The Clinical and Prognostic Impact of the Choice of Surgical Approach to Fourth Ventricular Tumors in a Single-Center, Single-Surgeon Cohort of 92 Consecutive Pediatric Patients

Abstract

Objective: A single-institution cohort of 92 consecutive pediatric patients harboring tumors involving the fourth ventricle, surgically treated via the telovelar or transvermian approach, was retrospectively reviewed in order to analyze the impact of surgical route on surgery-related outcomes and cumulative survival.

Methods: Clinical, radiological, surgical, and pathology details were retrospectively analyzed. We selected n = 6 surgery-related clinical and radiological outcomes: transient and permanent neurological deficits, duration of assisted ventilation, postoperative new onset medical events, postoperative cerebellar mutism, and extent of resection. We built univariate and multivariate logistic models to analyze the significance of relationships between the surgical routes and the outcomes. Cumulative survival (CS) was estimated by the cohort approach.

Results: There were 53 girls and 39 boys (mean age, 83 months). Telovelar approach was performed in 51 cases and transvermian approach in 41 cases. Early postoperative MRI studies showed complete removal in 57 cases (62%) and measurable residual tumor in 35 cases (38%). The average tumor residual volume was 1,316 cm³ (range, 0.016-4.231 cm³; median value, 0.9875 cm³). Residual disease was more often detected on immediate postop MRI after telovelar approach, but the difference was not significant. Cerebellar mutism was observed in 10 cases (11%). No significant difference in the onset of cerebellar mutism was detected between telovelar and transvermian approach. The choice of surgical approach did not significantly modify any other postoperative outcome and 1-/3-year CS of high-grade surgically treated tumors.

Conclusions: With the limitation of a single-center, single-surgeon retrospective series, our findings offer significant data to reconsider the real impact of the choice of the surgical route to the fourth ventricle on the incidence of cerebellar mutism and surgery-related morbidity. This seems to be in line with some recent reports in the literature. Surgical approach to the fourth ventricle should be individualized according to the location of the tumor, degree of vermian infiltration, and lateral and upward extension. Telovelar and transvermian approaches should not be considered alternative but complementary. Pediatric neurosurgeons should fully master both approaches and choose the one that they consider the best for the patient based on a thorough and careful evaluation of pre-operative imaging.

Keywords: cerebellar mutism; children; fourth ventricle; telovelar; transvermian.

Figure 1

Anatomical classification of posterior fossa tumors requiring surgical access to the fourth ventricle. (A) Mainly/purely intraventricular, without evident brainstem infiltration or extensive vermian infiltration. (B) Midbrain/intra-aqueductal tumor with significant bulging in the upper part of the fourth ventricle. (C) Cerebellar/vermian tumor with extensive parenchymal/vermian infiltration and secondary bulging into the fourth ventricle. (D) Cerebellopontine angle tumors extending into the fourth ventricle through the Luschka foramen/foramina. (E) Brainstem tumors with dorsally exophytic fourth ventricular component. (F) Giant tumors with extensive posterior fossa involvement, including a significant fourth ventricle component..

Figure 2

Kaplan–Meier survival estimates. CS, cumulative survival; CI, confidence interval; y, year. (A) Kaplan–Meier survival estimates (histological grading): 1- and 3-year CS are analyzed in relation to histological grading (WHO 1–2 vs. WHO 3–4), showing a better 1- and 3-year CS for low-grade group (log-rank test, p = 0.0001). Low-grade tumors: 1-year CS, 97.6% (95% CI: 83.9%–99.7%); 3-year CS, 97.6% (95% CI: 83.9%–99.7%). High-grade tumors: 1-year CS, 82% (95% CI: 68.2%–90.2%); 3-year CS: 62% (95% CI: 46.4%–74.2%). (B) Kaplan-Meier survival estimates (histology): 1- and 3-year CS are analyzed in relation to histology (log-rank test, p = 0.03). Pilocytic astrocytoma: 1- and 3-year CS, 100%. Non-anaplastic ependymomas: 1-y CS, 85.7% (95% CI: 33.4%–97.9%); 3-year CS, 85.7% (95% CI, 33.4%–97.9%). Medulloblastomas: 1-year CS, 91.9% (95% CI, 76.9%–97.3%); 3-year CS, 73.6% (95% CI: 55.1%–85.4%). Anaplastic ependymomas: 1-year CS, 100%; 3-year CS, 66.7% (95% CI: 19.5%–90.4%). (C) Kaplan–Meier survival estimates (surgical approach for high-grade tumors): Considering the choice of surgical approach for high-grade tumors (WHO 3–4), we found better 1- and 3-years CS of transvermian approach when compared to telovelar approach with statistical significance (log-rank test, p = 0.048). Telovelar approach: 1-year CS, 76.9% (95% CI: 55.7%–88,9%); 3-year CS, 48.8% (95% CI, 28.5%–66.4%). Transvermian approach: 1-year CS, 87.5% (95% CI: 66.1%–95.8%); 3-year CS, 78% (95% CI, 54.8%–90.3%). The power of predicting factors (surgical approach for high-grade tumors and residual tumor volume) is evaluated in a Cox proportional hazard model ( Table 4 ).