Purpose

To evaluate the impact of computed tomography (CT) slice thickness on reconstruction accuracy of high-density tandem-ovoid applicators and resulting dose deviations in cervical cancer brachytherapy.

Material and methods

Seventeen cervical cancer patients treated with CT-based brachytherapy using stainless steel tandem-ovoid applicators (2.50 mm slice thickness) were retrospectively analyzed. Homologous 0.625 mm, 1.25 mm, and 5.00 mm image sets were reconstructed from original CT raw data. Applicators were reconstructed on each thickness, and treatment plans from 2.50 mm images were copied to other sets. Applicator displacements, reconstruction uncertainties, and dose deviations at 1.25 mm, 2.50 mm, and 5.00 mm were compared with a 0.625 mm reference.

Results

Slice thickness-induced partial volume effect caused applicator reconstruction uncertainties, predominantly in superior-inferior ovoid displacements, with maximum uncertainties approaching slice thickness. Each 1 mm superior-inferior displacement caused high-risk clinical target volume (HR-CTV) D₉₀ deviations of 1.23% (right) and 2.47% (left). Each 1 mm anterior-posterior displacement resulted in rectal D_2cc deviations of 1.71% (right) and 2.34% (left). Thicker slices (2.50 and 5.0 mm) led to average HR-CTV D₉₀ overestimations of 2.86% and 11.11% (maximum 7.06% and 25.55%). Rectal D_2cc was underestimated by an average of 1.40% and 4.43% (maximum: 6.67% and 8.39%).

Conclusions

CT slice thickness ≥ 2.50 mm introduces significant reconstruction inaccuracies, leading to clinically relevant dose deviations for HR-CTV and rectum. Reconstructing both 0.625 mm (for applicator reconstruction) and 2.50 mm (for contouring) images from a single scan is recommended for high-density applicator brachytherapy.