Posterior fossa lateral ependymoma in childhood.

Over a period extending from 1984 to 1993, 16 children ranging in ages from 2 months to 12 years with posterior fossa ependymoma were treated. Four of these tumors were classified as lateral ependymomas because of their configuration and suspected site of origin. These 4 patients' ages ranged from 2 months to 5 years. Signs and symptoms of increased intracranial pressure and cranial nerve dysfunction were the hallmark of their presentation. Their imaging evaluation included a preoperative and postoperative CT san and/or MRI. A gross total surgical resection' was completed in the 4 children. At least a 1-year follow-up was available for all the patients. Intraoperative brain stem evoked potentials and facial and glossopharyngeal nerve monitoring were used in all the surgeries. Three children required a ventriculoperitoneal shunt postoperatively. Chemotherapy was used postoperatively in 3 children. The 4th child did not receive chemotherapy due to parental refusal and succumbed to a recurrence 1 years postoperatively. Radiation therapy was given to 2 children. Tumor recurrences developed in all children at 12 months, 18 months, 3 years and 4 years postoperatively in proximity to the original tumor site. Histologic sections in 3 patients, including those at the time of recurrence, were consistent with a low mitotic index. Only 1 child's histologic diagnosis was consistent with a high-grade ependymoma. The lower cranial nerve dysfunction transiently worsened in all the patients and was permanent in 2. The child's age, tumor histology and extent of resection play an important role in the prognosis of posterior fossa ependymoma. The lateral posterior fossa ependymoma has a particularly poor prognosis due to its location and its significant postoperative morbidity. The authors propose a combined midline and lateral suboccipital approach in order to expose the fourth ventricular floor, upper cervical spine, lateral recess, cerebellomedullary fissure, cerebellopontine and the cerebellomedullary cistern in an attempt at a "total' resection with a focus on the site of origin. The usage of neurophysiological monitoring appears to be useful in limiting and predicting the extent of postoperative complications. However, prolonged morbidity is likely and is commonly related to lower cranial nerve deficits. The author proposes a proactive approach in order to limit the sequelae of these complications.