Session Item

In this work the intra-fraction (IF) motion was estimated through intra-fraction CBCT (Elekta Medical) when DIBH was performed using Catalyst (CRad) or Active Breathing Coordinator (Elekta Medical) in liver SBRT treatments.

160 patients with liver metastases were considered. In 55 of them Catalyst was utilized and for the other 105 ABC was employed. A total of 232 (479) IF CBCTs were recorded for Catalyst (ABC). For every treatment session an IF CBCT was synchronized with the first treatment arc. The full treatment was composed of two arcs. The patient position was corrected before the second arc was imparted. Using these corrections the IF motion for each patient and treatment session was estimated.

Median values and standard deviations were computed for each spatial component (SI, LR and AP). Furthermore, the IF offset percentage lower than 5mm (O_<5) and the IF offset percentage greater than 8mm (O_>8) were also calculated. Moreover, systematic and random errors were obtained for each DIBH systems employed. To elucidate whether there were significant differences in the IF offset distribution widths between Catalyst and ABC IF motion data a Fisher-Snedecor F test was performed with a 0.05 significance level.

Table 1 shows medians, standard deviations and systematic and random errors computed for SI, LR and AP. O_<5 and O_>8 values were recorded in table 2. Both tables were arranged by spatial component and DIBH system used. Distribution widths were statistically distinct, as p-values computed through Fisher F-test were much lower than 0.001. Systematic and random errors were higher in Catalyst treatments for all spatial components. For both DIBH systems systematic and random errors were higher for SI direction. At least 81% (90%) of the IF offsets were lower than 5mm and no more than 11% (4%) were higher than 8mm in Catalyst (ABC) treatments. Lower O_<5 and higher O_>8 values were found for SI in both DIBH systems.

Table 1: Medians, standard deviations and systematic (Σ_error) and random (σ_error) errors arranged by DIBH system and spatial component.

Table 2: Percentage of IF offsets greater than 8mm (O_>8mm) and lower than 5mm (O_<5mm) arranged by DIBH system and spatial component.

DIBH reduces IF motion in liver SBRT but does not eliminate it completely, as some residual motion is still present due to the imperfectness of breath-hold and internal liver motion. Although both systems are perfectly suitable to perform DIBH in liver SBRT, ABC is more precise than Catalyst. These differences might be associated with the distinct behavior of the respiratory cycle in the gated area. When utilizing ABC the air flux is stopped and a completely flat respiratory pattern is obtained. However, Catalyst forces the patient to hold its breath and to maintain the cycle inside a gating window. Inside that window the respiratory cycle moves and so does the patient. Further investigation should be perform to unravel IF liver motion during SBRT treatments.