Risk Factors for Postoperative Hydrocephalus Following Subependymal Giant Cell Astrocytoma Resection: A Study of Under 18-Year-Old Patients in China

Background: Hydrocephalus may occur after subependymal giant cell astrocytoma (SEGA) resection. In existing literatures, SEGA almost always occurred in patients with tuberous sclerosis complex (TSC), however, many SEGA also occurred alone in our Chinese pediatric patients. Objective: To discuss the risk factors of postoperative hydrocephalus following SEGA resection and the relationship between SEGA and TSC in Chinese children. Materials and methods: A total of 35 children ( ≤ 18-year-old) who underwent SEGA resection were selected. From 3 months postoperatively until December 2020 all patients received telephone or clinical follow-up. Related risk factors were rst screened by univariate analysis and then analyzed by multivariate logistic regression. Results: The ratio of males to females was 3:2 and the mean age was 11.6 years. Twenty cases were associated with TSC and 15 were not. The mean maximum diameter of the SEGA for patients with and without associated TSC was 49.7mm and 30.5mm, respectively (Z=-3.293, P=0.001). Twenty-eight patients had preoperative hydrocephalus. Sixteen patients developed postoperative hydrocephalus, and amongst these, 2 did not have hydrocephalus before surgery. Multivariate analysis showed that association with TSC [odds ratio (OR), 18.81, P=0.048] and tumor resection rate (OR, 0.042, P=0.025) were independent risk factors for postoperative hydrocephalus.


Introduction
Subependymal giant cell astrocytoma (SEGA), a tumor arising in the wall of the lateral ventricles adjacent to the foramen of Monro, accounts for about 1-2% of all pediatric brain tumors. [1,2] It typically occurs during the rst two decades of life and infrequently arises after the age of 20-25 years. [3,4] The International Agency for Research on Cancer of World Health Organization Classi cation of Tumors (revised 4th edition) classi ed SEGA as grade I tumor in 2016. [5] SEGA is composed of different cell lineages and is not purely astrocytic in nature.
[6] SEGA has been considered to be a pathognomonic nding of tuberous sclerosis complex (TSC), and its incidence in patients with TSC varies from 5-25%. [4,7,8] TSC is a neurocutaneous syndrome which mainly involves the central nervous system (CNS) where SEGA, subependymal nodules (SEN), and cortical tubers may be present. In 2013, the experts of an international committee de ned the imaging features of SEGA as a lesion at the caudothalamic groove larger than 1 cm in any direction or subependymal lesions which showed serial growth on consecutive imaging in any position. [9] Because of their location and growth potential, SEGA can cause increased intracranial pressure, obstructive hydrocephalus, focal neurologic de cits, even death. [10] Symptomatic lesions tend to have greater morbidity, [11] in which cases surgical resection is the preferred modality of treatment. [12] However, surgery can have associated postoperative complications, such as intracranial infection, seizures and hydrocephalus. Many patients who develop hydrocephalus after surgery require a second operation to relieve symptoms. The previous studies reported that SEGA occurs almost exclusively in patients with TSC, [1,2] however in our study cohort, nearly one-third of Chinese pediatric patients did not have associated TSC. The purpose of this study was to investigate the risk factors for hydrocephalus after SEGA resection and the relationship between SEGA and TSC in Chinese children younger than 18 years old.

Patients
We included 35 children (≤ 18 years) who underwent SEGA resection at our institutions from January 2002 to December 2020. All patients with TSC received con rmation through genetic testing. Clinical data, including age, sex, onset symptoms, radiological results, operation ndings, surgical approach, postoperative complications and related treatments, were collected from the hospital database. In order to evaluate the extent of the resection and for future comparisons, all patients underwent a cranial magnetic resonance imaging (MRI) enhanced scan 24-48 hours postoperatively.

Statistical methods
Statistical analysis was performed with SPSS 25.0. Continuous variables were analyzed using the nonparametric Mann-Whitney U-test, and categorical variables were analyzed using chi-square test. Univariate analyses were performed to examine the relationship between hydrocephalus postoperatively and prognostic factors. Variables were included in a logistic regression model if their P value in the univariate analysis was < .1. The odds ratio (OR) and 95% con dence interval (CI) were calculated. P < .05 was considered statistically signi cant.

Results
In total, 35 patients (21 males, 14 females) were eligible for the nal analysis in the present study. The patients' age ranged from 5 to 18 years (mean 11.6 years), the disease duration ranged from 2 weeks to 168 months (mean 27.6 months), and the follow-up period ranged from 3 to 168 months (mean 78.2 months). A total of 20 patients had associated TSC and 15 patients had no associated TSC. All data regarding age, sex, disease duration, onset symptoms, tumor location, tumor size and tumor components are presented in Table 1. Sixteen patients developed hydrocephalus after SEGA resection during the follow-up period. In order to relieve the hydrocephalus of these patients, ventriculoperitoneal shunt (VPS) was performed in 7 cases, Ommaya reservoir insertion was performed in 6 cases, ventricularostomy was performed in 2 cases, and Ommaya reservoir and ventricularostomy insertion was performed in one case. All data regarding age, sex, association with TSC, preoperative hydrocephalus, tumor location, tumor size, the extent of resection and postoperative complications are presented in Table 2 There were no signi cant differences in age, sex, disease duration, onset symptoms, preoperative hydrocephalus, tumor location or tumor components between patients with and without associated TSC (Table 1). There were signi cant differences in tumor size (patients with associated TSC: 49.7mm vs. patients without associated TSC: 30.5mm, P = 0.001). Furthermore, the maximum diameter of SEGA was larger in patients with associated TSC than without associated TSC. There were no signi cant differences in age, sex, preoperative hydrocephalus or postoperative intracranial infection between patients with and without hydrocephalus after SEGA resection. The presence or absence of associated TSC, tumor size, tumor location and the extent of resection were included in a logistic regression model (Table 2). Logistic regression analysis showed that GTR (OR, 0.042; 95% CI, 0.003-0.670; P = 0.025) was a protective factor for postoperative hydrocephalus after SEGA resection, while association with TSC was a risk factor (OR, 18.814; 95% CI, 1.020-347.120; P = 0.048) ( Table 3).

Discussion
SEGA is a clinically infrequent benign brain tumor and the incidence is slightly lower in females than in males which is consistent with our research. [13] The mean age at diagnosis was 11.6 years in our study.
[18] However, in our cohort, 20 cases of SEGA had associated TSC, and 15 cases of SEGA occurred alone.
This difference from previous studies [1,[14][15][16][17] might be due to the different ethnic groups in China. We analyzed the age, sex, disease duration and other characteristics of patients with and without associated TSC and found that there were signi cant differences in the maximum tumor diameter between the two groups. The maximum diameter of SEGA with associated TSC was larger than that of patients without associated TSC. SEGA tends to occur in the area of the lateral ventricle near the foramen of Monro, which easily causes hydrocephalus. [19] Associated morbidity and mortality are sometimes due to obstructive hydrocephalus rather than tumor growth. [20] Tumor resection is the main treatment method, [12] however, there could be recurrence or new discovery of hydrocephalus after resection in some patients, and severe hydrocephalus can be fatal. [6,11,21,22] In our study of 35 patients, 28 had preoperative hydrocephalus, amongst these 14 had relief after resection while 14 had recurrence. Sixteen patients in our study had postoperative hydrocephalus, amongst which 2 cases were newly discovered. For the 16 patients with postoperative hydrocephalus, a second operation was carried out. All patients effectively had relief and no obvious abnormalities were found during the follow-up period.
We analyze the risk factors of postoperative hydrocephalus and found that association with TSC and extent of resection are independent risk factors. Patients with associated TSC are more likely to develop hydrocephalus after resection than patients without associated TSC. The reason might be that for patients with associated TSC, in addition to SEGA, there could have SEN or other neurostructural disorders. Compared with SEGA alone, patients with concurrent SEN or other disorders are more likely to have obstructive hydrocephalus. As for the extent of surgical resection, GTR is less prone to develop hydrocephalus after resection than STR. In general, total resection should be the goal of surgery, but for some tumors especially those with large volume and unclear boundaries with adjacent brain tissue, it may be di cult to resect the entire tumor by surgery, and there is a greater probability of hydrocephalus after surgery. For patients with these risk factors, VPS or Ommaya reservoir insertion could be considered during the rst SEGA resection in order to avoid or reduce injury caused by a second operation. For patients with associated TSC, the mechanistic target of rapamycin (mTOR) inhibitors everolimus could be considered, which was approved effectively by the Food and Drug Administration in 2010. [23] Conclusion The incidence of SEGA is slightly more predominant in males than in females. The lesion tends to be located in the lateral ventricle near the foramen of Monro. SEGA could have associated TSC or occur alone. The maximum diameter of SEGA with associated TSC is larger than that without associated TSC. Hydrocephalus is a common onset symptom of SEGA, which can also occur after tumor resection. Some patients need further VPS or Ommaya reservoir insertion. Association with TSC and STR could place patients at high risk for hydrocephalus after SEGA resection. For these patients, VPS or Ommaya reservoir insertion could be considered in conjunction with SEGA resection.

Declarations Funding
The research was supported by Yunnan Fundamental Research Projects (202001AY070001-247).

Con icts of interest/Competing interests
The authors have no con icts of interest to declare that are relevant to the content of this article.

Ethics approval
This retrospective chart review study involving human participants was in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. Ethics Committee of Xinhua Hospital A liated to Shanghai Jiaotong University School of Medicine approved this study (Approval No. XHEC-D-2020-052).

Consent to publish
The authors a rm that human research participants provided informed consent for publication of the images in Figures 1e-f.

Data availability
The datasets generated during and analysed during the current study are available from the corresponding author on reasonable request.

Author contributions
All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Fangjie Shen, Jia Wang, Huatao Niu and Xiaoqiang Wang. Fangjie Shen, Gang cui and Xu kang collected and analyzed the raw clinical data. The rst draft of the manuscript was written by Fangjie Shen, Jia Wang and Loren Skudder-Hill, all authors commented on previous versions of the manuscript. All authors read and approved the nal manuscript.   Postoperative T1 enhancement showed that no obvious abnormal enhancement was observed in the operative area after tumor resection. c: MRI showed hydrocephalus 2 months after surgery. d: The tumor consisted of polygonal cells and ganglionic-like cells with fascicles (Hematoxylin and eosin stain, ×400 magni cation). e: Shagreen patch could be seen on left back, hypopigmented macule could be seen on right back. f: Angio bromas could be seen on the face.