Multivariate Analysis of Factors Affecting BrainStem Integral Dose in Stereotactic-Treated Patients

Abstract

The dose contributed to the brainstem (B.S.) in stereotactic radiosurgery/radiotherapy (SRS/SRT) is an essential part of treatment plan evaluation. Yet, there is a lack of a stratifying schema for the lesions near the brain stem, based on doses and volume irradiated. We investigated the possible determining factors affecting brain stem integral dose (BSID). Materials and Methods: Treatment plans of 141 patients treated by SRS or SRT were reviewed Selection for this analysis was based upon lesions within 4 cm radius from the center of brain stem in all directions. Thirty-two patients were eligible. Out of 32 patients, 2 were treated with SRS and 6 with SRT. Lesions were stratified into five groups according to P-P distance (periphery of lesion to periphery of brain stem): group I (−1.5–0 cm) constituted lesions overlapping or interfacing brain stem; group II (0.1–0.9 cm); group III (1–1.9 cm); group IV (2–2.9 cm); group V (3–4cm). Conformity index (PITV) volume of brain stem receiving the mean dose and BSID were calculated. Univariate and multi-variate analyses were performed to correlate BSID with its determinants, namely PITV, P-P distance, C-P distance (center of lesion to periphery of brain stem), and mean collimator size. Results: The highest mean BSID was observed in group I and the least in group V; 1 patient in group I with BSID of 16.8 Gy/cc developed brain stem injury. In univariate analysis, all the determinant factors for BSID (P-P distance, C-P distance, collimator size, and PITV) were shown to impact its value, but on multivariate analysis only P-P distance was the significant factor (p = 0.03). Conclusion: Utilization of the P-P distance as a stratification method for lesions within 4 cm radius from the center of the brain stem is valuable in determining the brain stem integral dose. This schema may suggest that lesions in group I with P-P distance (−1.5–0.0 cm) are better treated with fractionated SRT rather than single-fraction SRS. In addition, plans requiring collimator size 730 mm with PITV > 1.5 should also be considered for fractionated regimens.