Herophilus-Galen line as a predictor of extent of resection in the occipital interhemispheric transtentorial approach to pineal tumors in children.

Objective: The goals of this study were to evaluate the extent of resection (EOR) obtained with an occipital interhemispheric transtentorial approach (OITA) in a series of pediatric patients with pineal region tumors and to define preoperative radiological factors predictive of the EOR.

Methods: This is a retrospective cohort study of a series of pediatric patients with pineal tumors who underwent surgery through a microsurgical OITA performed by the senior author during the period from January 2006 to January 2020. The tumor volume was measured preoperatively, and then on sagittal midline cuts the authors identified the most cranial point of the torcular Herophili (defined as the "Herophilus point") and the lowest point of the inferior profile of the vein of Galen (defined as the "Galen point"). The line joining these two points (defined as the "Herophilus-Galen line" [H-G line]) was used to identify the "Herophilus-Galen plane" (H-G plane) perpendicular to the sagittal plane. Tumor volumes located below and above this plane were measured. EOR was evaluated by measuring residual tumor volume visible on T1 volumetric injected sequences of immediate postoperative MRI.

Results: Thirty patients were selected for study inclusion. The preoperative mean tumor volume was 15.120 cm3 (range 0.129-104.3 cm3). The mean volumes were 2.717 cm3 (range 0-31 cm3) above the H-G plane and 12.40 cm3 (median 5.27 cm3, range 0.12-72.87 cm3) below the H-G plane. Three patients underwent only biopsy. Of the remaining 27 patients, gross-total resection (GTR; 100% tumor volume) was achieved in 20 patients (74%). In the remaining 7 patients, the mean residual tumor volume was 7.3 cm3 (range 0.26-17.88 cm3). In 3 of these patients, GTR was accomplished after further surgical procedures (1 in 2 patients, 3 in 1 patient) for an overall GTR rate of 85.18%. Larger tumor volume was significantly associated with incomplete resection (p < 0.001). A tumor volume ≤ 2 cm3 above the H-G plane (p = 0.003), linear extension ≤ 1 mm above the H-G line, and pineal histology were predictive of GTR at first OITA procedure (p = 0.001).

Conclusions: The H-G line is an intuitive, easy-to-use, and reliable indicator of the superior anatomical limit of visibility during the microsurgical OITA. This anatomical landmark may be useful as a predictor of EOR for pineal tumors performed through this approach. The main limitations of this study are the small number of patients and the exclusively pediatric age of the patient population.