Correlation of anatomical involvement patterns of insular gliomas with subnetworks of the limbic system.

OBJECTIVE: Gliomas frequently involve the insula both primarily and secondarily by invasion. Despite the high connectivity of the human insula, gliomas do not spread randomly to or from the insula but follow stereotypical anatomical involvement patterns. In the majority of cases, these patterns correspond to the intrinsic connectivity of the limbic system, except for tumors with aggressive biology. On the basis of these observations, the authors hypothesized that these different involvement patterns may be correlated with distinct outcomes and analyzed these correlations in an institutional cohort. METHODS: Fifty-nine patients who had undergone surgery for insular diffuse gliomas and had complete demographic, pre- and postoperative imaging, pathology, molecular genetics, and clinical follow-up data were included in the analysis (median age 37 years, range 21-71 years, M/F ratio 1.68). Patients with gliomatosis and those with only minor involvement of the insula were excluded. The presence of T2-hyperintense tumor infiltration was evaluated in 12 anatomical structures. Hierarchical biclustering was used to identify co-involved structures, and the findings were correlated with established functional anatomy knowledge. Overall survival was evaluated using Kaplan-Meier and Cox proportional hazards regression analysis (17 parameters). RESULTS: The tumors involved the anterior insula (98.3%), posterior insula (67.8%), temporal operculum (47.5%), amygdala (42.4%), frontal operculum (40.7%), temporal pole (39%), parolfactory area (35.6%), hypothalamus (23.7%), hippocampus (16.9%), thalamus (6.8%), striatum (5.1%), and cingulate gyrus (3.4%). A mean 4.2 ± 2.6 structures were involved. On the basis of hierarchical biclustering, 7 involvement patterns were identified and correlated with cortical functional anatomy (pure insular [11.9%], olfactocentric [15.3%], olfactoopercular [33.9%], operculoinsular [15.3%], striatoinsular [3.4%], translimbic [11.9%], and multifocal [8.5%] patterns). Cox regression identified hippocampal involvement (p = 0.006) and postoperative tumor volume (p = 0.027) as significant negative independent prognosticators of overall survival and extent of resection (p = 0.015) as a significant positive independent prognosticator. CONCLUSIONS: The study findings indicate that insular gliomas primarily involve the olfactocentric limbic girdle and that involvement in the hippocampocentric limbic girdle is associated with a worse prognosis.

Sporadic spinal hemangioblastomas can be effectively treated by microsurgery alone.

OBJECTIVE: Clinical characteristics and management of hemangioblastomas of the spinal cord associated with von Hippel-Lindau syndrome have been extensively covered in the literature. This report aims to analyze the characteristics and surgical treatment results of sporadic spinal hemangioblastomas (SSHB). METHODS: This is a retrospective analysis of 14 patients with SSHB (8 men and 6 women) operated on during a span of 23 years. The median age was 41.5 years (24-70 years). von Hippel Lindau syndrome was excluded by imaging in all patients. The median follow-up was 4 years (1-23 years). We also conducted a meta-analysis of all 271 SSHB cases reported in the English-speaking language literature from 1967 to 2011. RESULTS: Nine (64.3%) lesions were cervical, 3 (28.5%) were thoracic, and 1 (7.1%) was lumbar. Eight (57.1%) tumors were dorsal intramedullary, 4 (28.6%) were exophytic, 1 (7.1%) was intradural extramedullary, and 1 (7.1%) was completely extradural. Diffuse segmental cord enlargement was present in 7 patients (50%) and a cyst/syrinx was present in 7 (50%). These 14 patients underwent 15 operations, and gross total resection was achieved in all operations. There was no mortality. Symptoms improved after 8 (53.3%) of 15 operations, remained the same after 5 (33.3%), and worsened after 2 (13.3%). The mean Karnofsky performance score improved from 79.3 (± 17.5) to 87.3 (± 12.2) after 6 months of follow-up. There was one recurrence 15 years after magnetic resonance imaging confirmed total resection. CONCLUSIONS: The SSHBs occur most often in the upper spinal cord. Excellent surgical results and long-term outcome can be achieved using microsurgery alone with only rare recurrences.

Asymptomatic remote cerebellar hemorrhage: CT and MRI findings.

The aim of this study was to evaluate the computed tomography (CT) and magnetic resonance imaging findings (MRI) of asymptomatic remote cerebellar hemorrhage (RCH) at the preoperative, early postoperative, and postoperative period. A total of 983 consecutive adult patients who underwent supratentorial craniotomies were included in the study. The ethics committee approved the study. The patient's clinical records and radiological examinations were retrospectively analyzed. All patients had preoperative CT and MRI examinations, immediate postoperative CT, and postoperative MRI within 24 h. The patients with the radiological diagnosis of RCH were followed up to 5 years. Eight asymptomatic RCH cases were recruited. The prevalence of asymptomatic RCH was 0.8% in our series. RCH was unilateral in two patients and bilateral in six patients. The postoperative CT was positive in two cases. The hemorrhage presented on MRI as folial linear hypointensities in six cases. In three cases (including one mixed case), punctate hypointense spots were identified at the superior cerebellar folia. Diffuse hemorrhage in the cerebellar tonsil, subarachnoid hemorrhage, and hemorrhage in the cerebellar vermis and the ventricles were also seen. The MRI findings were stable up to 5 years. The prevalence of asymptomatic RCH is higher than previously reported. Immediate postoperative CT is usually unremarkable; however, MRI demonstrates various hemorrhagic patterns at the cerebellum other than classical "zebra sign". This condition is self-limiting and no further investigation or follow-up study is required. In the proper clinical setting, the awareness of different hemorrhagic patterns in patients with RCH would prevent unnecessary investigations.