the early versus late obliteration cohorts. The post-SRS outcomes of the two cohorts were not significantly different. The crude rate of post-SRS hemorrhage was 50% lower in the early (3.2%) versus late (6.4%) obliteration cohort, but this difference was not significant (p = 0.248). Comparison of early versus late obliteration cohorts showed similar rates of radiological RICs (34.7% vs 42.8%; p = 0.174), symptomatic RICs (10.5% vs 9.6%; p = 0.797), permanent RICs (7.4% vs 6.8%; p = 0.853), cyst formation (1.1% vs 2.8%; p = 0.688), and tumor formation (1.1% vs 0.4%; p = 0.472). The incidences of post-SRS hemorrhage in the early versus late obliteration cohorts were 2.16 versus 1.45 hemorrhages per 100 patient-years, respectively (p = 0.258). There were 6 postobliteration hemorrhages (1.7%) that occurred at a median interval of 3.5 years after obliteration. Although these hemorrhages could be attributed to recurrent AVMs, parenchymal/cyst wall hemorrhage, or angiographically occult micronidi, their presence could not be verified based on available data. There were no deaths reported in either cohort.
Comparison of outcomes in the early versus late obliteration cohorts
| Outcome | Early Obliteration Cohort (n = 95) | Late Obliteration Cohort (n = 250) | OR (95% CI) | p Value |
|---|---|---|---|---|
| Post-SRS hemorrhage | 3/95 (3.2) | 16/250 (6.4) | 0.477 (0.136–1.675) | 0.248 |
| RIC | 33/95 (34.7) | 107/250 (42.8) | 0.711 (0.435–1.162) | 0.174 |
| Symptomatic | 10/95 (10.5) | 24/250 (9.6) | 1.108 (0.508–2.414) | 0.797 |
| Permanent | 7/95 (7.4) | 17/250 (6.8) | 1.090 (0.437–2.719) | 0.853 |
| Cyst | 1/93 (1.1) | 7/248 (2.8) | — | 0.688* |
| Symptomatic | 0/93 (0) | 2/248 (0.8) | — | 1.000* |
| Required intervention | 0/93 (0) | 2/248 (0.8) | — | 1.000* |
| Tumor | 1/93 (1.1) | 1/248 (0.4) | — | 0.472* |
Cohort values are presented as the number (%) of patients.
Fisher’s exact test.
Predictors of Early AVM Obliteration
Table 3 summarizes the independent predictors of early obliteration. Female sex (Lasso coefficient = 0.170), a single SRS procedure (Lasso coefficient = 0.735), higher margin dose (Lasso coefficient = 0.088), higher isodose line (Lasso coefficient = 0.020), and deep AVM location (Lasso coefficient = 0.338) were positive predictors of early obliteration. Larger AVM volume was a negative predictor of early obliteration (Lasso coefficient = −0.036).
Independent predictors of early AVM obliteration in logistic Lasso regression model
| Predictor | Logistic Lasso Coefficient |
|---|---|
| Female sex | 0.170 |
| AVM vol | −0.036 |
| Single SRS procedure | 0.735 |
| Margin dose | 0.088 |
| Isodose line | 0.020 |
| Deep AVM location | 0.338 |
Discussion
Complete obliteration of an AVM nidus is the primary goal of any nonpalliative AVM intervention, as this radiological endpoint virtually eliminates the risk of hemorrhage.32 SRS is one of the definitive treatments for AVMs, but it induces nidal obliteration in a delayed fashion.33 A subset of patients respond early to SRS, with obliteration occurring in the first 4–18 months, whereas late responders do not achieve obliteration for up to 5 years after SRS.26,34 Since the risk of hemorrhage after SRS persists for as long as the AVM remains patent, early obliteration potentially reduces the morbidity and mortality of posttreatment hemorrhage by curtailing the latency period. In light of the theoretical benefit of early obliteration, we performed a multicenter, retrospective analysis of a large cohort of 345 pediatric patients with obliterated AVMs following intervention with SRS. We identified several independent predictors of early obliteration, which was defined in this study as complete AVM occlusion within 24 months of SRS. Independent predictors of early obliteration were female sex, a single SRS treatment, higher margin dose, higher isodose line, deep AVM location, and smaller AVM volume. The crude rate of post-SRS hemorrhage was 50% lower in the early (3.2%) versus late (6.4%) obliteration cohorts, but this difference was not statistically significant (p = 0.248). The other outcomes of the early versus late obliteration cohorts were similar, with respect to symptomatic RICs, cyst formation, and tumor formation.
Although our study is the first, to date, to evaluate early obliteration in a cohort composed exclusively of pediatric AVMs, some of the predictors (i.e., higher margin dose and smaller AVM volume) overlap with those identified in a previous IRRF study that included both adult and pediatric AVMs.26 It is important to note that the time interval cutoff used to define early obliteration after SRS in the prior study was ≤ 18 months, which was different than our study. Using an early obliteration cutoff of 18 months in the present study would not have provided a sufficient number of subjects in the early obliteration cohort to adequately power the statistical analysis. Therefore, we extended the cutoff to 24 months in order to provide a more robust analysis. The latency interval between SRS and obliteration can extend up to 4 years or longer.35 As such, we believe that defining early obliteration at an interval of ≤ 24 months represents a reasonable clinical benefit that halves the potential at-risk period for post-SRS hemorrhage. In congruence with the previous IRRF study, AVM location is a relevant factor in the timing of obliteration for pediatric patients, and we showed that deep location was predictive of early obliteration.26 Patients who underwent a single SRS treatment were more likely to have early obliteration, which is consistent with a prior single-center study that reported lower obliteration rates after repeat versus initial SRS for angioarchitecturally comparable AVMs.36
A higher isodose line, which is complementary to margin dose, was associated with early obliteration in the present analysis. In a single-center, retrospective cohort study of pediatric AVMs treated with SRS, Hasegawa et al. reported an association between higher margin dose and obliteration.14 The obliteration rates at 5 and 10 years were 71% and 88%, respectively, for pediatric AVMs treated with a margin dose ≥ 21.8 Gy, compared with 60% and 78%, respectively, for those treated with a margin dose < 21.8 Gy (p = 0.041).14 The early obliteration cohort in our study had a similar mean margin dose of 21.6 Gy. The importance of margin dose in pediatric AVM obliteration after SRS was underscored in a previous IRRF study, wherein AVMs treated with a margin dose ≥ 22 Gy versus < 22 Gy had obliteration rates of 81.6% versus 51.4%, respectively.37
A single-center, retrospective analysis of pediatric patients with unruptured AVMs who underwent SRS reported a higher margin dose, fewer draining veins, and lower VRAS scores as independent predictors of obliteration in the multivariable model.20 Similarly, AVMs treated with ≥ 22 Gy were more likely to be obliterated.20 Another single-center, retrospective study of SRS-treated pediatric AVMs reported lack of prior embolization, smaller nidus volume, and higher margin dose as independent predictors of obliteration.9 Smyth et al. found that pediatric AVMs treated with a radiosurgical margin dose ≥ 18 Gy had a 10-fold increase in the likelihood of obliteration.38 Potts et al. reported obliteration rates of 52% versus 16% for pediatric AVMs treated with a margin dose of 18–20 Gy versus < 18 Gy, respectively.39 In contrast, Dinca et al. did not extrapolate a relationship between dose and obliteration for SRS-treated pediatric AVMs, with obliteration rates of 86%, 78%, and 83% for margin doses of 25, 22.5, and 20 Gy, respectively (p = 0.43).10 Although the majority of studies showed a positive correlation between margin dose and obliteration rate, the relationship between dose and timing of obliteration is less rigorously examined. As such, the present analysis provides a valuable addition to the relatively sparse literature regarding this topic.
Shin et al. reported a negative association between nidus size and obliteration rate after SRS for pediatric AVMs, and they identified a mean nidus diameter of ≤ 2 cm, nidus volume of ≤ 3.8 cm3, and maximum diameter < 3 cm as predictors of obliteration.40 Other studies have illustrated the relationship between smaller AVM size and improved obliteration rates.18,23,41 Most prior pediatric AVM SRS series assessed predictors of overall obliteration, whereas the predictors identified in our study pertain specifically to early obliteration. While many studies may suffer from overfitting of their models in order to identify predictors of obliteration, our study provides a better out-of-sample model performance via the use of Lasso regression.
Since patients with early obliteration have a shorter latency period, we hypothesized that this subset of pediatric AVMs would have a lower risk of post-SRS hemorrhage compared with those with late obliteration. Although the crude rate of post-SRS hemorrhage was 50% lower in the early (3.2%) versus late (6.4%) obliteration cohort, this difference was not significant (p = 0.248). We suspect this is secondary to inadequate power. Additionally, the annual post-SRS hemorrhage rates were statistically similar between the two cohorts. The similarity in results between early and late obliterators mirrors the findings of the previous IRRF study that comprised adult and pediatric patients.26 The other outcomes, including RICs, delayed cyst development, SRS-induced tumor formation, and mortality, were also similar between pediatric AVMs with early versus late obliteration. This partly contrasts with the previous IRRF study of early obliteration, which reported a higher rate of radiological RICs in the late obliteration cohort.26 Another difference between the findings of our analysis and those of the prior one is the predictive capability of commonly used AVM SRS grading scales for early obliteration. Of note, the SM grade was not a predictor of early obliteration in either study, and therefore, it appears to correlate poorly with the timing of obliteration after AVM SRS.
It is important to note the limitations of our study. The results depend on the accuracy and reliability of retrospectively acquired data from the contributing centers without independent verification, thereby subjecting the data to reporting bias. Since each contributing center is experienced with AVM SRS, our study’s generalizability could be inhibited by this lack of data verification. The treated AVMs represent a highly selected patient population who are unique to referral patterns of the tertiary SRS institutions comprising the IRRF, and this could subject our findings to referral, selection, and treatment biases related to the specific centers and their physicians. Additionally, due to the inconsistent quality of clinical follow-up that is commonly available to tertiary referral centers for AVM SRS, we were unable to compare detailed clinical follow-up (e.g., functional and educational outcomes) between the early versus late obliteration cohorts. Unfortunately, the collected data were not granular enough to ascertain the symptomatology of each post-SRS hemorrhage.
The timing of follow-up imaging to assess obliteration was subject to variations across centers, and this was particularly the case for confirmatory cerebral angiograms. We acknowledge that DSA remains the gold standard for evaluating obliteration after AVM SRS. However, MRI has been found to be an acceptably accurate alternative to DSA for the assessment of post-SRS nidal patency.42–44 Since the vast majority of obliterations in each cohort were determined by DSA and the between-group rates of DSA-confirmed obliteration were not significantly different, we do not believe that our interpretation of the results was appreciably impacted by the minority of cases followed with MRI alone. One should also note that MRI and CTA can misjudge AVM recurrence or progression. As such, variations in follow-up duration and the lack of uniformly rigorous angiographic follow-up across all cases could have led to an underestimation of AVM recurrence in our analysis. Although all collected variables were entered into the Lasso regression model to identify independent predictors of early obliteration, other relevant variables that were not captured in the database and thus went unaccounted for in our model could exist. Our results are limited to the pediatric AVM population, and therefore, they cannot be generalized to AVMs in adults. Furthermore, our findings may not be generalizable to high-grade AVMs, since SM grade II and III AVMs comprised the majority of the study cohort. We have previously published the outcomes of the entire IRRF pediatric AVM cohort.45 Since the present study only included obliterated AVMs in order to focus on the timing of post-SRS obliteration, our findings are not applicable to those with partial regression of the AVM nidus.
Conclusions
Pediatric AVM patients in whom nidal obliteration is achieved after SRS will reach this radiological endpoint within 24 months of intervention in approximately one-quarter of the cases. We identified patient-, AVM-, and SRS-specific predictors of early obliteration, including female sex, smaller AVM volume, deep AVM location, higher margin dose, higher isodose line, and single SRS procedure. These predictors of early obliteration could aid in the prognostication and management of pediatric AVMs. Pediatric AVM patients with early, as compared with late, obliteration have a lower crude rate of post-SRS hemorrhage.
Disclosures
Dr. Grills reports < 5% stock ownership in and service on the executive board of directors of Greater Michigan Gamma Knife. Dr. Lunsford reports stock ownership in Elekta AB and being a consultant for Insightec and DSMB. Dr. Kondziolka reports funding from Brainlab for research support in brain tumor imaging (not related to this study).
Author Contributions
Conception and design: Burke, Chen, Ding, Starke, Sheehan. Acquisition of data: Lee, Kano, Kearns, Tzeng, Yang, Huang, Kondziolka, Mathieu, Iorio-Morin, Grills, Feliciano, Barnett, Lunsford. Analysis and interpretation of data: Burke, Chen, Ding, Buell, Sokolowski, Lee, Kano, Tzeng, Yang, Huang, Kondziolka, Ironside, Mathieu, Iorio-Morin, Grills, Feliciano, Barnett, Starke, Lunsford, Sheehan. Drafting the article: Burke, Chen, Ding, Sheehan. Critically revising the article: all authors.
References
- 1↑
Meyer-Heim AD, Boltshauser E. Spontaneous intracranial haemorrhage in children: aetiology, presentation and outcome. Brain Dev. 2003;25(6):416–421.
- 2↑
El-Ghanem M, Kass-Hout T, Kass-Hout O, et al. Arteriovenous malformations in the pediatric population: review of the existing literature. Intervent Neurol. 2016;5(3-4):218–225.
- 3↑
Chen CJ, Lee CC, Ding D, et al. Stereotactic radiosurgery for unruptured versus ruptured pediatric brain arteriovenous malformations. Stroke. 2019;50(10):2745–2751.
- 4↑
Ding D, Ilyas A, Sheehan JP. Contemporary management of high-grade brain arteriovenous malformations. Neurosurgery. 2018;65(CN_suppl_1):24–33.
- 5↑
Nicolato A, Lupidi F, Sandri MF, et al. Gamma knife radiosurgery for cerebral arteriovenous malformations in children/adolescents and adults. Part I: Differences in epidemiologic, morphologic, and clinical characteristics, permanent complications, and bleeding in the latency period. Int J Radiat Oncol Biol Phys. 2006;64(3):904–913.
- 6
Stapf C, Khaw AV, Sciacca RR, et al. Effect of age on clinical and morphological characteristics in patients with brain arteriovenous malformation. Stroke. 2003;34(11):2664–2669.
- 7↑
Celli P, Ferrante L, Palma L, Cavedon G. Cerebral arteriovenous malformations in children. Clinical features and outcome of treatment in children and in adults. Surg Neurol. 1984;22(1):43–49.
- 8
Nicolato A, Lupidi F, Sandri MF, et al. Gamma Knife radiosurgery for cerebral arteriovenous malformations in children/adolescents and adults. Part II: Differences in obliteration rates, treatment-obliteration intervals, and prognostic factors. Int J Radiat Oncol Biol Phys. 2006;64(3):914–921.
- 9↑
Yen CP, Monteith SJ, Nguyen JH, et al. Gamma Knife surgery for arteriovenous malformations in children. J Neurosurg Pediatr. 2010;6(5):426–434.
- 10↑
Dinca EB, de Lacy P, Yianni J, et al. Gamma knife surgery for pediatric arteriovenous malformations: a 25-year retrospective study. J Neurosurg Pediatr. 2012;10(5):445–450.
- 11
Kondziolka D, Humphreys RP, Hoffman HJ, et al. Arteriovenous malformations of the brain in children: a forty year experience. Can J Neurol Sci. 1992;19(1):40–45.
- 12
Lindqvist M, Karlsson B, Guo WY, et al. Angiographic long-term follow-up data for arteriovenous malformations previously proven to be obliterated after gamma knife radiosurgery. Neurosurgery. 2000;46(4):803–810.
- 13
Ding D, Starke RM, Kano H, et al. International multicenter cohort study of pediatric brain arteriovenous malformations. Part 1: Predictors of hemorrhagic presentation. J Neurosurg Pediatr. 2017;19(2):127–135.
- 14↑
Hasegawa T, Kato T, Naito T, et al. Long-term outcomes for pediatric patients with brain arteriovenous malformations treated with Gamma Knife radiosurgery, Part 1: Analysis of nidus obliteration rates and related factors. World Neurosurg. 2019;126:e1518–e1525.
- 15↑
Foy AB, Wetjen N, Pollock BE. Stereotactic radiosurgery for pediatric arteriovenous malformations. Neurosurg Clin N Am. 2010;21(3):457–461.
- 16
Börcek AO, Çeltikçi E, Aksoğan Y, Rousseau MJ. Clinical outcomes of stereotactic radiosurgery for cerebral arteriovenous malformations in pediatric patients: systematic review and meta-analysis. Neurosurgery. 2019;85(4):E629–E640.
- 17
Levy EI, Niranjan A, Thompson TP, et al. Radiosurgery for childhood intracranial arteriovenous malformations. Neurosurgery. 2000;47(4):834–842.
- 18↑
Hanakita S, Koga T, Shin M, et al. The long-term outcomes of radiosurgery for arteriovenous malformations in pediatric and adolescent populations. J Neurosurg Pediatr. 2015;16(2):222–231.
- 19
Riva D, Pantaleoni C, Devoti M, et al. Radiosurgery for cerebral AVMs in children and adolescents: the neurobehavioral outcome. J Neurosurg. 1997;86(2):207–210.
- 20↑
Ding D, Xu Z, Yen CP, et al. Radiosurgery for unruptured cerebral arteriovenous malformations in pediatric patients. Acta Neurochir (Wien). 2015;157(2):281–291.
- 21
Börcek AO, Emmez H, Akkan KM, et al. Gamma Knife radiosurgery for arteriovenous malformations in pediatric patients. Childs Nerv Syst. 2014;30(9):1485–1492.
- 22
Kemeny AA, Dias PS, Forster DM. Results of stereotactic radiosurgery of arteriovenous malformations: an analysis of 52 cases. J Neurol Neurosurg Psychiatry. 1989;52(5):554–558.
- 23↑
Pan DH, Kuo YH, Guo WY, et al. Gamma Knife surgery for cerebral arteriovenous malformations in children: a 13-year experience. J Neurosurg Pediatr. 2008;1(4):296–304.
- 24↑
Kano H, Kondziolka D, Flickinger JC, et al. Stereotactic radiosurgery for arteriovenous malformations, Part 2: Management of pediatric patients. J Neurosurg Pediatr. 2012;9(1):1–10.
- 25↑
Ding D, Starke RM, Kano H, et al. Radiosurgery for cerebral arteriovenous malformations in A Randomized Trial of Unruptured Brain Arteriovenous Malformations (ARUBA)-eligible patients: a multicenter study. Stroke. 2016;47(2):342–349.
- 26↑
Cohen-Inbar O, Starke RM, Paisan G, et al. Early versus late arteriovenous malformation responders after stereotactic radiosurgery: an international multicenter study. J Neurosurg. 2017;127(3):503–511.
- 27↑
Chen CJ, Ding D, Kano H, et al. Stereotactic radiosurgery for pediatric versus adult brain arteriovenous malformations. Stroke. 2018;49(8):1939–1945.
- 28
Starke RM, Yen CP, Ding D, Sheehan JP. A practical grading scale for predicting outcome after radiosurgery for arteriovenous malformations: analysis of 1012 treated patients. J Neurosurg. 2013;119(4):981–987.
- 29↑
Spetzler RF, Martin NA. A proposed grading system for arteriovenous malformations. J Neurosurg. 1986;65(4):476–483.
- 30↑
Wegner RE, Oysul K, Pollock BE, et al. A modified radiosurgery-based arteriovenous malformation grading scale and its correlation with outcomes. Int J Radiat Oncol Biol Phys. 2011;79(4):1147–1150.
- 31↑
Steiner L, Lindquist C, Adler JR, et al. Clinical outcome of radiosurgery for cerebral arteriovenous malformations. J Neurosurg. 1992;77(1):1–8.
- 32↑
Ding D, Chen CJ, Starke RM, et al. Risk of brain arteriovenous malformation hemorrhage before and after stereotactic radiosurgery. Stroke. 2019;50(6):1384–1391.
- 33↑
Ding D, Starke RM, Sheehan JP. Radiosurgery for the management of cerebral arteriovenous malformations. Handb Clin Neurol. 2017;143:69–83.
- 34↑
Yamamoto M, Jimbo M, Ide M, et al. Postradiation volume changes in gamma unit-treated cerebral arteriovenous malformations. Surg Neurol. 1993;40(6):485–490.
- 35↑
Rubin BA, Brunswick A, Riina H, Kondziolka D. Advances in radiosurgery for arteriovenous malformations of the brain. Neurosurgery. 2014;74(suppl 1):S50–S59.
- 36↑
Ding D, Xu Z, Shih HH, et al. Worse outcomes after repeat vs initial stereotactic radiosurgery for cerebral arteriovenous malformations: a retrospective matched-cohort study. Neurosurgery. 2016;79(5):690–700.
- 37↑
Starke RM, Ding D, Kano H, et al. International multicenter cohort study of pediatric brain arteriovenous malformations. Part 2: Outcomes after stereotactic radiosurgery. J Neurosurg Pediatr. 2017;19(2):136–148.
- 38↑
Smyth MD, Sneed PK, Ciricillo SF, et al. Stereotactic radiosurgery for pediatric intracranial arteriovenous malformations: the University of California at San Francisco experience. J Neurosurg. 2002;97(1):48–55.
- 39↑
Potts MB, Sheth SA, Louie J, et al. Stereotactic radiosurgery at a low marginal dose for the treatment of pediatric arteriovenous malformations: obliteration, complications, and functional outcomes. J Neurosurg Pediatr. 2014;14(1):1–11.
- 40↑
Shin M, Kawamoto S, Kurita H, et al. Retrospective analysis of a 10-year experience of stereotactic radio surgery for arteriovenous malformations in children and adolescents. J Neurosurg. 2002;97(4):779–784.
- 41↑
Nataf F, Schlienger M, Lefkopoulos D, et al. Radiosurgery of cerebral arteriovenous malformations in children: a series of 57 cases. Int J Radiat Oncol Biol Phys. 2003;57(1):184–195.
- 42
Pollock BE, Kondziolka D, Flickinger JC, et al. Magnetic resonance imaging: an accurate method to evaluate arteriovenous malformations after stereotactic radiosurgery. J Neurosurg. 1996;85(6):1044–1049.
- 43
Lee CC, Reardon MA, Ball BZ, et al. The predictive value of magnetic resonance imaging in evaluating intracranial arteriovenous malformation obliteration after stereotactic radiosurgery. J Neurosurg. 2015;123(1):136–144.
- 44
OʼConnor TE, Friedman WA. Magnetic resonance imaging assessment of cerebral arteriovenous malformation obliteration after stereotactic radiosurgery. Neurosurgery. 2013;73(5):761–766.
- 45↑
Chen CJ, Lee CC, Kano H, et al. Stereotactic radiosurgery for pediatric brain arteriovenous malformations: long-term outcomes. J Neurosurg Pediatr. 2020;25(5):497–505.
