This study evaluated the impact of respiratory movement on four-dimensional computed tomography (4D-CT) acquisition and radiotherapy delivery. QUASAR^™ Respiratory Motion Phantom with nine sinusoidal waveforms that combined three amplitudes (10, 20, and 30 mm) and three breaths per minute (BPMs) (10, 15, and 20) was used. We acquired 4D-CT for nine respective waveforms and measured and compared geometrical factors (volume, shape, and motion) to reference static CT. We optimized and delivered radiotherapy plans on a Vital Beam and evaluated gamma for dose distribution analysis with three acceptance criteria: 3%/3 mm, 3%/3 mm, and 5%/5 mm. Geometrical analysis revealed that an increase in amplitude decreased 4D-CT acquisition accuracy. Geometrical factors for an amplitude of ≤ 20 mm were within tolerance limits, however, amplitude 30 mm resulted in significant deviations over tolerance limits. BPM had a minimal effect on geometrical accuracy when accurate pitch values for respective BPMs were selected for scanning. Dose distribution analysis also indicated that an increase in amplitude decreased the gamma passing rates. Amplitude 30 mm failed to achieve a 95% passing rate for all three gamma evaluation criteria adopted in the present study. Overall, our results demonstrate that respiratory motion has significant effects on the accuracy of both 4D-CT acquisition and planned radiotherapy delivery. Small amplitude of ≤ 20 mm is required to ensure accuracy.

Background: In postmastectomy radiotherapy, radiation reaches an already-reconstructed breast, and implant-based reconstruction remains feasible in selected patients. Far less studied is reconstruction within a previously irradiated field: after breast-conserving therapy, ipsilateral breast tumor recurrence (IBTR), and salvage mastectomy. We compared implant-based and autologous reconstruction in this setting and propose a hypothesis-generating treatment algorithm. Methods: From 2219 consecutive breast reconstructions (2012–2026), we cross-matched implant-removal, radiotherapy, and local-recurrence registries to identify all immediate total reconstructions of a previously irradiated breast after salvage mastectomy for IBTR (n = 13; 8 implant-based [all stand-alone, without vascularized flap coverage], 5 autologous [all latissimus dorsi-based]). The primary endpoint was reconstruction failure, defined in a modality-neutral manner as definitive loss or abandonment of the reconstruction; implant-based reconstructions of breasts not previously irradiated at the time of reconstruction (n = 689) served as a descriptive institutional reference cohort. Results: Seven of eight implants (87.5%) were removed at 3 to 51 months; the single retained implant has a persistent wound; none achieved complication-free success. Failure modes comprised early envelope necrosis, periprosthetic infection, and late attritional failure; chronic retained-capsule seroma followed removal without capsulectomy. All five autologous reconstructions healed without failure or unplanned reoperation (reconstruction failure, 7/8 [87.5%; 95% CI, 47.3–99.7%] vs. 0/5 [0%; 95% CI, 0–52.2%], p = 0.005; log-rank p = 0.039; median follow-up, 32.0 vs. 10.9 months; failure or major complication, 8/8 vs. 0/5, p = 0.0008). The implant-removal rate also exceeded that of the institutional reference cohort (87.5% vs. 11.2%; unadjusted, descriptive comparison). Conclusions: In this small consecutive series, stand-alone implant-based reconstruction of the previously irradiated breast was associated with a very high failure rate, whereas reconstruction importing well-vascularized autologous tissue healed in every case, a pattern biologically consistent with chronic radiation injury. These findings constitute a strong cautionary signal rather than definitive evidence. Our findings support an autologous-first approach after salvage mastectomy of the irradiated breast; stand-alone implant-based reconstruction should be offered only to carefully selected, fully counseled patients, and the proposed algorithm requires external validation.