Clinical Neurology and Neurosurgery 126 (2014) 110–114
Contents lists available at ScienceDirect
Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro
Gamma knife radiosurgery of the symptomatic brain stem cavernous angioma with low marginal dose Byung Sup Kim, Je Young Yeon, Jong-Soo Kim, Seung-Chyul Hong, Jung-Il Lee ∗ Department of Neurosurgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, South Korea
a r t i c l e
i n f o
Article history: Received 28 May 2014 Received in revised form 4 August 2014 Accepted 25 August 2014 Available online 6 September 2014 Keywords: Brain stem Cavernous angioma Gamma knife radiosurgery
a b s t r a c t Objective: To analyze the outcome of gamma knife radiosurgery (GKS) using low marginal dose for the symptomatic brain stem cavernous angioma (BSCA). Methods: 39 patients (16 males, 23 females) were treated with GKS for BSCA from January 1997 to September 2012. Clinical data were analyzed retrospectively. The mean age was 41.5 years. All patients had a history of symptomatic bleeding once or more before performing GKS. Mean volume of BSCA was 1095.3 mm3 and median prescribed marginal dose was 13 Gy. Results: Mean follow-up period since diagnosis was 4.1 years. The number of hemorrhagic events between initial diagnosis and GKS was 5 over a total of 14.9 patients-years with annual hemorrhagic rate of 33.6%. Following GKS, there were five hemorrhagic events within the first 2 years (8.1%/year) and two after the first 2 years (2.4%/year). The difference was not statistically significant. Neurologic status improved in 24 patients (61.5%), and stationary in eleven (28.2%). 4 patients (10.3%) experienced the exacerbation of symptoms at the last follow-up and none of them were related to the radiation injury. Significant volume reduction after GKS was observed in 24 patients (61.5%). Surgical excision was performed in one patient due to swelling and rebleeding after GKS. Age at presentation, sex, mass size of BSCA, and location, GKS dose did not affect post-GKS hemorrhage. Conclusions: GKS for BSCA using relatively low marginal dose is safe and effective. Long-term prospective study is needed to confirm the optimal dose for BSCA. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Cavernous angioma or cavernoma is a benign vascular malformation, which can be found at any region within the brain [1]. Cavernous angioma accounts for 5%–15% of all central nervous system vascular malformations [2,3]. Brain stem cavernous angioma (BSCA) is rare, affecting only 0.4%–0.9% of the population [4–9] and representing 9%–35% of intracranial cavernomas [10–14]. BSCA is most frequently found in the pons. The current treatment of choice for most symptomatic cavernous angioma is microsurgical resection [15]; however, morbidity and mortality associated with the operation are especially high in a surgically inaccessible lesions such as brain stem. Therefore, radiosurgery has been used as an alternative method of treatment and the primary goal of the radiosurgery is the same as in the case of AVMs; the prevention
∗ Corresponding author at: Department of Neurosurgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, 81 Irwon-ro, Gangnam-gu, Seoul 135-710, South Korea. Tel.: +82 2 3410 3494; fax: +82 2 3410 0048. E-mail address: [email protected] (J.-I. Lee). http://dx.doi.org/10.1016/j.clineuro.2014.08.028 0303-8467/© 2014 Elsevier B.V. All rights reserved.
of repeated hemorrhage. Though the natural history of BSCA and the use of radiosurgery in the management are still controversial issues recent studies reported reduction of recurrent hemorrhage after radiosurgery. Marginal dose used in radiosurgery, however, varies according to authors or institutions and optimal dose is not yet clearly determined. The purpose of this study is to investigate the outcome of GKS with low marginal dose for BSCA. 2. Materials and methods 2.1. Patient population For this study, an Institutional Review Board approved retrospective chart of 39 patients undergoing gamma knife radiosurgery for brain stem cavernous angioma from January 1997 to September 2012 at Samsung Medical Center was reviewed. Mean age of the patients was 41.5 years ranging from 18 to 64. The ratio of men to women was 1:1.4. The most common location of BSCA was pons (30 patients, 76.9%) followed by medulla (4 patients, 10.3%), midbrain (3 patients, 7.7%), and pontomedullary junction (2 patients, 5.1%). Four patients (10.3%) had undergone previous partial microsurgical
B.S. Kim et al. / Clinical Neurology and Neurosurgery 126 (2014) 110–114
resection before GKS. Abnormal neurological symptom or sign was present in all patients. Most patients (64.1%) were presented with complex neurological deficit. Initial neurological manifestation was vertigo (15 patients), sensory deficit (12), diplopia (10), dysarthria (9), hemiparesis (7), headache (7), facial palsy (6), 6th nerve palsy (5), general weakness (1), and tinnitus (1). Bleeding events were defined as the development of hemorrhage in follow-up imaging associated with new neurological deficit. There were 44 bleeding events in 39 patients reported before GKS. All patients had a history of symptomatic bleeding once or more before performing GKS and bleeding happened repeatedly in 5 patients. Mean volume of BSCA at the time of GKS was 1095.3 mm3 (range 31.5–5400). Additional lesions outside the brain stem were found in 3 patients. Adjacent venous anomalies were seen through MRI in 10 patients (25.6%). 2.2. Radiosurgical technique GKS was performed in the following procedure. After application of Leksell stereotactic frame type G under local anesthesia, MRI scanning was done. Stereotactic MR imaging including T2weighted images and postcontrast FLASH (Fast Low Angle SHot) sequences were used for dose planning. Upon administrating double dose contrast medium, we obtained a Fast Low Angle Shot images (FLASH, 1 mm-thick slices with no gap, 512 × 512 pixel). Images were transferred to planning workstations for Leksell Gamma Knife (Elekta Instruments) and dose planning was performed using GammaPlan versions 5.31, 5.32, 5.34, and 9.0. Radiosurgery dose plans, with single or multiple isocenters (range 1–9), were created and the targeted edge of the BSCA was considered to be the region characterized by mixed signal change within the T2-weighted signal-defined hemosiderin ring. This margin served as the 50% or greater isodose line. The median marginal dose was 13 Gy (range 11–18 Gy) and median isodose 50%. Irradiation was performed with Leksell Gamma Knife type B, C or Perfexion. Each patient received a 10 mg loading dose of dexamethasone in the morning on the day of GKS followed by 5 mg after completion. All patients were discharged from the hospital within 24 h. 2.3. Follow-up Follow-up MR imaging was performed at 6-month intervals during the first year after GKS. It was recommended on an annual basis after the first year. Mean clinical follow-up period since diagnosis was 4.1 years (range 0.4–16 years). 23 patients (59%) had follow-up period longer than 2 years and 11 (28.5%) longer than 5 years. Mean follow-up duration till the last MRI was 3.4 years (range 0.5–16). 2.4. Statistical analysis We analyzed for variables which might affect the rebleeding of BSCA such as location, sex, age at presentation, mass size, and marginal dose using Cox-regression analysis. Tied time was considered using Efron method. We analyzed using Stata 13.0 (StataCorp LP, 4905 Lakeway Drive, College Station, TX 77845-4512, USA). Bleeding events was defined as imaging evidence corresponding to a newly developing neurological sign or symptom. We calculated the annual rate of recurrent hemorrhage between diagnosis and treatment using the following formula: total number of hemorrhages (excluding the first hemorrhage at diagnosis) in all patients/total number of patient-years between diagnosis and GKS. Post-treatment annual hemorrhage rate was calculated as a total number of hemorrhages after GKS in all patients/total number of patient-years after GKS. Hemorrhage rates were analyzed
111
before and after GKS using Poisson distribution. Exact Poisson 95% confidence intervals were calculated and 95% confidence intervals (95% CI) were corrected using Bonferroni’s method due to multiple testing. Finally, we statistically analyzed the difference of an annual hemorrhage rates after GKS between marginal doses below 13 Gy and higher than 14 Gy using Poisson distribution. Exact Poisson 95% confidence intervals were calculated. 3. Results 3.1. Pretreatment hemorrhage rate Observation periods before GKS comprised of the time from hemorrhage in MRI associated with the first symptom to the time of GKS. Total of 14.9 patient-years were observed with a mean pretreatment observation period of 0.4 years (range 0–5.6 years). All patients had at least one hemorrhagic event prior to GKS. Because of the five patients presented with a recurrent hemorrhage before GKS, there were 44 hemorrhagic events documented during this time (1.1 hemorrhagic events per patient). After exclusion of the first hemorrhage, the calculated annual rate of recurrent hemorrhage was 33.6% (95% CI: 8.26–89.80). 3.2. Post-treatment hemorrhage rate The observation period following GKS was considered to be the time following the treatment until the most recent clinical followup. Thus, the mean follow-up period after GKS was 3.7 years (range 0.3–15.9 years) with an overall observation period of 145.9 personyears. During the follow-up after GKS, no patients died. Seven hemorrhagic events in six patients were documented during this period. Only two hemorrhagic events developed after 2 years in each patient. The annual hemorrhage rate during the first 2 years after GKS was calculated to be 8.1%. (5 hemorrhages/61.4 personyears, 95% CI: 2.64–19.00) and the annual hemorrhage rate after the initial 2 year was calculated to be 2.4% (2 hemorrhages/84.5 personyears, 95% CI: 0.29–8.55). However, statistical analysis using 95% confidence interval did not reveal a significant reduction in the annual hemorrhage rate within 2 year after GKS (8.1%, 95% CI: 2.64–19.00) compared to before GKS (33.6%, 95% CI: 8.26–89.80). Also, statistical analysis did not reveal a significant reduction in the annual hemorrhage rate within a 2-year period (8.1%, 95% CI: 2.64–19.00) compared to following the 2-year latency period after GKS (2.4%, 95% CI: 0.29–8.55). 3.3. Factors associated with rebleeding Univariate analysis using the Cox-regression proportional hazard model did not reveal statistically significant association of factors including age at presentation (p = 0.1428), sex (p = 0.5926), GKS dose (p = 0.3319), and mass size (p = 0.4182) with rebleeding after GKS. When we compared the annual hemorrhage rates between the marginal doses below 13 Gy (3.64%, 95% CI: 0.75–10.64) and higher than 14 Gy (9.75%, 95% CI: 3.58–21.21) using 95% confidence interval calculated by Poisson distribution, there was no statistically significant difference. 3.4. Mass shrinkage (Fig. 1) Volume of the cavernous angioma before and after the treatment was measured on MR FLASH images using Gamma plan. We defined significant mass shrinkage or increment as change of total volume of more than 20% from the baseline. In all patients, followup MRI was available. Mass shrinkage was revealed in 24 patients
112
B.S. Kim et al. / Clinical Neurology and Neurosurgery 126 (2014) 110–114
among 39 patients (61.5%), deteriorated in 4 (10.3%), and stationary in 11 patients (28.2%). No patients died at the time of this writing. 3.6. Adverse radiation effects Two patients (5.1%) experienced new neurologic symptoms following GKS in the absence of a new hemorrhage. That neurological deterioration was transient and was fully recovered after a short course of dexamethasone. 4. Discussion
Fig. 1. A representative case of gamma knife radiosurgery with low marginal dose for a 39-year-old woman with a pontine cavernous angioma is shown. (A) Gamma knife radiosurgery plan is shown on T2-weighted axial image. Prescribed marginal dose (the innermost isodose line) is 13 Gy and additional outer isodose lines indicate 10 Gy and 5 Gy each respectively. (B) T2-weighted images obtained at 40 months after treatment reveals marked shrinkage of the lesion.
(61.5%), stationary in 13 patients (33.4%) and mass increment in 2 patients (5.1%) during the last follow-up images.
3.5. Neurologic outcome One patient (2.6%) underwent operation after GKS due to rebleeding and edema. Neurologic symptoms were improved in 24
The natural history of BSCA is not known in detail. The growth velocity of BSCA is slow but they may be neurologically devastated [16]. Annual hemorrhage rates of BSCA are 0.7–6% and annual rebleeding rate is about 5–30% [6,16–18]. Also, it has a 20-fold increased incidence of rebleeding compared to a supratentorial lesion [16,19]. BSCA becomes symptomatic with progressive neurological deficits due to mass effect on brain stem tracts and nuclei [16]. According to other studies [16,18,20], symptomatic BSCA have a high rate of rehemorrhage (up to a 60% annual risk) and corresponding neurological deficits. The aim of gamma knife radiosurgery for BSCA is the prevention of repeated bleeding and neurologic deterioration. There are a few reports about the beneficial effect of gamma knife radiosurgery on BSCA. It was suggested that cavernous angioma may respond to SRS in a similar way to other vascular lesions via delayed luminal closure of vascular channels [21]. Obliteration of AVMs usually occurring in the period of one to three years following treatment was proved in many studies [22–26] and cavernous angioma after SRS revealed similar temporal course with significant reduction of rebleeding after a latency period [5,8,27]. Our study showed similar outcome corresponding to previous literatures. A pre-GKS annual hemorrhage rate of 33.6% declined to 8.1% in the first 2 years after GKS and further decreased to 2.4%. There was no statistical significance, however, between the first 2 years and the period after 2 years latency. While clinical factors such as age at presentation, sex, mass size, location, and GKS dose could influence hemorrhage, we could not find the correlation between the postulated risk factors and rebleeding. Venous anomalies are associated with 8% to 100% of BSCA [11,16]. These venous anomalies may induce the formation of cavernous malformations and have a role in their recurrence [16]. Unlike the contribution of venous anomaly to the pathogenesis of cavernous angioma, we observed rebleeding after GKS in only one patient associated with venous anomaly. Also it deserves to be mentioned that there was no dose effect relationship within dose range from 11 to 18 Gy. Shrinkage of the lesion after GKS was observed in 61.5% of the patients, though it is not clear whether this change was really induced by radiation or absorption of hemorrhage. Majority of the patients were neurologically improved or stable at the last followup. Therefore, overall outcome after GKS in our study was favorable and compatible with those of other publications (Table 1). Different implication of our study includes the optimal level of marginal dose. In majority of other studies, mean or median marginal dose ranged from 15 to 18 Gy and annual hemorrhage rate after GKS was ranged from 1.1 to 22.4% with significant reduction of hemorrhage after 2 years. However, there was substantial rate of radiation-related complication in 11 to 59% of the patients. Only one paper reported the outcome of BSCA treated with relatively lower mean marginal dose [28]. Annual hemorrhage rate was 10.3% during the first 2 years after GKS and 3.3% afterwards. Meanwhile, complication was reported as only 2.5%. That result was reproduced in our study using 13 Gy of median marginal dose with transient adverse radiation effects in 5.1% of the patients. Relatively low marginal dose used in our study resulted in the reduction of annual hemorrhage
59 5.1 34.45 33.6
GKS using marginal dose as low as 11 Gy was safe without serious radiation-related complication while its efficacy in reduction of annual hemorrhage rate, mass shrinkage and improvement of neurologic symptoms was compatible with that of high marginal dose. Further prospective study is needed for revealing natural history of BSCA and optimal marginal dose.
NA 28.25 (17.5–37.7)
References
0.7–4.7 1.1 (0.03–5.4)
3.12 (0.032–25.9)
1.42 (0.09–4.8) 2.25 (0.08–15.2) 3.1 (0.2–38.8)
a
18 (16–23) 13 (11–18)
2.5 29.2 NA
27.3 11 26.5 35.5 9.4 17.3 25.7 (18–50) NA 18.3 (8–36.7) a
16.1 (8–24) 18 (7–40) 15 (6.5–30) a 16.5 (7.2–33) 12.1 (9–20) a
5. Conclusions
53 (24–78) 49.2 (5–192)
64.8 (9–122.4)
a
38.3 (21–67) 90 (9–166) 64.8 (4–147)
2.1 (0.5–7.9)
Median, NA: not applicable, RS: radiosurgery. a
15 39 Garcia-Munoz et al. (2007) [34] Present study
22 57 95 Kim et al. (2002) [31] Chang et al. (1998) [32] Amin-Hanjani et al. (1998) [33]
125
17 Pollock et al. (2000) [30]
Liu et al. (2005) [28]
47 Kondziolka et al. (1995) [27]
No. of patients Author and year
Table 1 Overall outcome after GKS in other publications.
113
rates compatible with other studies with significantly low risk of complication. This dose risk relationship is also supported by observation of Liscak et al. [29] who reported an increased incidence of collateral edema by marginal dose exceeding 15 Gy. Therefore, our data suggest that adequate marginal dose may be as low as 11 Gy though optimal marginal dose for BSCA cannot be exactly defined. To get more reasonable guidelines, prospective study in a large number of patients with a long-term follow-up is needed.
a
51
a
2.1 (0.2–8.1) 43.2 (4–77)
Mean follow-up (mos)
Mean lesion vol. (cm3 )
a
18 (16–18)
a
32 (20–36)
40.1
59
26 After RS
8.8 (0–2 years) 1.1 (2–6 years) 8.8 (0–2 years) 2.9 (after 2 years) 1.55 1.6 (after 3 years) 22.4 (0–2 years) 4.5 (after 2 years) 10.3 (0–2 years) 3.3 (after 2 years) 7.17 8.1 (0–2 years) 2.4 (after 2 years)
Before RS
16 (10–20)
32 (20–40)
32
Radiation-related complications (%) Radiation-related complications (%) Mean marginal dose (Gy)
Mean maximal dose (Gy)
Annual hemorrage rate (%)
B.S. Kim et al. / Clinical Neurology and Neurosurgery 126 (2014) 110–114
[1] Moran NF, Fish DR, Kitchen N, Shorvon S, Kendall BE, Stevens JM. Supratentorial cavernous haemangiomas and epilepsy: a review of the literature and case series. J Neurol Neurosurg Psychiatry 1999;66:561–8. [2] Di Rocco C, Iannelli A, Tamburrini G. Cavernous angiomas of the brain stem in children. Pediatr Neurosurg 1997;27:92–9. [3] Moriarity JL, Clatterbuck RE, Rigamonti D. The natural history of cavernous malformations. Neurosurg Clin N Am 1999;10:411–7. [4] Cantore G, Missori P, Santoro A. Cavernous angiomas of the brain stem intraaxial anatomical pitfalls and surgical strategies. Surg Neurol 1999;52:84–93, discussion 93–4. [5] Del Curling Jr O, Kelly Jr DL, Elster AD, Craven TE. An analysis of the natural history of cavernous angiomas. J Neurosurg 1991;75:702–8. [6] Kim DS, Park YG, Choi JU, Chung SS, Lee KC. An analysis of the natural history of cavernous malformations. Surg Neurol 1997;48:9–17, discussion 17–8. [7] McCormick WF. The pathology of vascular (“arteriovenous”) malformations. J Neurosurg 1966;24:807–16. [8] Robinson JR, Awad IA, Little JR. Natural history of the cavernous angioma. J Neurosurg 1991;75:709–14. [9] Sage MR, Brophy BP, Sweeney C, Phipps S, Perrett LV, Sandhu A, et al. Cavernous haemangiomas (angiomas) of the brain: clinically significant lesions. Australas Radiol 1993;37:147–55. [10] Fahlbusch R, Strauss C, Huk W, Rockelein G, Kompf D, Ruprecht KW. Surgical removal of pontomesencephalic cavernous hemangiomas. Neurosurgery 1990;26:449–56, discussion 456–7. [11] Fritschi JA, Reulen HJ, Spetzler RF, Zabramski JM. Cavernous malformations of the brain stem. A review of 139 cases. Acta Neurochir (Wien) 1994;130:35–46. [12] Karlsson B, Kihlstrom L, Lindquist C, Ericson K, Steiner L. Radiosurgery for cavernous malformations. J Neurosurg 1998;88:293–7. [13] Kondziolka D, Lunsford LD, Kestle JR. The natural history of cerebral cavernous malformations. J Neurosurg 1995;83:820–4. [14] Simard JM, Garcia-Bengochea F, Ballinger Jr WE, Mickle JP, Quisling RG. Cavernous angioma: a review of 126 collected and 12 new clinical cases. Neurosurgery 1986;18:162–72. [15] Kashiwagi S, van Loveren HR, Tew Jr JM, Wiot JG, Weil SM, Lukin RA. Diagnosis and treatment of vascular brain-stem malformations. J Neurosurg 1990;72:27–34. [16] Porter RW, Detwiler PW, Spetzler RF, Lawton MT, Baskin JJ, Derksen PT, et al. Cavernous malformations of the brainstem: experience with 100 patients. J Neurosurg 1999;90:50–8. [17] Clatterbuck RE, Moriarity JL, Elmaci I, Lee RR, Breiter SN, Rigamonti D. Dynamic nature of cavernous malformations: a prospective magnetic resonance imaging study with volumetric analysis. J Neurosurg 2000;93:981–6. [18] Wang CC, Liu A, Zhang JT, Sun B, Zhao YL. Surgical management of brain-stem cavernous malformations: report of 137 cases. Surg Neurol 2003;59:444–54, discussion 454. [19] Bertalanffy H, Benes L, Miyazawa T, Alberti O, Siegel AM, Sure U. Cerebral cavernomas in the adult. Review of the literature and analysis of 72 surgically treated patients. Neurosurg Rev 2002;25:1–53, discussion 54–5. [20] Hauck EF, Barnett SL, White JA, Samson D. Symptomatic brainstem cavernomas. Neurosurgery 2009;64:61–70, discussion 70–1. [21] Monaco EA, Khan AA, Niranjan A, Kano H, Grandhi R, Kondziolka D, et al. Stereotactic radiosurgery for the treatment of symptomatic brainstem cavernous malformations. Neurosurg Focus 2010;29:E11. [22] Colombo F, Pozza F, Chierego G, Casentini L, De Luca G, Francescon P. Linear accelerator radiosurgery of cerebral arteriovenous malformations: an update. Neurosurgery 1994;34:14–20, discussion 20–1. [23] Friedman WA, Bova FJ, Mendenhall WM. Linear accelerator radiosurgery for arteriovenous malformations: the relationship of size to outcome. J Neurosurg 1995;82:180–9.
114
B.S. Kim et al. / Clinical Neurology and Neurosurgery 126 (2014) 110–114
[24] Pollock BE, Lunsford LD, Kondziolka D, Maitz A, Flickinger JC. Patient outcomes after stereotactic radiosurgery for “operable” arteriovenous malformations. Neurosurgery 1994;35:1–7, discussion 7–8. [25] Pozzati E, Acciarri N, Tognetti F, Marliani F, Giangaspero F. Growth, subsequent bleeding, and de novo appearance of cerebral cavernous angiomas. Neurosurgery 1996;38:662–9, discussion 669–70. [26] Steiner L, Lindquist C, Adler JR, Torner JC, Alves W, Steiner M. Clinical outcome of radiosurgery for cerebral arteriovenous malformations. J Neurosurg 1992;77:1–8. [27] Kondziolka D, Lunsford LD, Flickinger JC, Kestle JR. Reduction of hemorrhage risk after stereotactic radiosurgery for cavernous malformations. J Neurosurg 1995;83:825–31. [28] Liu KD, Chung WY, Wu HM, Shiau CY, Wang LW, Guo WY, et al. Gamma knife surgery for cavernous hemangiomas: an analysis of 125 patients. J Neurosurg 2005;Suppl. 102:81–6. [29] Liscak R, Vladyka V, Simonova G, Vymazal J, Novotny Jr J. Gamma knife radiosurgery of the brain stem cavernomas. Minim Invasive Neurosurg 2000;43:201–7.
[30] Pollock BE, Garces YI, Stafford SL, Foote RL, Schomberg PJ, Link MJ. Stereotactic radiosurgery for cavernous malformations. J Neurosurg 2000;93: 987–91. [31] Kim DG, Choe WJ, Paek SH, Chung HT, Kim IH, Han DH. Radiosurgery of intracranial cavernous malformations. Acta Neurochir (Wien) 2002;144:869–78, discussion 878. [32] Chang SD, Levy RP, Adler Jr JR, Martin DP, Krakovitz PR, Steinberg GK. Stereotactic radiosurgery of angiographically occult vascular malformations: 14-year experience. Neurosurgery 1998;43:213–20, discussion 220–1. [33] Amin-Hanjani S, Ogilvy CS, Candia GJ, Lyons S, Chapman PH. Stereotactic radiosurgery for cavernous malformations: Kjellberg’s experience with proton beam therapy in 98 cases at the Harvard Cyclotron. Neurosurgery 1998;42:1229–36, discussion 1236–8. [34] Garcia-Munoz L, Velasco-Campos F, Lujan-Castilla P, Enriquez-Barrera M, Cervantes-Martinez A, Carrillo-Ruiz J. Radiosurgery in the treatment of brain cavernomas. Experience with 17 lesions treated in 15 patients. Neurochirurgie 2007;53:243–50.
Contents lists available at ScienceDirect
Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro
Gamma knife radiosurgery of the symptomatic brain stem cavernous angioma with low marginal dose Byung Sup Kim, Je Young Yeon, Jong-Soo Kim, Seung-Chyul Hong, Jung-Il Lee ∗ Department of Neurosurgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, South Korea
a r t i c l e
i n f o
Article history: Received 28 May 2014 Received in revised form 4 August 2014 Accepted 25 August 2014 Available online 6 September 2014 Keywords: Brain stem Cavernous angioma Gamma knife radiosurgery
a b s t r a c t Objective: To analyze the outcome of gamma knife radiosurgery (GKS) using low marginal dose for the symptomatic brain stem cavernous angioma (BSCA). Methods: 39 patients (16 males, 23 females) were treated with GKS for BSCA from January 1997 to September 2012. Clinical data were analyzed retrospectively. The mean age was 41.5 years. All patients had a history of symptomatic bleeding once or more before performing GKS. Mean volume of BSCA was 1095.3 mm3 and median prescribed marginal dose was 13 Gy. Results: Mean follow-up period since diagnosis was 4.1 years. The number of hemorrhagic events between initial diagnosis and GKS was 5 over a total of 14.9 patients-years with annual hemorrhagic rate of 33.6%. Following GKS, there were five hemorrhagic events within the first 2 years (8.1%/year) and two after the first 2 years (2.4%/year). The difference was not statistically significant. Neurologic status improved in 24 patients (61.5%), and stationary in eleven (28.2%). 4 patients (10.3%) experienced the exacerbation of symptoms at the last follow-up and none of them were related to the radiation injury. Significant volume reduction after GKS was observed in 24 patients (61.5%). Surgical excision was performed in one patient due to swelling and rebleeding after GKS. Age at presentation, sex, mass size of BSCA, and location, GKS dose did not affect post-GKS hemorrhage. Conclusions: GKS for BSCA using relatively low marginal dose is safe and effective. Long-term prospective study is needed to confirm the optimal dose for BSCA. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Cavernous angioma or cavernoma is a benign vascular malformation, which can be found at any region within the brain [1]. Cavernous angioma accounts for 5%–15% of all central nervous system vascular malformations [2,3]. Brain stem cavernous angioma (BSCA) is rare, affecting only 0.4%–0.9% of the population [4–9] and representing 9%–35% of intracranial cavernomas [10–14]. BSCA is most frequently found in the pons. The current treatment of choice for most symptomatic cavernous angioma is microsurgical resection [15]; however, morbidity and mortality associated with the operation are especially high in a surgically inaccessible lesions such as brain stem. Therefore, radiosurgery has been used as an alternative method of treatment and the primary goal of the radiosurgery is the same as in the case of AVMs; the prevention
∗ Corresponding author at: Department of Neurosurgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, 81 Irwon-ro, Gangnam-gu, Seoul 135-710, South Korea. Tel.: +82 2 3410 3494; fax: +82 2 3410 0048. E-mail address: [email protected] (J.-I. Lee). http://dx.doi.org/10.1016/j.clineuro.2014.08.028 0303-8467/© 2014 Elsevier B.V. All rights reserved.
of repeated hemorrhage. Though the natural history of BSCA and the use of radiosurgery in the management are still controversial issues recent studies reported reduction of recurrent hemorrhage after radiosurgery. Marginal dose used in radiosurgery, however, varies according to authors or institutions and optimal dose is not yet clearly determined. The purpose of this study is to investigate the outcome of GKS with low marginal dose for BSCA. 2. Materials and methods 2.1. Patient population For this study, an Institutional Review Board approved retrospective chart of 39 patients undergoing gamma knife radiosurgery for brain stem cavernous angioma from January 1997 to September 2012 at Samsung Medical Center was reviewed. Mean age of the patients was 41.5 years ranging from 18 to 64. The ratio of men to women was 1:1.4. The most common location of BSCA was pons (30 patients, 76.9%) followed by medulla (4 patients, 10.3%), midbrain (3 patients, 7.7%), and pontomedullary junction (2 patients, 5.1%). Four patients (10.3%) had undergone previous partial microsurgical
B.S. Kim et al. / Clinical Neurology and Neurosurgery 126 (2014) 110–114
resection before GKS. Abnormal neurological symptom or sign was present in all patients. Most patients (64.1%) were presented with complex neurological deficit. Initial neurological manifestation was vertigo (15 patients), sensory deficit (12), diplopia (10), dysarthria (9), hemiparesis (7), headache (7), facial palsy (6), 6th nerve palsy (5), general weakness (1), and tinnitus (1). Bleeding events were defined as the development of hemorrhage in follow-up imaging associated with new neurological deficit. There were 44 bleeding events in 39 patients reported before GKS. All patients had a history of symptomatic bleeding once or more before performing GKS and bleeding happened repeatedly in 5 patients. Mean volume of BSCA at the time of GKS was 1095.3 mm3 (range 31.5–5400). Additional lesions outside the brain stem were found in 3 patients. Adjacent venous anomalies were seen through MRI in 10 patients (25.6%). 2.2. Radiosurgical technique GKS was performed in the following procedure. After application of Leksell stereotactic frame type G under local anesthesia, MRI scanning was done. Stereotactic MR imaging including T2weighted images and postcontrast FLASH (Fast Low Angle SHot) sequences were used for dose planning. Upon administrating double dose contrast medium, we obtained a Fast Low Angle Shot images (FLASH, 1 mm-thick slices with no gap, 512 × 512 pixel). Images were transferred to planning workstations for Leksell Gamma Knife (Elekta Instruments) and dose planning was performed using GammaPlan versions 5.31, 5.32, 5.34, and 9.0. Radiosurgery dose plans, with single or multiple isocenters (range 1–9), were created and the targeted edge of the BSCA was considered to be the region characterized by mixed signal change within the T2-weighted signal-defined hemosiderin ring. This margin served as the 50% or greater isodose line. The median marginal dose was 13 Gy (range 11–18 Gy) and median isodose 50%. Irradiation was performed with Leksell Gamma Knife type B, C or Perfexion. Each patient received a 10 mg loading dose of dexamethasone in the morning on the day of GKS followed by 5 mg after completion. All patients were discharged from the hospital within 24 h. 2.3. Follow-up Follow-up MR imaging was performed at 6-month intervals during the first year after GKS. It was recommended on an annual basis after the first year. Mean clinical follow-up period since diagnosis was 4.1 years (range 0.4–16 years). 23 patients (59%) had follow-up period longer than 2 years and 11 (28.5%) longer than 5 years. Mean follow-up duration till the last MRI was 3.4 years (range 0.5–16). 2.4. Statistical analysis We analyzed for variables which might affect the rebleeding of BSCA such as location, sex, age at presentation, mass size, and marginal dose using Cox-regression analysis. Tied time was considered using Efron method. We analyzed using Stata 13.0 (StataCorp LP, 4905 Lakeway Drive, College Station, TX 77845-4512, USA). Bleeding events was defined as imaging evidence corresponding to a newly developing neurological sign or symptom. We calculated the annual rate of recurrent hemorrhage between diagnosis and treatment using the following formula: total number of hemorrhages (excluding the first hemorrhage at diagnosis) in all patients/total number of patient-years between diagnosis and GKS. Post-treatment annual hemorrhage rate was calculated as a total number of hemorrhages after GKS in all patients/total number of patient-years after GKS. Hemorrhage rates were analyzed
111
before and after GKS using Poisson distribution. Exact Poisson 95% confidence intervals were calculated and 95% confidence intervals (95% CI) were corrected using Bonferroni’s method due to multiple testing. Finally, we statistically analyzed the difference of an annual hemorrhage rates after GKS between marginal doses below 13 Gy and higher than 14 Gy using Poisson distribution. Exact Poisson 95% confidence intervals were calculated. 3. Results 3.1. Pretreatment hemorrhage rate Observation periods before GKS comprised of the time from hemorrhage in MRI associated with the first symptom to the time of GKS. Total of 14.9 patient-years were observed with a mean pretreatment observation period of 0.4 years (range 0–5.6 years). All patients had at least one hemorrhagic event prior to GKS. Because of the five patients presented with a recurrent hemorrhage before GKS, there were 44 hemorrhagic events documented during this time (1.1 hemorrhagic events per patient). After exclusion of the first hemorrhage, the calculated annual rate of recurrent hemorrhage was 33.6% (95% CI: 8.26–89.80). 3.2. Post-treatment hemorrhage rate The observation period following GKS was considered to be the time following the treatment until the most recent clinical followup. Thus, the mean follow-up period after GKS was 3.7 years (range 0.3–15.9 years) with an overall observation period of 145.9 personyears. During the follow-up after GKS, no patients died. Seven hemorrhagic events in six patients were documented during this period. Only two hemorrhagic events developed after 2 years in each patient. The annual hemorrhage rate during the first 2 years after GKS was calculated to be 8.1%. (5 hemorrhages/61.4 personyears, 95% CI: 2.64–19.00) and the annual hemorrhage rate after the initial 2 year was calculated to be 2.4% (2 hemorrhages/84.5 personyears, 95% CI: 0.29–8.55). However, statistical analysis using 95% confidence interval did not reveal a significant reduction in the annual hemorrhage rate within 2 year after GKS (8.1%, 95% CI: 2.64–19.00) compared to before GKS (33.6%, 95% CI: 8.26–89.80). Also, statistical analysis did not reveal a significant reduction in the annual hemorrhage rate within a 2-year period (8.1%, 95% CI: 2.64–19.00) compared to following the 2-year latency period after GKS (2.4%, 95% CI: 0.29–8.55). 3.3. Factors associated with rebleeding Univariate analysis using the Cox-regression proportional hazard model did not reveal statistically significant association of factors including age at presentation (p = 0.1428), sex (p = 0.5926), GKS dose (p = 0.3319), and mass size (p = 0.4182) with rebleeding after GKS. When we compared the annual hemorrhage rates between the marginal doses below 13 Gy (3.64%, 95% CI: 0.75–10.64) and higher than 14 Gy (9.75%, 95% CI: 3.58–21.21) using 95% confidence interval calculated by Poisson distribution, there was no statistically significant difference. 3.4. Mass shrinkage (Fig. 1) Volume of the cavernous angioma before and after the treatment was measured on MR FLASH images using Gamma plan. We defined significant mass shrinkage or increment as change of total volume of more than 20% from the baseline. In all patients, followup MRI was available. Mass shrinkage was revealed in 24 patients
112
B.S. Kim et al. / Clinical Neurology and Neurosurgery 126 (2014) 110–114
among 39 patients (61.5%), deteriorated in 4 (10.3%), and stationary in 11 patients (28.2%). No patients died at the time of this writing. 3.6. Adverse radiation effects Two patients (5.1%) experienced new neurologic symptoms following GKS in the absence of a new hemorrhage. That neurological deterioration was transient and was fully recovered after a short course of dexamethasone. 4. Discussion
Fig. 1. A representative case of gamma knife radiosurgery with low marginal dose for a 39-year-old woman with a pontine cavernous angioma is shown. (A) Gamma knife radiosurgery plan is shown on T2-weighted axial image. Prescribed marginal dose (the innermost isodose line) is 13 Gy and additional outer isodose lines indicate 10 Gy and 5 Gy each respectively. (B) T2-weighted images obtained at 40 months after treatment reveals marked shrinkage of the lesion.
(61.5%), stationary in 13 patients (33.4%) and mass increment in 2 patients (5.1%) during the last follow-up images.
3.5. Neurologic outcome One patient (2.6%) underwent operation after GKS due to rebleeding and edema. Neurologic symptoms were improved in 24
The natural history of BSCA is not known in detail. The growth velocity of BSCA is slow but they may be neurologically devastated [16]. Annual hemorrhage rates of BSCA are 0.7–6% and annual rebleeding rate is about 5–30% [6,16–18]. Also, it has a 20-fold increased incidence of rebleeding compared to a supratentorial lesion [16,19]. BSCA becomes symptomatic with progressive neurological deficits due to mass effect on brain stem tracts and nuclei [16]. According to other studies [16,18,20], symptomatic BSCA have a high rate of rehemorrhage (up to a 60% annual risk) and corresponding neurological deficits. The aim of gamma knife radiosurgery for BSCA is the prevention of repeated bleeding and neurologic deterioration. There are a few reports about the beneficial effect of gamma knife radiosurgery on BSCA. It was suggested that cavernous angioma may respond to SRS in a similar way to other vascular lesions via delayed luminal closure of vascular channels [21]. Obliteration of AVMs usually occurring in the period of one to three years following treatment was proved in many studies [22–26] and cavernous angioma after SRS revealed similar temporal course with significant reduction of rebleeding after a latency period [5,8,27]. Our study showed similar outcome corresponding to previous literatures. A pre-GKS annual hemorrhage rate of 33.6% declined to 8.1% in the first 2 years after GKS and further decreased to 2.4%. There was no statistical significance, however, between the first 2 years and the period after 2 years latency. While clinical factors such as age at presentation, sex, mass size, location, and GKS dose could influence hemorrhage, we could not find the correlation between the postulated risk factors and rebleeding. Venous anomalies are associated with 8% to 100% of BSCA [11,16]. These venous anomalies may induce the formation of cavernous malformations and have a role in their recurrence [16]. Unlike the contribution of venous anomaly to the pathogenesis of cavernous angioma, we observed rebleeding after GKS in only one patient associated with venous anomaly. Also it deserves to be mentioned that there was no dose effect relationship within dose range from 11 to 18 Gy. Shrinkage of the lesion after GKS was observed in 61.5% of the patients, though it is not clear whether this change was really induced by radiation or absorption of hemorrhage. Majority of the patients were neurologically improved or stable at the last followup. Therefore, overall outcome after GKS in our study was favorable and compatible with those of other publications (Table 1). Different implication of our study includes the optimal level of marginal dose. In majority of other studies, mean or median marginal dose ranged from 15 to 18 Gy and annual hemorrhage rate after GKS was ranged from 1.1 to 22.4% with significant reduction of hemorrhage after 2 years. However, there was substantial rate of radiation-related complication in 11 to 59% of the patients. Only one paper reported the outcome of BSCA treated with relatively lower mean marginal dose [28]. Annual hemorrhage rate was 10.3% during the first 2 years after GKS and 3.3% afterwards. Meanwhile, complication was reported as only 2.5%. That result was reproduced in our study using 13 Gy of median marginal dose with transient adverse radiation effects in 5.1% of the patients. Relatively low marginal dose used in our study resulted in the reduction of annual hemorrhage
59 5.1 34.45 33.6
GKS using marginal dose as low as 11 Gy was safe without serious radiation-related complication while its efficacy in reduction of annual hemorrhage rate, mass shrinkage and improvement of neurologic symptoms was compatible with that of high marginal dose. Further prospective study is needed for revealing natural history of BSCA and optimal marginal dose.
NA 28.25 (17.5–37.7)
References
0.7–4.7 1.1 (0.03–5.4)
3.12 (0.032–25.9)
1.42 (0.09–4.8) 2.25 (0.08–15.2) 3.1 (0.2–38.8)
a
18 (16–23) 13 (11–18)
2.5 29.2 NA
27.3 11 26.5 35.5 9.4 17.3 25.7 (18–50) NA 18.3 (8–36.7) a
16.1 (8–24) 18 (7–40) 15 (6.5–30) a 16.5 (7.2–33) 12.1 (9–20) a
5. Conclusions
53 (24–78) 49.2 (5–192)
64.8 (9–122.4)
a
38.3 (21–67) 90 (9–166) 64.8 (4–147)
2.1 (0.5–7.9)
Median, NA: not applicable, RS: radiosurgery. a
15 39 Garcia-Munoz et al. (2007) [34] Present study
22 57 95 Kim et al. (2002) [31] Chang et al. (1998) [32] Amin-Hanjani et al. (1998) [33]
125
17 Pollock et al. (2000) [30]
Liu et al. (2005) [28]
47 Kondziolka et al. (1995) [27]
No. of patients Author and year
Table 1 Overall outcome after GKS in other publications.
113
rates compatible with other studies with significantly low risk of complication. This dose risk relationship is also supported by observation of Liscak et al. [29] who reported an increased incidence of collateral edema by marginal dose exceeding 15 Gy. Therefore, our data suggest that adequate marginal dose may be as low as 11 Gy though optimal marginal dose for BSCA cannot be exactly defined. To get more reasonable guidelines, prospective study in a large number of patients with a long-term follow-up is needed.
a
51
a
2.1 (0.2–8.1) 43.2 (4–77)
Mean follow-up (mos)
Mean lesion vol. (cm3 )
a
18 (16–18)
a
32 (20–36)
40.1
59
26 After RS
8.8 (0–2 years) 1.1 (2–6 years) 8.8 (0–2 years) 2.9 (after 2 years) 1.55 1.6 (after 3 years) 22.4 (0–2 years) 4.5 (after 2 years) 10.3 (0–2 years) 3.3 (after 2 years) 7.17 8.1 (0–2 years) 2.4 (after 2 years)
Before RS
16 (10–20)
32 (20–40)
32
Radiation-related complications (%) Radiation-related complications (%) Mean marginal dose (Gy)
Mean maximal dose (Gy)
Annual hemorrage rate (%)
B.S. Kim et al. / Clinical Neurology and Neurosurgery 126 (2014) 110–114
[1] Moran NF, Fish DR, Kitchen N, Shorvon S, Kendall BE, Stevens JM. Supratentorial cavernous haemangiomas and epilepsy: a review of the literature and case series. J Neurol Neurosurg Psychiatry 1999;66:561–8. [2] Di Rocco C, Iannelli A, Tamburrini G. Cavernous angiomas of the brain stem in children. Pediatr Neurosurg 1997;27:92–9. [3] Moriarity JL, Clatterbuck RE, Rigamonti D. The natural history of cavernous malformations. Neurosurg Clin N Am 1999;10:411–7. [4] Cantore G, Missori P, Santoro A. Cavernous angiomas of the brain stem intraaxial anatomical pitfalls and surgical strategies. Surg Neurol 1999;52:84–93, discussion 93–4. [5] Del Curling Jr O, Kelly Jr DL, Elster AD, Craven TE. An analysis of the natural history of cavernous angiomas. J Neurosurg 1991;75:702–8. [6] Kim DS, Park YG, Choi JU, Chung SS, Lee KC. An analysis of the natural history of cavernous malformations. Surg Neurol 1997;48:9–17, discussion 17–8. [7] McCormick WF. The pathology of vascular (“arteriovenous”) malformations. J Neurosurg 1966;24:807–16. [8] Robinson JR, Awad IA, Little JR. Natural history of the cavernous angioma. J Neurosurg 1991;75:709–14. [9] Sage MR, Brophy BP, Sweeney C, Phipps S, Perrett LV, Sandhu A, et al. Cavernous haemangiomas (angiomas) of the brain: clinically significant lesions. Australas Radiol 1993;37:147–55. [10] Fahlbusch R, Strauss C, Huk W, Rockelein G, Kompf D, Ruprecht KW. Surgical removal of pontomesencephalic cavernous hemangiomas. Neurosurgery 1990;26:449–56, discussion 456–7. [11] Fritschi JA, Reulen HJ, Spetzler RF, Zabramski JM. Cavernous malformations of the brain stem. A review of 139 cases. Acta Neurochir (Wien) 1994;130:35–46. [12] Karlsson B, Kihlstrom L, Lindquist C, Ericson K, Steiner L. Radiosurgery for cavernous malformations. J Neurosurg 1998;88:293–7. [13] Kondziolka D, Lunsford LD, Kestle JR. The natural history of cerebral cavernous malformations. J Neurosurg 1995;83:820–4. [14] Simard JM, Garcia-Bengochea F, Ballinger Jr WE, Mickle JP, Quisling RG. Cavernous angioma: a review of 126 collected and 12 new clinical cases. Neurosurgery 1986;18:162–72. [15] Kashiwagi S, van Loveren HR, Tew Jr JM, Wiot JG, Weil SM, Lukin RA. Diagnosis and treatment of vascular brain-stem malformations. J Neurosurg 1990;72:27–34. [16] Porter RW, Detwiler PW, Spetzler RF, Lawton MT, Baskin JJ, Derksen PT, et al. Cavernous malformations of the brainstem: experience with 100 patients. J Neurosurg 1999;90:50–8. [17] Clatterbuck RE, Moriarity JL, Elmaci I, Lee RR, Breiter SN, Rigamonti D. Dynamic nature of cavernous malformations: a prospective magnetic resonance imaging study with volumetric analysis. J Neurosurg 2000;93:981–6. [18] Wang CC, Liu A, Zhang JT, Sun B, Zhao YL. Surgical management of brain-stem cavernous malformations: report of 137 cases. Surg Neurol 2003;59:444–54, discussion 454. [19] Bertalanffy H, Benes L, Miyazawa T, Alberti O, Siegel AM, Sure U. Cerebral cavernomas in the adult. Review of the literature and analysis of 72 surgically treated patients. Neurosurg Rev 2002;25:1–53, discussion 54–5. [20] Hauck EF, Barnett SL, White JA, Samson D. Symptomatic brainstem cavernomas. Neurosurgery 2009;64:61–70, discussion 70–1. [21] Monaco EA, Khan AA, Niranjan A, Kano H, Grandhi R, Kondziolka D, et al. Stereotactic radiosurgery for the treatment of symptomatic brainstem cavernous malformations. Neurosurg Focus 2010;29:E11. [22] Colombo F, Pozza F, Chierego G, Casentini L, De Luca G, Francescon P. Linear accelerator radiosurgery of cerebral arteriovenous malformations: an update. Neurosurgery 1994;34:14–20, discussion 20–1. [23] Friedman WA, Bova FJ, Mendenhall WM. Linear accelerator radiosurgery for arteriovenous malformations: the relationship of size to outcome. J Neurosurg 1995;82:180–9.
114
B.S. Kim et al. / Clinical Neurology and Neurosurgery 126 (2014) 110–114
[24] Pollock BE, Lunsford LD, Kondziolka D, Maitz A, Flickinger JC. Patient outcomes after stereotactic radiosurgery for “operable” arteriovenous malformations. Neurosurgery 1994;35:1–7, discussion 7–8. [25] Pozzati E, Acciarri N, Tognetti F, Marliani F, Giangaspero F. Growth, subsequent bleeding, and de novo appearance of cerebral cavernous angiomas. Neurosurgery 1996;38:662–9, discussion 669–70. [26] Steiner L, Lindquist C, Adler JR, Torner JC, Alves W, Steiner M. Clinical outcome of radiosurgery for cerebral arteriovenous malformations. J Neurosurg 1992;77:1–8. [27] Kondziolka D, Lunsford LD, Flickinger JC, Kestle JR. Reduction of hemorrhage risk after stereotactic radiosurgery for cavernous malformations. J Neurosurg 1995;83:825–31. [28] Liu KD, Chung WY, Wu HM, Shiau CY, Wang LW, Guo WY, et al. Gamma knife surgery for cavernous hemangiomas: an analysis of 125 patients. J Neurosurg 2005;Suppl. 102:81–6. [29] Liscak R, Vladyka V, Simonova G, Vymazal J, Novotny Jr J. Gamma knife radiosurgery of the brain stem cavernomas. Minim Invasive Neurosurg 2000;43:201–7.
[30] Pollock BE, Garces YI, Stafford SL, Foote RL, Schomberg PJ, Link MJ. Stereotactic radiosurgery for cavernous malformations. J Neurosurg 2000;93: 987–91. [31] Kim DG, Choe WJ, Paek SH, Chung HT, Kim IH, Han DH. Radiosurgery of intracranial cavernous malformations. Acta Neurochir (Wien) 2002;144:869–78, discussion 878. [32] Chang SD, Levy RP, Adler Jr JR, Martin DP, Krakovitz PR, Steinberg GK. Stereotactic radiosurgery of angiographically occult vascular malformations: 14-year experience. Neurosurgery 1998;43:213–20, discussion 220–1. [33] Amin-Hanjani S, Ogilvy CS, Candia GJ, Lyons S, Chapman PH. Stereotactic radiosurgery for cavernous malformations: Kjellberg’s experience with proton beam therapy in 98 cases at the Harvard Cyclotron. Neurosurgery 1998;42:1229–36, discussion 1236–8. [34] Garcia-Munoz L, Velasco-Campos F, Lujan-Castilla P, Enriquez-Barrera M, Cervantes-Martinez A, Carrillo-Ruiz J. Radiosurgery in the treatment of brain cavernomas. Experience with 17 lesions treated in 15 patients. Neurochirurgie 2007;53:243–50.
