If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Communications should be addressed to: Dr. Katarzyna Kotulska; Department of Science; Department of Neurology and Epileptology; The Children's Memorial Health Institute; Al. Dzieci Polskich 20; 04-730 Warsaw; Poland.
Subependymal giant cell astrocytoma is a brain tumor associated with tuberous sclerosis complex. There are two treatment options for subependymal giant cell astrocytomas: surgery or mammalian target of rapamycin inhibitor. The analysis of outcome of subependymal giant cell astrocytoma surgery may help characterize the patients who may benefit from pharmacotherapy.
Sixty-four subependymal giant cell astrocytoma surgeries in 57 tuberous sclerosis complex patients with at least a 12-month follow-up were included in the study. The tumor size, age of the patients, mutation in the TSC1 or TSC2 gene, indication for the surgery, and postsurgical complications were analyzed.
The mean age of patients at surgery was 9.7 years. Mean follow-up after surgery was 63.7 months. Thirty-seven (57.8%) tumors were symptomatic and 27 (42.2%) were asymptomatic. Patients with TSC2 mutations developed subependymal giant cell astrocytoma at a significantly younger age than individuals with TSC1 mutations. Four patients (6.2% of all surgeries) died after surgery. Surgery-related complications were reported in 0%, 46%, 83%, 81%, and 67% of patients with tumors <2 cm, between 2 and 3 cm, between 3 and 4 cm, >4 cm, and bilateral subependymal giant cell astrocytomas, respectively, and were most common in children younger than 3 years of age. The most common complications included hemiparesis, hydrocephalus, hematoma, and cognitive decline.
Our study indicates that subependymal giant cell astrocytoma surgery is associated with significant risk in individuals with bilateral subependymal giant cell astrocytomas, tumors bigger than 2 cm, and in children younger than 3 years of age. Therefore, tuberous sclerosis complex patients should be thoroughly screened for subependymal giant cell astrocytoma growth, and early treatment should be considered in selected patients.
There are no established predictors of prognosis of SEGA surgery, and there is no consensus on the best timing for surgery. There are also no biomarkers to identify the best candidates for pharmacotherapy with mTOR inhibitor or elective surgery.
The aim of this study was to analyze our large cohort of TSC patients who underwent SEGA surgery who were followed at the Children's Memorial Health Institute, Warsaw, to establish the safety and efficacy of surgical treatment of SEGA in TSC patients.
Materials and Methods
The study was approved by The Children's Memorial Health Institute Ethics Committee. The records of patients with a history of SEGA surgery who were followed at the Department of Neurology and Epileptology, the Children's Memorial Health Institute, Warsaw, between 2000 and 2012 were retrospectively reviewed. The inclusion criteria were histologically proven diagnosis of SEGA, clinically definite TSC based on Roach's criteria,
presurgical neuroimaging and surgery report available, and at least a 12-month follow-up after surgery with full neurological examination and brain magnetic resonance imaging/computed tomography performed. The patients who died in the first year after surgery and the relation of death to surgery assessed as probable were also included in the analysis.
The analyzed data included patient demographics; mutational analysis results, if available; the presenting symptoms; size of the tumor; surgical approach and the extent of surgery; any adverse events; results of follow-up neurological examination; and neuroimaging studies. Mutational analysis was performed in either of two laboratories: Genetics Laboratory, Translational Medicine Division, Brigham and Women Hospital, Boston, MA, or the Institute of Medical Genetics, Cardiff University School of Medicine, Cardiff, Great Britain.
Results were analyzed statistically using two-proportion Z-test with significance set at P ≤ 0.05.
Fifty-seven patients with a history of SEGA surgery were included in the study. All patients underwent SEGA surgery between 1994 and 2011. Forty-four patients were operated on at the Department of Neurosurgery, The Children's Memorial Health Institute; eight at the Department of Paediatric Neurosurgery, Silesian Medical University; and five in other neurosurgical departments in Poland. Altogether, 64 surgeries were analyzed because seven patients had two separate SEGA surgeries: two because of regrowth of partially removed tumor and five because of new contralateral tumor development.
There were 25 (43.9%) females and 32 (56.1%) males. Mean follow-up was 63.7 months (median 60 months), ranging from 12 to 168 months, with the exception of four patients who died in association with surgery. Three deaths occurred within 7 days after surgery and one 3 months after surgery.
The mean age of the patients at surgery was 9.7 years (range 6 weeks to 26 years). Six (9.4%) patients underwent surgery before the age of 3 years, 31 (48.4%) at between age 3 and 10 years, 23 (35.9%) between age 10 and 16 years, and four (6.2%) at older than age 16 years.
Mutational analysis was performed in 37 patients. The TSC1 mutation was found in 10 patients (27.0%), TSC2 mutation in 21 patients (56.7%, including five [13.5%] with a large deletion affecting the PKD1 gene), and no mutation was identified in six patients (16.2%). The patient age distribution at first SEGA surgery among the patients with TSC1, TSC2, and TSC2/PKD1 mutations and patients with no mutation identified is shown in Figure 2. Patients with a TSC2 mutation required surgery at a younger age (average 6.8 years) than did patients with a TSC1 mutation (12.9 years, P = 0.01) and patients with no mutation identified (11.3 years; P = 0.02). Patients with a TSC2/PKD1 mutation underwent surgery earlier (average age 3.6 years) than did patients with other TSC2 mutations (average age 7.8 years; P < 0.05), a TSC1 mutation (12.9 years, P < 0.05), and patients with no mutation identified (11.3 years; P < 0.05).
Thirty-seven (57.8%) tumors were operated on because the patients developed clinical symptoms, and 27 (42.2%) tumors were removed because of documented tumor growth and/or hydrocephalus revealed on neuroimaging. Six patients had shunts implanted before SEGA surgery. In 13 (22.8%) patients, symptomatic SEGA was the first sign of TSC. Two of the patients were infants, and nine were older children with very mild presentations of other symptoms of TSC. In five patients, mutational analysis was undertaken; in three of these patients, it disclosed a TSC1 mutation, in one patient a TSC2 mutation, and in one patient no mutation was identified. Symptoms of SEGA included headache, nausea and vomiting, visual disturbances and/or visual loss, hemiparesis, seizures, cognitive functions deterioration, and syncope (Table 1).
Table 1Signs and symptoms of subependymal giant cell astrocytoma development in tuberous sclerosis complex patients
Incidence; n (%)
Nausea and/or vomiting
New seizures or increased number of seizures
Thirty-seven (57.8%) tumors were removed when clinical symptoms were present.
The maximum diameter of SEGA was <2 cm in 13 (20%) patients, between 2 and 3 cm in 13 (20%) patients, between 3 and 4 cm in 12 (18.7%) patients, and >4 cm in 26 (40.6%) patients. Nine patients (9/57; 15.8%) presented with bilateral (“mirror”) tumors simultaneously and in all of them both tumors were removed at the same surgery.
In 58 (90.6%) tumors, gross total resection (Fig 1B) was performed; no regrowth of tumor was observed in this group. Six tumors (9.4%) were removed subtotally; five (83.3%) of them regrew in 3 to 12 months requiring either second surgery (two SEGAs) or mTOR inhibitor (three patients). One subtotally removed tumor remained stable for 9 years. All partially removed SEGAs exceeded 2 cm in diameter.
In most cases (56 surgeries, 87%) the transcallosal approach was used to remove SEGA. Five tumors (7.8%) were removed via transcortical approach and three (4.7%) via endoscopy.
Surgery-related complications were observed in 37 (57.8%) patients. Some patients suffered from more than one adverse event, and altogether 48 complications were reported. Twenty-seven surgeries were uneventful.
Four patients (6.2% of all surgeries) died within 1 year after surgery. Three patients died within 7 days postoperatively: one because of drug-resistant status epilepticus, one because of massive intracerebral bleeding, and one because of cardiac arrest. One patient died 3 months after partial tumor resection because of tumor regrowth and acute hydrocephalus. The complications observed during the first year postoperatively included hydrocephalus requiring shunt implantation, hemiparesis, intracranial bleeding, cognitive decline, meningitis, diabetes insipidus, seizures, precocious puberty, and neuropathic headache. Most of these complications resolved without sequelae within 1 year. After 12 months, the persistent complications included hemiparesis, cognitive decline, precocious puberty, and neuropathic headache. Table 2 presents the incidence of surgery-related complications. There was statistically significant correlation between the risk of complication and the size of SEGA. Patients operated on for tumors smaller than 2 cm, regardless the presence of symptoms, did not experience any surgery-related complications. Complications were noted in four (30.8%) patients with tumors between 2 and 3 cm in diameter, in eight (66.7%) patients with tumors between 3 and 4 cm, in 19 (73%) patients with tumors bigger than 4 cm, and in six patients (67%) with bilateral SEGAs (Fig 3). Complications were also more frequently seen in young children younger than 3 years of age (in five of six tumors; 83.3%) than in children between 3 and 10 years (in 16 of 31 tumors; 51.6%), children between 10 and 16 years (in 15 of 23 tumors, 65.2%) and older patients (in one of four tumors, 25%), but the differences were not statistically significant.
Table 2Complications related to surgery in first 12 months and persisting beyond the first year after surgery in tuberous sclerosis complex patients operated on for SEGA
Adverse events were more frequent in patients operated on for symptomatic SEGA (in 26 of 37, 70.2%) than with asymptomatic tumors (9 of 27, 33.3%; P < 0.05).
Complications were observed in 33 (58.9%) and 4 (80%) patients operated via transcallosal and transcortical approach, respectively. Patients operated by means of endoscopic approach did not experience complications, but none of them presented with SEGA bigger than 2 cm.
Fifteen patients (26.3%) developed contralateral SEGA in 6 to 120 months after the first surgery and 10 of them required either second surgical intervention (three SEGAs) or treatment with mTOR inhibitor (seven patients).
Recent clinical research that showed efficacy of mTOR inhibitors in the treatment of SEGA associated with TSC has opened a discussion on benefits and risks of SEGA surgery and pharmacological therapy. It has also exposed the urgent need of analyses of surgical treatment outcomes identification of risk factors for poor prognosis associated with surgery. Our cohort of TSC patients operated on for SEGA is the largest published. Moreover, most (77%) of surgeries we analyzed were performed by the same team of neurosurgeons; this significantly reduced the expertise-related bias.
Our study indicates that surgical treatment of SEGA >3 cm is burdened with more than 67% risk of surgery-related complication. Surgery on tumors >4 cm was associated with a 73% risk of adverse events. Bilateral SEGAs, regardless their size, were associated with 67% risk of complications after surgery. In patients undergoing surgery for SEGAs smaller than 2 cm, no complications were observed. This is in accordance with other previous study of Cuccia et al.,
reported that symptomatic SEGA is associated with significantly higher risk of surgery-related complications than asymptomatic tumors removed because of documented growth on serial neuroimaging studies. Currently, brain magnetic resonance imaging or computed tomography is recommended every 2 years in TSC patients,
and unpredictable until now, research focused on risk factors for SEGA development and prognostic factors of SEGA growth are urgently needed to identify the patients who may benefit from early surgery and those in whom “watch and wait” approach could be more advantageous.
The surgery-related complications reported in our study included persisting hydrocephalus requiring shunt implantation, focal deficits, intracranial bleeding, cognitive decline, meningitis, diabetes insipidus, and seizures. Most of them were temporary and were not observed beyond 12 months after surgery. However, focal deficits, precocious puberty, and neuropathic headache persisted in some patients for a longer time.
The incidence of surgery-related infections was low (3.1%) in comparison to the data reported by Sun et al.
They reviewed medical data of 47 TSC patients undergoing surgery for SEGA by analyzing three large US national health care claims databases. In their study, during the first postsurgery year, 48.9% of patients developed postoperative complications, including 6.4% of postoperative infections, 17.0% subdural empyemas, and 2.1% epidural abscesses. However, the main limitation of their study is that it was based only on data in a database. The size of SEGA and the surgical approach were not known. Moreover, patients had surgery in many different centers, each likely having different experience in surgery of SEGA.
In our patients with SEGA, TSC1 mutation was identified in 27% of all patients who underwent mutational analysis. TSC2 mutation was found in 56.7% of patients. This suggests that SEGAs develop significantly more frequently in individuals with TSC1 mutations than previously reported (15%).
In 11 of our patients, symptomatic SEGA was the first symptom of TSC, and among those who had mutational analysis done, the TSC1 mutation accounted for three of five patients. This might suggest that TSC is misdiagnosed in some patients with a TSC1 mutation and mild clinical presentation of the disease.
We showed that a TSC2 mutation is associated with SEGA development at a younger age than with a TSC1 mutation. Therefore, we recommend more frequent neuroimaging examinations in children with TSC2 mutations. Large genomic mutations affecting both TSC2 and PKD1 genes are rare and account for 2% to 3% of all TSC patients,
but in our cohort TSC2/PKD1 mutations were found in 13.5% of patients. Moreover, in this group of patients, SEGA developed at significantly younger age than in individuals with other TSC2 or TSC1 mutations. The prevalence of SEGA among patients with mutations in TSC1, TSC2, or TSC2/PKD1 and the mechanisms underlying the differences in SEGA development risk between the groups requires further studies. Nevertheless, our results indicate that patients with polycystic kidneys should be screened for SEGA from birth.
We observed more surgery-related complications in patients younger than 3 years of age than in older children. Goh et al.
reported more complication in patients older than age 11 years; however, they analyzed 11 patients only. Poor outcome in younger children observed in our study may be at least partly associated with more rapid SEGA growth or overall more severe TSC presentation.
The risk of SEGA regrowth in our study was high in patients with partial tumor removal. Partial removal was done only in patients with tumors larger than 2 cm. In all patients in whom gross total SEGA removal was achieved, surgery appeared to be curative. In some individuals with large or bilateral SEGAs, the induction therapy with mTOR inhibitor might enable subsequent complete surgery; however, such an approach requires clinical studies.
reported a high rate of postsurgery diagnosis of SEGA (34%) and a high rate of second SEGA surgery in the first 12 months after first surgery (12%). However, it is not known whether the second diagnosis was because of SEGA regrowth or appearance of a new SEGA. It is also not known what the proportion of gross total and partial first surgeries was. In our cohort, 15 patients (26.3%) developed contralateral SEGA 6 to 120 months after first surgery; 10 of them required either second surgical intervention (three SEGAs) or treatment with mTOR inhibitor (seven patients). Given the frequency of bilateral tumors at first surgery (nine patients, 15.8%), the overall incidence of bilateral SEGA in our study was 42.1%. This is more than reported by Pascual-Castroviejo,
and may suggest that some yet unknown factors predispose TSC patients to SEGA development.
In conclusion, we showed that SEGA surgery is safe and effective in patients presenting with tumors smaller than 2 cm. Therefore, periodic brain neuroimaging is recommended for TSC patients to identify growing SEGAs. Patients with TSC2 mutations, and especially with TSC2/PKD1 mutations, develop SEGA earlier in childhood than do patients with a TSC1 mutation and should be screened for SEGA from birth. Partial removal of SEGA is associated with a high risk of tumor regrowth, and these patients should be thoroughly followed by an experienced neurologist and have neuroimaging done more frequently. The risk factors for poor outcome of SEGA surgery identified in our study include: age <3 years, bilateral tumors, tumor size exceeding 2 cm, symptomatic SEGA, and partial SEGA surgery. In patients presenting any of these features, apart from acute hydrocephalus, pharmacotherapy with an mTOR inhibitor should be considered as a treatment option.
Courage is what it takes to stand up and speak; courage is also what it takes to sit down and listen.
We thank Lauren D'Angelo for her excellent and kind technical support.
Tuberous sclerosis complex: advances in diagnosis, genetics, and management.