Clinical Neurology and Neurosurgery 127 (2014) 59–64
Contents lists available at ScienceDirect
Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro
Cavernous sinus cavernous hemangioma: Imaging features and therapeutic effect of Gamma Knife radiosurgery Xiao Anqi a , Shangfu Zhang b , Xiao Jiahe c , You Chao a,∗ a b c
Department of Neurosurgery, West China Hospital, Sichuan University, Chengdu 610041, PR China Department of Pathology, West China Hospital, Sichuan University, Chengdu 610041, PR China Department of Radiology, West China Hospital, Sichuan University, Chengdu 610041, PR China
a r t i c l e
i n f o
Article history: Received 6 August 2014 Received in revised form 2 September 2014 Accepted 26 September 2014 Available online 5 October 2014 Keywords: Cavernous sinus Cavernous hemangioma MRI Gamma Knife radiosurgery
a b s t r a c t Object: To investigate the imaging features of cavernous sinus cavernous hemangioma (CSCH) and evaluate the therapeutic effect of Gamma Knife radiosurgery (GKRS) in treatment of CSCH. Methods and materials: Fifteen patients with CSCH treated by GKRS in our institute, including 6 males and 9 females, age range 20–77 years old, were analyzed retrospectively. Three of them were given craniotomies as the initial therapy. All cases had performed conventional and contrast-enhanced MRI and 5 patients underwent dynamic enhanced MRI preoperatively. In 6 cases, the multi-directional continuous data of axial, coronal and sagittal enhanced MRI were acquired. Three cases performed digital subtraction angiography (DSA) simultaneously. The diagnoses of lesions were determined mainly depending on typical imaging features. In 3 patients, the diagnoses of CSCH were confirmed histopathologically. The radiation dosimetry was done with a goal of conformal and selective coverage of the lesion with a 50% prescription isodose line. The mean marginal dose constituted 13.4 Gy (range 10–16 Gy). After GKRS was performed, all patients were arranged regular clinical and MRI follow-up every 6 months during the first 12 months, and once per year thereafter. Results: On MRI, the lesions were typically demonstrated as iso/hypo-intensities on T1WI and remarkable hyper-intensities on T2WI, and apparent homogeneous enhancement. The phenomenon of dynamic enhancement was found in 11 cases. The progressive enhancing process from heterogeneous to uniform was displayed in the 5 patients performed same-slice dynamic MRI, including imaging characteristics of ‘edge to center’ enhancement in 2 case. In the other 6 cases, the delayed homogeneous enhancement of lesion was observed. Ten patients obtained radiological follow-up results after GKRS. Reviewing the follow-up data of 8 patients during the period of 3–6 months, the lesions were apparently shrunk in 5 patients with shrinkage rate of 20.8–46.8%. In 4 patients with imaging follow-up during the period of 6–12 months, the lesions of 3 patients were remarkably shrunk with shrinkage rate of 53.5–81.7%. Four patients had imaging follow-up data over 12 months, and all their lesion sizes were reduced with shrinkage rate of 19–83.6%. The clinical presentations of all patients after GKRS were followed up during the period of 1–30 months. In 7 of 9 cases with headache, the symptom was improved; in 5 of 6 cases, facial hypesthesia was improved; in 6 of 9 cases with visual impairments, the visions were markedly improved; and in 8 cases with preoperative diplopia, the symptoms were all resolved. Conclusion: Although bright hyper-intensities on T2WI and significant homogeneous enhancement on contrast-enhanced T1WI are considered as typical imaging characteristics of CSCH, the dynamic process of progressive delayed enhancement on contrast-enhanced MR is more persuasive in diagnosis. According to our study, GKRS could be chosen as an effective and safe alternative treatment for CSCH. We consider that using relatively low marginal dose may get better effects in tumor shrinkage and protection of cranial nerves. © 2014 Elsevier B.V. All rights reserved.
∗ Corresponding author at: Department of Neurosurgery, West China Hospital, Sichuan University, 37# Guo Xue Xiang Street, Chengdu, Sichuan, 610041, PR China. Tel.: +86 028 85422490; fax: +86 28 85164009. E-mail address: [email protected] (Y. Chao). http://dx.doi.org/10.1016/j.clineuro.2014.09.025 0303-8467/© 2014 Elsevier B.V. All rights reserved.
60
X. Anqi et al. / Clinical Neurology and Neurosurgery 127 (2014) 59–64
1. Introduction Cavernous sinus cavernous hemangioma (CSCH) is a rare extraaxial vascular tumor, accounting for 2–3% of all cavernous sinus neoplasms [1,2]. It is prone to occur in middle age with predilection sex of female [3]. As a benign and complete capsular vascular neoplasm, the impact on central nervous system is mainly caused by its mass effect. Symptoms are commonly found as headache and cranial nerve involvements such as visual impairment, diplopia or facial hypoesthesia. Currently, the treatment methods include microsurgical resection, interventional embolization, fractionated radiotherapy and Gamma Knife radiosurgery (GKRS) [1]. However, the optimal treatment is still controversial. Though some CSCH cases are reported achieving good results by surgical treatment, the uncontrollable intraoperative bleeding and complex relationship between the lesion and cranial nerves are still the challenge for neurosurgeons [1,4]. In contrast, Gamma Knife treatment is relatively safer compared with craniotomy. The purpose of our study is to track the CSCH patients received Gamma Knife therapy in our institute to evaluate the therapeutic effect of GKRS for CSCH and explore the imaging features of CSCH.
2. Methods and materials We retrospectively analyzed all the 15 patients given GKRS treatment from January 2010 to March 2014, 3 of them were given craniotomies as the initial therapy. Of all, 15 patients, age range 20–77 years old with a mean age of 41 years, include 6 males and 9 females. Clinical manifestations include headache in 9 patients, visual impairment in 9, diplopia in 8 and facial hypoesthesia in 6. Duration of symptoms ranged from 2 months to 5 years. All cases had performed conventional and contrast-enhanced MRI and 5 patients underwent dynamic enhanced MRI preoperatively. MR devices included the 1.5 T (Avanto and Sonata, Siemens; MRT200SP5, Toshiba) and 3.0 T superconducting MR (SIGNA EXCITE, GE; Trio Tim, Siemens; Achieva, Philips). The sequences included T1 Weighted Imaging (T1WI), T2 Weighted Imaging (T2WI), Fluid Attenuated Inversion Recovery (FLAIR), contrastenhanced T1WI with gadopentetate dimeglumine (0.1 mmol/Kg) as the contrast agent, acquiring axial, coronal and sagittal data. The scanning of dynamic enhanced MRI started 10 s after injection of contrast agent with thickness of 3 mm and intervals of 19–20 s, achieving 5–6 dynamic contrast-enhanced imagings perslice. In 6 cases, the multi-directional data of axial, coronal and sagittal enhanced MRI were acquired during the continuous interval time of 90–250 s. Three cases performed digital subtraction angiography (DSA) simultaneously. In 14 patients, the CSCHs were diagnosed based on the typical imaging characteristics, in the other one, the lesion was misdiagnosed as a pituitary tumor. In 3 cases who were given craniotomy initially, 2 of them only underwent partial excision via subdural pterion approach, due to the excessive intraoperative bleeding. The other patient with the initial diagnosis of pituitary tumor was performed transphenoidal approach, and the surgery was timely terminated due to the intraoperative findings of cavernous hemangioma. Eventually, the pathological diagnosis was confirmed as CSCH in all 3 surgical patients (Figs. 1 and 2). Our 15 patients with CSCH were treated with GKRS. The radiation dosimetry was made with a goal of conformal and selective coverage of the lesion with a 50% prescription isodose line by the multi-isocenter technique. The Gamma Knife device was Leksell-C type (Elekta, Stickholm, Sweden). As our treatment plan, all patients were discharged from hospital within 24 h after GKRS, and arranged regular clinical and MRI follow-up
every 6 months during the first 12 months, and once per year thereafter. The tumor dimensions were measured in maximum length of axial, sagittal, and coronal planes and calculated using formula of length × width × height × 0.5. The pre-treatment and posttreatment volumes were compared in order to obtain the change rate of tumor volume. 3. Results On MRI, all the lesions of 15 patients were displayed as welldefined expansion parasellar masses. The involvement of sella in 9 cases, orbital apex in 10,and ipsilateral Meckel cavity and CPA in 8 were observed respectively. The initial tumor volumes ranged from 8.5 to 138 cm3 with an average of 29.3 cm3 . The lesions were demonstrated as iso/hypo-intensities on T1WI and remarkable hyper-intensities on T2WI, and appeared apparent homogeneous enhancement after injection of contrast agent. The phenomenon of dynamic enhancement was found in 11 cases. The progressive enhancing process from heterogeneous to uniform was displayed in the 5 patients who performed same-slice dynamic MRI, including imaging characteristics of ‘edge to center’ enhancement in 2 case. In the other 6 cases, the delay of lesion enhancement was observed during a period of consecutive time on contrast-enhanced MRI. The radiation dosimetry was done with a goal of conformal and selective coverage of the lesion with a 50% prescription isodose line. The mean marginal dose constituted 13.4 Gy (range 10–16 Gy) and the irradiation dose was kept below 8 Gy for region of the optic nerve. Ten patients did radiological follow-up after GKRS (range 3–27 months). Reviewing the follow-up data of 8 patients during the period of 3–6 months, the lesions were apparently shrunk in 5 patients with shrinkage rate of 20.8–46.8%, while the other lesion sizes seldom changed in the others. 4 patients had follow-up imaging during the period of 6–12 months. Apart from one lesion was unchanged in size in 1 patient, the others were remarkably shrunk with shrinkage rate of 53.5–81.7%. Four patients had follow-up data over 12 months, and all their lesion sizes were reduced with shrinkage rate of 19–83.6%. In 1 patient, the change of shrinkage rate was just 1.9% in comparison of post-operative size at 14th month with 27th month. In 2 patients, their lesion sizes appeared few changes on their 6th month and 10th month follow-up imaging, while on 14th month and 18th month follow-up, the tumor shrinkage of 26.2% and 19% were separately observed radiologically. The post-operative symptoms of all the patients were tracked (follow-up time range 1–30 months, with an average of 13 months). In 7 of 9 cases with headache, the symptom was improved; in 5 of 6 cases, facial hypesthesia was improved; in 6 of 9 cases with visual impairments, their visions were markedly improved; and in 8 cases with preoperative diplopia, the symptoms were all resolved (Table 1). 4. Discussion Cavernous sinus cavernous hemangioma (CSCH) is a rare extraaxial lesion, which entirely locates within the cavernous sinus dura, accounting for 2–3% of all cavernous sinus neoplasms [1,2]. Its occurrence has tendency of female and middle-age patients [3]. Consistent with the literatures, the average age of our patients was about 41-years-old, and the majority were female (N = 9). The main symptoms include headache, vision impairment, diplopia and facial hypesthesia. In spite of a vascular malformation, the bleeding is extremely rare seen in its natural course [1,5]. Pathologically,
X. Anqi et al. / Clinical Neurology and Neurosurgery 127 (2014) 59–64
61
Fig. 1. Patient No 4. The preoperative MRI displayed an occupying mass located at the right cavernous sinus with hypointensity on T1WI(A) and bright hyperintensity on T2WI(B). The consecutive multi-directional sequences of enhanced MRI revealed that the enhancement of the lesion was heterogeneous and peripheral initially (C), and then progressively expanded from margin to center into homogeneous enhancement on the delayed imaging of 140 s (D) and 220 s (E) later. After partial resection, histopathology confirmed the lesion as cavernous sinus cavernous hemangioma with characteristics of multiple vascular channels lined by endothelial cells with interval fibrous connective tissue matrix (hematoxylin–eosin staining ×200) (F). Compared with lesion volume before Gamma Knife treatment (G), progressive shrinkage of lesion was observed on contrast-enhanced T1WI at 6 months (H) and 12 months after GKRS.
the CSCH was divided the into 2 types by scholars: Type A has soft texture and high tension with large numbers of thin-walled vascular sinusoids inside, lined by a single layer of endothelium; while Type B has hard texture and abundant connective tissue within thicker sinal walls [6,7]. Based on the success experience of radiotherapy in treating CSCH, Iwai et al. first reported successful treatment of CSCH by Gamma Knife in 1999. Since then, GKRS has become an alternative treatment besides of craniotomy [8–10]. The preoperative radiological diagnosis of CSCH is crucial in determining the treatment plan, therefore understanding its
imaging performance is very important for identifying the CSCH from other lesions in such area. On CT, CSCHs appear as iso/hyperdensities and homogeneous enhancement after administration of contrast agent. Sometimes, the petrosal erosion could be seen simultaneously. On MRI, CSCHs had well-defined boundaries usually without ‘dural tail’ signs. On T1WI, the lesions are seen as uniform hypointense in contrast with brain parenchyma. While on T2WI, they appear brightly hyperintense typically, and intratumoral linear septum-like hypointensities may be found. After the injection of contrast agent, they are significantly and uniformly enhanced. On selective angiography, delayed mild vascular
62
X. Anqi et al. / Clinical Neurology and Neurosurgery 127 (2014) 59–64
Fig. 2. Patient No 7. A mass located at the left cavernous sinus with homogeneous hypointensity on T1WI(A) and uniform hyperintensity on T2WI(B), with typical imaging features of cavernous sinus cavernous hemangioma. On the same-slice dynamic contrast-enhanced T1WI at the interval of 19 s, the progressive enhancement extending from the lesion edge gradually toward the center and finally becoming the uniformity was observed (C–G). The lesion was shrunk at 6 months after Gamma Knife treatment on contrast-enhanced T1WI(H).
blush could be observed in majority of patients. Due to the numerous of sinal cavities and fibrous septums inside the CSCHs, on dynamic contrast-enhanced MRI, the dynamic enhanced process of the lesions could be displayed as heterogeneity gradually expanding into homogeneity with the time delay after contrast injection. This imaging phenomenon of delay enhancement is corresponding to the pathological features of CSCH, and therefore has a good value in identification of CSCH from the others. In our study, dynamic enhancement was observed in 11 cases. As was mentioned in some literatures, dynamic enhanced way of ‘periphery to the center’ was found in 2 of them, similar to imaging findings of hepatic cavernous hemangioma [11,12]. The same-plane dynamic enhancement is valuable for diagnosis by
accurately reflecting the enhancing process of CSCH. Besides of same-plane dynamic imaging, dynamic change of lesion enhancement could also be observed during a period of continuous time on the multi-sequencing MRI, which is also helpful for determining CSCH. In our series, consistent with the literature, the signs of dynamic enhancement were found in 6 cases through viewing their different sequences. Some scholars believe that filling time of tumor enhancement varies along with different subtypes of CSCH, therefore different treatments could be chosen according to delay situation of enhancement time on dynamic contrast-enhanced MRI. CSCHs are commonly misdiagnosed as meningioma, schwannoma or parasellar pituitary tumor. On T2WI, CSCH appears
X. Anqi et al. / Clinical Neurology and Neurosurgery 127 (2014) 59–64
63
Table 1 Clinical data and post-GKRS follow-up of 15 patients with CSCH. Case number
Age (yr)/Sex
Initial tumor volume (cm3 )
Pre-GKRS clinical features
Pre-GKRS treatment
Marginal radiation dose (Gy)
Follow-up period (mo)
Tumor volume change on MRI/%
3–6 mo 1
30/F
42
2
77/F
42.5
3
37/F
138
4
49/M
8.5
5
48/F
28.5
6
33/F
12
7 8
26/F 36/M
9.5 6.5
9
46/M
19
10
20/M
8
a
11
54/F
58
a
12
59/F
14.5
a
13
23/F
25
a
14
24/M
16
a
15
44/M
12
Hch; FHy Dip; Hypo Hch; Dip; Hypo Hch; Hypo Hch; FHy Hch; Hypo Dip Dip
None
13
3
None
10
14
Craniotomy
11
12
Craniotomy
13
27
None
14
6
Craniotomy
12
8
None None
13 16
6 14
Shr/20.8 UC
Dip; Hypo Hch; FHy; Dip; Hypo FHy; Hypo Hch; Dip; Hypo Hch; Hypo FHy; Dip Hch; FHy
None
10
6
Shr/35.6
None
13
18
None
14
10
None
16
7
None
16
6
None
15
3
None
15
2
7–12 mo
12–27 mo
UC Shr/44.5
Shr/ 71.9 Shr/ 53.5
Shr/ 29.5–46.8 Shr/28
Shr/ 81.7
Shr/ 83.5
Shr/75.2
UC
Shr/ 26.2
UC
Clinical response
Shr/ 19
Hch: Stable FHy: Improved Dip: Resolved Hypo: Improved Hch: Improved Dip: Resolved Hypo: Improved Hch: Improved Hypo: Improved Hch: Improved FHy: Improved Hch; Improved Hypo: Improved Dip: Resolved Dip: Resolved Dip: Resolved Hypo: Improved Hch: Improved FHy: Stable Dip: Resolved Hypo: Stable FHy; Improved Hypo: Improved Hch: Improved Dip: Resolved Hypo: Stable Hch: Improved Hypo: Stable FHy; Improved Dip: Resolved Hch: Stable FHy: Improved
M: Male; F: Female; yr: year; Mo: Months; Hch: Headache; FHy: Facial Hypesthesia; Dip: Diplopia; Hypo: Hypopsia; UC: Unchanged; Shr: Shrinkage. a Only clinical follow-up, no MR imaging review temporarily.
significant homogeneous hyperintense, which is apparently brighter than meningioma and pituitary tumor. Though displayed as hyperintense on T2WI, the signals of schwannoma are heterogeneous comparing with CSCH. Moreover, on contrast-enhanced MRI, CSCHs are homogeneously enhanced, also, the progressive expansion of enhancement could be found on dynamic enhanced MRI. Comparing craniotomy with GKRS, the latter is relatively safer in treating CSCH. Based on the limited follow-up information, the effect of GKRS in our patients is optimistic. In 8 cases with radiological follow-up of 3–6 months after GKRS, the positive effects of tumor shrinkage accounted for 62.5% (5/8), and the maximum shrinkage rate reached 46.8%. In 4 cases postoperatively followed during 6–12 months, the tumor shrinkage rate in 3 cases reached over 50% with maximum rate of 81.7%. In 4 cases with follow-up data of 12–27 months, the sizes of lesions were all reduced. In one of them, the postoperative shrinkage rate of lesion increased to 83.6% at 27 months from 81.7% at 12 months with the change rate of only 1.9%. As seen in our results, the CSCHs could be shrunk at 3 months after gamma knife treatment, and the most significant shrinkage occurs within the 1st year, especially from 6 to 12 months. We also noticed that the lesion sizes of 2 patients were unchanged at 6 months and 10 months of their follow-up, tumor shrinkage only reached 26.2% and 19% at 14 months and 18 months separately. Compared to the other patients’, occurrence of the shrinkage was relative smaller and later. Though limited in
the patients and lack of evidence, we speculate that their different sensitivities to Gamma ray might be related to subtypes of CSCHs (Fig. 3). As for symptom improvements after GKRS of our patients, preoperative symptoms of intracranial hypertension and cranial nerves compression, including headaches, facial hypesthesia, visual impairments and diplopia, were relieved and resolved in most patients. In addition, none of postoperative adverse reaction was observed in our patients. Because of the close relationship with the optic nerves and cavernous sinus, high radiation doses are generally restricted, especially in the treatment of large-size CSCH. Some scholars consider the high rate of remarkable tumor shrinkage is related to the high marginal dose (a median dose of 14.5 Gy), which is an optimal dose for rapid volume reduction and subsequent mass effect relief. Also they suggest that using a higher dose than a 14 Gy marginal dose could obtain remarkable early tumor shrinkage of more than 50% of the initial tumor volume [8,13]. However, in our study we noticed that the marginal dose of lesions with significant volume reduction (volume shrinkage over 50%) was only 11.5 Gy on average, even lower than the total average marginal dose. Our result is consistent with findings mentioned in some literatures, that remarkable tumor shrinkage could be obtained using a relatively low marginal dose of 10–12 Gy [14,15]. Till now, the relationship of radiation dose and tumor shrinkage is still under research.
64
X. Anqi et al. / Clinical Neurology and Neurosurgery 127 (2014) 59–64
Fig. 3. Patient No 2. A mass occupying the left cavernous sinus was showed as homogeneous hypointensity on T1WI(A) and hyperintensity on T2WI(B) with significant enhancement on contrast-enhanced T1WI(C), consistent with MRI features of cavernous sinus cavernous hemangioma. After Gamma Knife treatment, the volume of the lesion was reduced at 3 months (D), and apparently shrunk at 14 months (E).
5. Conclusion CSCH is a rare extra-axial vascular tumor, which is commonly misdiagnosed as meningioma or schwannoma on the imaging. Although bright hyperintensities on T2WI and significant homogeneous enhancement on contrast-enhanced T1WI are considered as typical imaging characteristics of CSCH, the dynamic process of progressive delayed enhancement on contrast-enhanced MR is more persuasive in diagnosis. According to our study, GKRS could be chosen as an effective and safe alternative treatment for CSCH. Though further studies need to be done, we consider that using relatively low marginal dose may get better effects in tumor shrinkage and protection of cranial nerves. Competing interests We declare that we have no competing interest. References [1] Bansal S, Suri A, Singh M, Kale SS, Agarwal D, Sharma MS, et al. Cavernous sinus hemangioma: a fourteen year single institution experience. J Clin Neurosci 2014;21(6):968–74. [2] Linskey ME, Sekhar LN. Cavernous sinus hemangiomas: a series, a review, and an hypothesis. Neurosurgery 1992;30:101–8. [3] Goel A. The extradural approach to lesions involving the cavernous sinus. Br J Neurosurg 1997;11:134–8. [4] Goel A, Muzumdar D, Sharma P. Extradural approach for cavernous hemangioma of the cavernous sinus: experience with 13 cases. Neurol Med Chir (Tokyo) 2003;43(3):112–8.
[5] Kida Y, Kobayashi T, Mori Y. Radiosurgery of cavernous hemangiomas in the cavernous sinus. Surg Neurol 2001;56:117–23. [6] Sohn CH, Kim SP, Kim IM, Lee JH, Lee HK, Characteristic MR. imaging findings of cavernous hemangiomas in the cavernous sinus. AJNR Am J Neuroradiol 2003;24(6):1148–51. [7] Shi J, Hang C, Pan Y, Liu C, Zhang Z. Cavernous hemangiomas in the cavernous sinus. Neurosurgery 1999;45:1308–12. [8] Song SW, Kim DG, Chung HT, Paek SH, Han JH, Kim YH, et al. Stereotactic radiosurgery for cavernous sinus hemangiomas. J Neurooncol 2014;118(May (1)):163–8. [9] Chou CW, Wu HM, Huang CI, Chung WY, Guo WY, Shih YH, et al. Gamma Knife surgery for cavernous hemangiomas in the cavernous sinus. Neurosurgery 2010;67(3):611–6. [10] Ivanov P, Chernov M, Hayashi M, Nakaya K, Izawa M, Murata N, et al. Lowdose gamma knife radiosurgery for cavernous sinus hemangioma: report of 3 cases and literature review. Minim Invasive Neurosurg 2008;51(3): 140–6. [11] Jinhu Y, Jianping D, Xin L, Yuanli Z. Dynamic enhancement features of cavernous sinus cavernous hemangiomas on conventional contrast-enhanced MR imaging. AJNR Am J Neuroradiol 2008;29(3):577–81. [12] Jeong MG, Yu JS, Kim KW. Hepatic cavernous hemangioma: temporal peritumoral enhancement during multiphase dynamic MR imaging. Radiology 2000;216:692–7. [13] Wang X, Mei G, Liu X, Dai J, Pan L, Wang E. The role of stereotactic radiosurgery in cavernous sinus hemangiomas: a systematic review and meta-analysis. J Neurooncol 2012;107(2):239–45. [14] Yamamoto M, Kida Y, Fukuoka S, Iwai Y, Jokura H, Akabane A, et al. Gamma Knife radiosurgery for hemangiomas of the cavernous sinus: a seven-institute study in Japan. J Neurosurg 2010;112(4):772–9. [15] Nakamura N, Shin M, Tago M, Terahara A, Kurita H, Nakagawa K, et al. Gamma Knife radiosurgery for cavernous hemangiomas in the cavernous sinus. Report of three cases. J Neurosurg 2002;97 (Suppl. 5):477–80.
Contents lists available at ScienceDirect
Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro
Cavernous sinus cavernous hemangioma: Imaging features and therapeutic effect of Gamma Knife radiosurgery Xiao Anqi a , Shangfu Zhang b , Xiao Jiahe c , You Chao a,∗ a b c
Department of Neurosurgery, West China Hospital, Sichuan University, Chengdu 610041, PR China Department of Pathology, West China Hospital, Sichuan University, Chengdu 610041, PR China Department of Radiology, West China Hospital, Sichuan University, Chengdu 610041, PR China
a r t i c l e
i n f o
Article history: Received 6 August 2014 Received in revised form 2 September 2014 Accepted 26 September 2014 Available online 5 October 2014 Keywords: Cavernous sinus Cavernous hemangioma MRI Gamma Knife radiosurgery
a b s t r a c t Object: To investigate the imaging features of cavernous sinus cavernous hemangioma (CSCH) and evaluate the therapeutic effect of Gamma Knife radiosurgery (GKRS) in treatment of CSCH. Methods and materials: Fifteen patients with CSCH treated by GKRS in our institute, including 6 males and 9 females, age range 20–77 years old, were analyzed retrospectively. Three of them were given craniotomies as the initial therapy. All cases had performed conventional and contrast-enhanced MRI and 5 patients underwent dynamic enhanced MRI preoperatively. In 6 cases, the multi-directional continuous data of axial, coronal and sagittal enhanced MRI were acquired. Three cases performed digital subtraction angiography (DSA) simultaneously. The diagnoses of lesions were determined mainly depending on typical imaging features. In 3 patients, the diagnoses of CSCH were confirmed histopathologically. The radiation dosimetry was done with a goal of conformal and selective coverage of the lesion with a 50% prescription isodose line. The mean marginal dose constituted 13.4 Gy (range 10–16 Gy). After GKRS was performed, all patients were arranged regular clinical and MRI follow-up every 6 months during the first 12 months, and once per year thereafter. Results: On MRI, the lesions were typically demonstrated as iso/hypo-intensities on T1WI and remarkable hyper-intensities on T2WI, and apparent homogeneous enhancement. The phenomenon of dynamic enhancement was found in 11 cases. The progressive enhancing process from heterogeneous to uniform was displayed in the 5 patients performed same-slice dynamic MRI, including imaging characteristics of ‘edge to center’ enhancement in 2 case. In the other 6 cases, the delayed homogeneous enhancement of lesion was observed. Ten patients obtained radiological follow-up results after GKRS. Reviewing the follow-up data of 8 patients during the period of 3–6 months, the lesions were apparently shrunk in 5 patients with shrinkage rate of 20.8–46.8%. In 4 patients with imaging follow-up during the period of 6–12 months, the lesions of 3 patients were remarkably shrunk with shrinkage rate of 53.5–81.7%. Four patients had imaging follow-up data over 12 months, and all their lesion sizes were reduced with shrinkage rate of 19–83.6%. The clinical presentations of all patients after GKRS were followed up during the period of 1–30 months. In 7 of 9 cases with headache, the symptom was improved; in 5 of 6 cases, facial hypesthesia was improved; in 6 of 9 cases with visual impairments, the visions were markedly improved; and in 8 cases with preoperative diplopia, the symptoms were all resolved. Conclusion: Although bright hyper-intensities on T2WI and significant homogeneous enhancement on contrast-enhanced T1WI are considered as typical imaging characteristics of CSCH, the dynamic process of progressive delayed enhancement on contrast-enhanced MR is more persuasive in diagnosis. According to our study, GKRS could be chosen as an effective and safe alternative treatment for CSCH. We consider that using relatively low marginal dose may get better effects in tumor shrinkage and protection of cranial nerves. © 2014 Elsevier B.V. All rights reserved.
∗ Corresponding author at: Department of Neurosurgery, West China Hospital, Sichuan University, 37# Guo Xue Xiang Street, Chengdu, Sichuan, 610041, PR China. Tel.: +86 028 85422490; fax: +86 28 85164009. E-mail address: [email protected] (Y. Chao). http://dx.doi.org/10.1016/j.clineuro.2014.09.025 0303-8467/© 2014 Elsevier B.V. All rights reserved.
60
X. Anqi et al. / Clinical Neurology and Neurosurgery 127 (2014) 59–64
1. Introduction Cavernous sinus cavernous hemangioma (CSCH) is a rare extraaxial vascular tumor, accounting for 2–3% of all cavernous sinus neoplasms [1,2]. It is prone to occur in middle age with predilection sex of female [3]. As a benign and complete capsular vascular neoplasm, the impact on central nervous system is mainly caused by its mass effect. Symptoms are commonly found as headache and cranial nerve involvements such as visual impairment, diplopia or facial hypoesthesia. Currently, the treatment methods include microsurgical resection, interventional embolization, fractionated radiotherapy and Gamma Knife radiosurgery (GKRS) [1]. However, the optimal treatment is still controversial. Though some CSCH cases are reported achieving good results by surgical treatment, the uncontrollable intraoperative bleeding and complex relationship between the lesion and cranial nerves are still the challenge for neurosurgeons [1,4]. In contrast, Gamma Knife treatment is relatively safer compared with craniotomy. The purpose of our study is to track the CSCH patients received Gamma Knife therapy in our institute to evaluate the therapeutic effect of GKRS for CSCH and explore the imaging features of CSCH.
2. Methods and materials We retrospectively analyzed all the 15 patients given GKRS treatment from January 2010 to March 2014, 3 of them were given craniotomies as the initial therapy. Of all, 15 patients, age range 20–77 years old with a mean age of 41 years, include 6 males and 9 females. Clinical manifestations include headache in 9 patients, visual impairment in 9, diplopia in 8 and facial hypoesthesia in 6. Duration of symptoms ranged from 2 months to 5 years. All cases had performed conventional and contrast-enhanced MRI and 5 patients underwent dynamic enhanced MRI preoperatively. MR devices included the 1.5 T (Avanto and Sonata, Siemens; MRT200SP5, Toshiba) and 3.0 T superconducting MR (SIGNA EXCITE, GE; Trio Tim, Siemens; Achieva, Philips). The sequences included T1 Weighted Imaging (T1WI), T2 Weighted Imaging (T2WI), Fluid Attenuated Inversion Recovery (FLAIR), contrastenhanced T1WI with gadopentetate dimeglumine (0.1 mmol/Kg) as the contrast agent, acquiring axial, coronal and sagittal data. The scanning of dynamic enhanced MRI started 10 s after injection of contrast agent with thickness of 3 mm and intervals of 19–20 s, achieving 5–6 dynamic contrast-enhanced imagings perslice. In 6 cases, the multi-directional data of axial, coronal and sagittal enhanced MRI were acquired during the continuous interval time of 90–250 s. Three cases performed digital subtraction angiography (DSA) simultaneously. In 14 patients, the CSCHs were diagnosed based on the typical imaging characteristics, in the other one, the lesion was misdiagnosed as a pituitary tumor. In 3 cases who were given craniotomy initially, 2 of them only underwent partial excision via subdural pterion approach, due to the excessive intraoperative bleeding. The other patient with the initial diagnosis of pituitary tumor was performed transphenoidal approach, and the surgery was timely terminated due to the intraoperative findings of cavernous hemangioma. Eventually, the pathological diagnosis was confirmed as CSCH in all 3 surgical patients (Figs. 1 and 2). Our 15 patients with CSCH were treated with GKRS. The radiation dosimetry was made with a goal of conformal and selective coverage of the lesion with a 50% prescription isodose line by the multi-isocenter technique. The Gamma Knife device was Leksell-C type (Elekta, Stickholm, Sweden). As our treatment plan, all patients were discharged from hospital within 24 h after GKRS, and arranged regular clinical and MRI follow-up
every 6 months during the first 12 months, and once per year thereafter. The tumor dimensions were measured in maximum length of axial, sagittal, and coronal planes and calculated using formula of length × width × height × 0.5. The pre-treatment and posttreatment volumes were compared in order to obtain the change rate of tumor volume. 3. Results On MRI, all the lesions of 15 patients were displayed as welldefined expansion parasellar masses. The involvement of sella in 9 cases, orbital apex in 10,and ipsilateral Meckel cavity and CPA in 8 were observed respectively. The initial tumor volumes ranged from 8.5 to 138 cm3 with an average of 29.3 cm3 . The lesions were demonstrated as iso/hypo-intensities on T1WI and remarkable hyper-intensities on T2WI, and appeared apparent homogeneous enhancement after injection of contrast agent. The phenomenon of dynamic enhancement was found in 11 cases. The progressive enhancing process from heterogeneous to uniform was displayed in the 5 patients who performed same-slice dynamic MRI, including imaging characteristics of ‘edge to center’ enhancement in 2 case. In the other 6 cases, the delay of lesion enhancement was observed during a period of consecutive time on contrast-enhanced MRI. The radiation dosimetry was done with a goal of conformal and selective coverage of the lesion with a 50% prescription isodose line. The mean marginal dose constituted 13.4 Gy (range 10–16 Gy) and the irradiation dose was kept below 8 Gy for region of the optic nerve. Ten patients did radiological follow-up after GKRS (range 3–27 months). Reviewing the follow-up data of 8 patients during the period of 3–6 months, the lesions were apparently shrunk in 5 patients with shrinkage rate of 20.8–46.8%, while the other lesion sizes seldom changed in the others. 4 patients had follow-up imaging during the period of 6–12 months. Apart from one lesion was unchanged in size in 1 patient, the others were remarkably shrunk with shrinkage rate of 53.5–81.7%. Four patients had follow-up data over 12 months, and all their lesion sizes were reduced with shrinkage rate of 19–83.6%. In 1 patient, the change of shrinkage rate was just 1.9% in comparison of post-operative size at 14th month with 27th month. In 2 patients, their lesion sizes appeared few changes on their 6th month and 10th month follow-up imaging, while on 14th month and 18th month follow-up, the tumor shrinkage of 26.2% and 19% were separately observed radiologically. The post-operative symptoms of all the patients were tracked (follow-up time range 1–30 months, with an average of 13 months). In 7 of 9 cases with headache, the symptom was improved; in 5 of 6 cases, facial hypesthesia was improved; in 6 of 9 cases with visual impairments, their visions were markedly improved; and in 8 cases with preoperative diplopia, the symptoms were all resolved (Table 1). 4. Discussion Cavernous sinus cavernous hemangioma (CSCH) is a rare extraaxial lesion, which entirely locates within the cavernous sinus dura, accounting for 2–3% of all cavernous sinus neoplasms [1,2]. Its occurrence has tendency of female and middle-age patients [3]. Consistent with the literatures, the average age of our patients was about 41-years-old, and the majority were female (N = 9). The main symptoms include headache, vision impairment, diplopia and facial hypesthesia. In spite of a vascular malformation, the bleeding is extremely rare seen in its natural course [1,5]. Pathologically,
X. Anqi et al. / Clinical Neurology and Neurosurgery 127 (2014) 59–64
61
Fig. 1. Patient No 4. The preoperative MRI displayed an occupying mass located at the right cavernous sinus with hypointensity on T1WI(A) and bright hyperintensity on T2WI(B). The consecutive multi-directional sequences of enhanced MRI revealed that the enhancement of the lesion was heterogeneous and peripheral initially (C), and then progressively expanded from margin to center into homogeneous enhancement on the delayed imaging of 140 s (D) and 220 s (E) later. After partial resection, histopathology confirmed the lesion as cavernous sinus cavernous hemangioma with characteristics of multiple vascular channels lined by endothelial cells with interval fibrous connective tissue matrix (hematoxylin–eosin staining ×200) (F). Compared with lesion volume before Gamma Knife treatment (G), progressive shrinkage of lesion was observed on contrast-enhanced T1WI at 6 months (H) and 12 months after GKRS.
the CSCH was divided the into 2 types by scholars: Type A has soft texture and high tension with large numbers of thin-walled vascular sinusoids inside, lined by a single layer of endothelium; while Type B has hard texture and abundant connective tissue within thicker sinal walls [6,7]. Based on the success experience of radiotherapy in treating CSCH, Iwai et al. first reported successful treatment of CSCH by Gamma Knife in 1999. Since then, GKRS has become an alternative treatment besides of craniotomy [8–10]. The preoperative radiological diagnosis of CSCH is crucial in determining the treatment plan, therefore understanding its
imaging performance is very important for identifying the CSCH from other lesions in such area. On CT, CSCHs appear as iso/hyperdensities and homogeneous enhancement after administration of contrast agent. Sometimes, the petrosal erosion could be seen simultaneously. On MRI, CSCHs had well-defined boundaries usually without ‘dural tail’ signs. On T1WI, the lesions are seen as uniform hypointense in contrast with brain parenchyma. While on T2WI, they appear brightly hyperintense typically, and intratumoral linear septum-like hypointensities may be found. After the injection of contrast agent, they are significantly and uniformly enhanced. On selective angiography, delayed mild vascular
62
X. Anqi et al. / Clinical Neurology and Neurosurgery 127 (2014) 59–64
Fig. 2. Patient No 7. A mass located at the left cavernous sinus with homogeneous hypointensity on T1WI(A) and uniform hyperintensity on T2WI(B), with typical imaging features of cavernous sinus cavernous hemangioma. On the same-slice dynamic contrast-enhanced T1WI at the interval of 19 s, the progressive enhancement extending from the lesion edge gradually toward the center and finally becoming the uniformity was observed (C–G). The lesion was shrunk at 6 months after Gamma Knife treatment on contrast-enhanced T1WI(H).
blush could be observed in majority of patients. Due to the numerous of sinal cavities and fibrous septums inside the CSCHs, on dynamic contrast-enhanced MRI, the dynamic enhanced process of the lesions could be displayed as heterogeneity gradually expanding into homogeneity with the time delay after contrast injection. This imaging phenomenon of delay enhancement is corresponding to the pathological features of CSCH, and therefore has a good value in identification of CSCH from the others. In our study, dynamic enhancement was observed in 11 cases. As was mentioned in some literatures, dynamic enhanced way of ‘periphery to the center’ was found in 2 of them, similar to imaging findings of hepatic cavernous hemangioma [11,12]. The same-plane dynamic enhancement is valuable for diagnosis by
accurately reflecting the enhancing process of CSCH. Besides of same-plane dynamic imaging, dynamic change of lesion enhancement could also be observed during a period of continuous time on the multi-sequencing MRI, which is also helpful for determining CSCH. In our series, consistent with the literature, the signs of dynamic enhancement were found in 6 cases through viewing their different sequences. Some scholars believe that filling time of tumor enhancement varies along with different subtypes of CSCH, therefore different treatments could be chosen according to delay situation of enhancement time on dynamic contrast-enhanced MRI. CSCHs are commonly misdiagnosed as meningioma, schwannoma or parasellar pituitary tumor. On T2WI, CSCH appears
X. Anqi et al. / Clinical Neurology and Neurosurgery 127 (2014) 59–64
63
Table 1 Clinical data and post-GKRS follow-up of 15 patients with CSCH. Case number
Age (yr)/Sex
Initial tumor volume (cm3 )
Pre-GKRS clinical features
Pre-GKRS treatment
Marginal radiation dose (Gy)
Follow-up period (mo)
Tumor volume change on MRI/%
3–6 mo 1
30/F
42
2
77/F
42.5
3
37/F
138
4
49/M
8.5
5
48/F
28.5
6
33/F
12
7 8
26/F 36/M
9.5 6.5
9
46/M
19
10
20/M
8
a
11
54/F
58
a
12
59/F
14.5
a
13
23/F
25
a
14
24/M
16
a
15
44/M
12
Hch; FHy Dip; Hypo Hch; Dip; Hypo Hch; Hypo Hch; FHy Hch; Hypo Dip Dip
None
13
3
None
10
14
Craniotomy
11
12
Craniotomy
13
27
None
14
6
Craniotomy
12
8
None None
13 16
6 14
Shr/20.8 UC
Dip; Hypo Hch; FHy; Dip; Hypo FHy; Hypo Hch; Dip; Hypo Hch; Hypo FHy; Dip Hch; FHy
None
10
6
Shr/35.6
None
13
18
None
14
10
None
16
7
None
16
6
None
15
3
None
15
2
7–12 mo
12–27 mo
UC Shr/44.5
Shr/ 71.9 Shr/ 53.5
Shr/ 29.5–46.8 Shr/28
Shr/ 81.7
Shr/ 83.5
Shr/75.2
UC
Shr/ 26.2
UC
Clinical response
Shr/ 19
Hch: Stable FHy: Improved Dip: Resolved Hypo: Improved Hch: Improved Dip: Resolved Hypo: Improved Hch: Improved Hypo: Improved Hch: Improved FHy: Improved Hch; Improved Hypo: Improved Dip: Resolved Dip: Resolved Dip: Resolved Hypo: Improved Hch: Improved FHy: Stable Dip: Resolved Hypo: Stable FHy; Improved Hypo: Improved Hch: Improved Dip: Resolved Hypo: Stable Hch: Improved Hypo: Stable FHy; Improved Dip: Resolved Hch: Stable FHy: Improved
M: Male; F: Female; yr: year; Mo: Months; Hch: Headache; FHy: Facial Hypesthesia; Dip: Diplopia; Hypo: Hypopsia; UC: Unchanged; Shr: Shrinkage. a Only clinical follow-up, no MR imaging review temporarily.
significant homogeneous hyperintense, which is apparently brighter than meningioma and pituitary tumor. Though displayed as hyperintense on T2WI, the signals of schwannoma are heterogeneous comparing with CSCH. Moreover, on contrast-enhanced MRI, CSCHs are homogeneously enhanced, also, the progressive expansion of enhancement could be found on dynamic enhanced MRI. Comparing craniotomy with GKRS, the latter is relatively safer in treating CSCH. Based on the limited follow-up information, the effect of GKRS in our patients is optimistic. In 8 cases with radiological follow-up of 3–6 months after GKRS, the positive effects of tumor shrinkage accounted for 62.5% (5/8), and the maximum shrinkage rate reached 46.8%. In 4 cases postoperatively followed during 6–12 months, the tumor shrinkage rate in 3 cases reached over 50% with maximum rate of 81.7%. In 4 cases with follow-up data of 12–27 months, the sizes of lesions were all reduced. In one of them, the postoperative shrinkage rate of lesion increased to 83.6% at 27 months from 81.7% at 12 months with the change rate of only 1.9%. As seen in our results, the CSCHs could be shrunk at 3 months after gamma knife treatment, and the most significant shrinkage occurs within the 1st year, especially from 6 to 12 months. We also noticed that the lesion sizes of 2 patients were unchanged at 6 months and 10 months of their follow-up, tumor shrinkage only reached 26.2% and 19% at 14 months and 18 months separately. Compared to the other patients’, occurrence of the shrinkage was relative smaller and later. Though limited in
the patients and lack of evidence, we speculate that their different sensitivities to Gamma ray might be related to subtypes of CSCHs (Fig. 3). As for symptom improvements after GKRS of our patients, preoperative symptoms of intracranial hypertension and cranial nerves compression, including headaches, facial hypesthesia, visual impairments and diplopia, were relieved and resolved in most patients. In addition, none of postoperative adverse reaction was observed in our patients. Because of the close relationship with the optic nerves and cavernous sinus, high radiation doses are generally restricted, especially in the treatment of large-size CSCH. Some scholars consider the high rate of remarkable tumor shrinkage is related to the high marginal dose (a median dose of 14.5 Gy), which is an optimal dose for rapid volume reduction and subsequent mass effect relief. Also they suggest that using a higher dose than a 14 Gy marginal dose could obtain remarkable early tumor shrinkage of more than 50% of the initial tumor volume [8,13]. However, in our study we noticed that the marginal dose of lesions with significant volume reduction (volume shrinkage over 50%) was only 11.5 Gy on average, even lower than the total average marginal dose. Our result is consistent with findings mentioned in some literatures, that remarkable tumor shrinkage could be obtained using a relatively low marginal dose of 10–12 Gy [14,15]. Till now, the relationship of radiation dose and tumor shrinkage is still under research.
64
X. Anqi et al. / Clinical Neurology and Neurosurgery 127 (2014) 59–64
Fig. 3. Patient No 2. A mass occupying the left cavernous sinus was showed as homogeneous hypointensity on T1WI(A) and hyperintensity on T2WI(B) with significant enhancement on contrast-enhanced T1WI(C), consistent with MRI features of cavernous sinus cavernous hemangioma. After Gamma Knife treatment, the volume of the lesion was reduced at 3 months (D), and apparently shrunk at 14 months (E).
5. Conclusion CSCH is a rare extra-axial vascular tumor, which is commonly misdiagnosed as meningioma or schwannoma on the imaging. Although bright hyperintensities on T2WI and significant homogeneous enhancement on contrast-enhanced T1WI are considered as typical imaging characteristics of CSCH, the dynamic process of progressive delayed enhancement on contrast-enhanced MR is more persuasive in diagnosis. According to our study, GKRS could be chosen as an effective and safe alternative treatment for CSCH. Though further studies need to be done, we consider that using relatively low marginal dose may get better effects in tumor shrinkage and protection of cranial nerves. Competing interests We declare that we have no competing interest. References [1] Bansal S, Suri A, Singh M, Kale SS, Agarwal D, Sharma MS, et al. Cavernous sinus hemangioma: a fourteen year single institution experience. J Clin Neurosci 2014;21(6):968–74. [2] Linskey ME, Sekhar LN. Cavernous sinus hemangiomas: a series, a review, and an hypothesis. Neurosurgery 1992;30:101–8. [3] Goel A. The extradural approach to lesions involving the cavernous sinus. Br J Neurosurg 1997;11:134–8. [4] Goel A, Muzumdar D, Sharma P. Extradural approach for cavernous hemangioma of the cavernous sinus: experience with 13 cases. Neurol Med Chir (Tokyo) 2003;43(3):112–8.
[5] Kida Y, Kobayashi T, Mori Y. Radiosurgery of cavernous hemangiomas in the cavernous sinus. Surg Neurol 2001;56:117–23. [6] Sohn CH, Kim SP, Kim IM, Lee JH, Lee HK, Characteristic MR. imaging findings of cavernous hemangiomas in the cavernous sinus. AJNR Am J Neuroradiol 2003;24(6):1148–51. [7] Shi J, Hang C, Pan Y, Liu C, Zhang Z. Cavernous hemangiomas in the cavernous sinus. Neurosurgery 1999;45:1308–12. [8] Song SW, Kim DG, Chung HT, Paek SH, Han JH, Kim YH, et al. Stereotactic radiosurgery for cavernous sinus hemangiomas. J Neurooncol 2014;118(May (1)):163–8. [9] Chou CW, Wu HM, Huang CI, Chung WY, Guo WY, Shih YH, et al. Gamma Knife surgery for cavernous hemangiomas in the cavernous sinus. Neurosurgery 2010;67(3):611–6. [10] Ivanov P, Chernov M, Hayashi M, Nakaya K, Izawa M, Murata N, et al. Lowdose gamma knife radiosurgery for cavernous sinus hemangioma: report of 3 cases and literature review. Minim Invasive Neurosurg 2008;51(3): 140–6. [11] Jinhu Y, Jianping D, Xin L, Yuanli Z. Dynamic enhancement features of cavernous sinus cavernous hemangiomas on conventional contrast-enhanced MR imaging. AJNR Am J Neuroradiol 2008;29(3):577–81. [12] Jeong MG, Yu JS, Kim KW. Hepatic cavernous hemangioma: temporal peritumoral enhancement during multiphase dynamic MR imaging. Radiology 2000;216:692–7. [13] Wang X, Mei G, Liu X, Dai J, Pan L, Wang E. The role of stereotactic radiosurgery in cavernous sinus hemangiomas: a systematic review and meta-analysis. J Neurooncol 2012;107(2):239–45. [14] Yamamoto M, Kida Y, Fukuoka S, Iwai Y, Jokura H, Akabane A, et al. Gamma Knife radiosurgery for hemangiomas of the cavernous sinus: a seven-institute study in Japan. J Neurosurg 2010;112(4):772–9. [15] Nakamura N, Shin M, Tago M, Terahara A, Kurita H, Nakagawa K, et al. Gamma Knife radiosurgery for cavernous hemangiomas in the cavernous sinus. Report of three cases. J Neurosurg 2002;97 (Suppl. 5):477–80.
