Instructional videos optimise patient compliance with respiratory-gated breast cancer radiotherapy.

Purpose/Objective
Respiratory gating techniques are used in breast cancer radiotherapy to reduce incidental radiation dose to organs of interest, particularly the heart and lungs [1], [2]. The most common gating technique in Denmark is deep inspiration breath-hold (DIBH), where the patient inhales and maintains the breath-hold during irradiation. This technique increases thoracic volume and displaces the heart inferiorly and posteriorly, thereby reducing cardiac and pulmonary dose while enhancing coverage of the target volume [2], [3], [4], [5]. These dosimetric advantages are clinically relevant in most, but not all, patients [6]. A high mean heart dose is associated with an increased risk of ischaemic heart disease and mortality, while higher lung doses may increase the risk of secondary lung cancer [7], [8], [9], [10]. Accordingly, DIBH is incorporated into routine practice for compliant patients.
Despite its benefits, DIBH is highly dependent on patient cooperation and reproducible breath-hold. Inconsistent breath-hold amplitude, difficulty sustaining the breath-hold, anxiety, and limited understanding of the technique can impair performance, potentially reducing expected dosimetric gains, prolonging treatment times, and occasionally preventing patients from complying with DIBH. The quality of the instructional process is therefore an important factor for patient compliance with respiratory gating [11], [12]. By increasing patient awareness, engagement, and active participation, improvements in treatment delivery and dosimetric outcomes may be achieved. We hypothesised that a revised patient education programme with the opportunity for video-guided home practice would improve DIBH performance, without compromising target coverage or doses to organs of interest. In addition, a systematic instructional approach can also help staff deliver consistent information and strengthen patient preparedness. This study aimed to assess the impact of the revised patient education programme for DIBH, with ipsilateral lung volume as the primary endpoint and secondary endpoints including doses to organs of interest, target coverage, and the proportion of eligible patients who underwent DIBH.

Materials and methods
DIBH was indicated, as per Danish Breast Cancer Group (DBCG) guidelines, for all patients requiring loco-regional or left-sided whole breast radiotherapy. This observational study evaluated the implementation of a revised patient education programme, including the addition of instructional videos designed to improve patient preparation for DIBH.
Before the implementation of the instructional videos, patients received their initial instruction during a group session, which provided a general introduction to radiotherapy and a brief segment on DIBH; no written material specifically addressed DIBH. Patients were instructed in DIBH during the planning CT and received additional guidance from a radiation therapist during the first treatment session and at subsequent treatments if needed.
After implementation of the instructional videos, guidance during the planning CT and treatment sessions remained unchanged; however, patients received additional support through the videos. Patients were provided with an online link before the planning CT and instructed to practise the breathing technique at home. During the group radiotherapy session, they received the same in-person instruction as before, supplemented by an introduction to the videos and a recommendation to rehearse according to the guidance.
The instructional material comprised two videos with substantial overlap: a 6-minute introductory video explaining respiratory gating and demonstrating the exercises (https://region-midtjylland.23video.com/video/119755308/breathing-exercises-radiation), and a 4-minute training video for repeated practice. Both videos taught a breathing sequence consisting of an initial, modest abdominal inspiration followed by chest wall inspiration.
The standard treatment technique consisted of opposed tangential fields supplemented with segments to achieve a homogeneous dose to the CTVp. When regional nodal irradiation (RNI) was indicated, the periclavicular region was treated using an opposed anterior–posterior/posterior–anterior (AP/PA) field arrangement. More advanced techniques, including intensity-modulated radiotherapy (IMRT) and volumetric modulated arc therapy (VMAT), were applied when clinically indicated.
The respiratory cycle was monitored and guided by Respiratory Gating for Scanners [RGSC, Varian Medical Systems, Palo Alto, Ca, USA] via a marker block placed on the sternal bone. Treatment delivery was triggered when the chest wall reached a predefined height within the gating window. Initiating the breath-hold abdominally resulted in caudal displacement of the diaphragm, shifting the lungs and heart away from the treatment fields before beam triggering. The subsequent thoracic breath-hold elevated the chest wall and triggered the accelerator, ensuring optimal positioning of both the target and organs of interest before beam delivery.
The revised programme was gradually introduced during 2024 and fully implemented as standard practice at Aarhus University Hospital (AUH) since late 2024. To assess the impact, two cohorts from AUH were compared: patients treated before the video (1 March–31 August 2023) and after the video implementation (1 March–31 August 2025).
DIBH effectiveness was assessed using ipsilateral lung volumes — calculated in quartiles (<2000 cm3, 2000–2299 cm3, 2300–2599 cm3, and ≥ 2600 cm3) — doses to organs of interest, and target coverage from the respiratory-gated planning CT, as well as DIBH eligibility rates. All consecutive patients treated with DIBH were included. Eligibility rates were assessed among all patients with an indication for DIBH. Doses were obtained from the first treatment plan, regardless of revisions or whether all fractions used gating.
Since heart and lung doses were expected to vary by laterality and nodal involvement, patients were grouped by tumour laterality and use of RNI: right-sided, left-sided whole breast irradiation (WBI), and left-sided WBI + RNI. Among the included patients with right-sided tumours, only eight did not indicate RNI (four in each cohort); seven had bilateral breast cancer, and one had unilateral right-sided cancer and pre-existing pulmonary disease. Therefore, all right-sided tumours were analysed together.
Dose was prescribed to the CTV as per DBCG guidelines. Coverage of the CTVp aimed for at least 98% of the volume receiving 95% of the prescribed dose (V95% ≥ 98%). For the nodal volumes (CTVn), which included Level 2–4 (and Level 1 in selected cases), interpectoral, and internal mammary nodes (IMN), coverage aimed at V90% ≥ 98%. For small, laterally located tumours, IMN coverage could be reduced to spare the heart and lungs.
Doses, based on the prescribed 40 Gy in 15 fractions, were reported as medians with interquartile ranges (IQR). Group comparisons used Pearson’s chi-square for categorical variables and the Mann-Whitney U test for continuous variables, with statistical significance defined as p < 0.05.

Results
Data were available for all 476 patients treated for 484 breast cancers using DIBH, comprising 195 cases in the 2023 cohort and 289 in the 2025 cohort. Most patients were treated with opposed tangential fields, only fifteen patients received Volumetric Modulated Arc Therapy (VMAT), including four in 2023 and eleven in 2025. Baseline characteristics were generally comparable across cohorts. The median age of patients treated with DIBH was lower in 2023 than in 2025 (59 vs. 64 years), whereas patients treated without DIBH had similar ages (64 vs. 65 years; data not shown). All patients received 40 Gy in 15 fractions. Other patient and treatment characteristics are summarised in Table 1.
The use of DIBH radiotherapy increased from 83% in 2023 to 95% in 2025 (p < 0.001), rising from 86% to 95% for left-sided and from 77% to 94% for right-sided treatments. The median gating window during DIBH was 2 mm, indicating consistent breath-hold performance across the cohorts. Replanning from DIBH to non-DIBH was rare (one patient [0.5%] in 2023; three patients [1.0%] in 2025), with no significant increase in replanning between the two cohorts (p = 0.541).
Overall, the median lung volume increased by 84 cm3 from 2240 cm3 (IQR, 1909–2456) in 2023 to 2324 cm3 (2033–2612) in 2025. The absolute increase in median lung volume was highest for right-sided tumours (+187 cm3) compared to left-sided tumours (+117 cm3) (Table 1). The proportion of patients with lung volumes ≥ 2600 cm3 increased by nine percentage points in 2025, while the proportion with volumes < 2000 cm3 decreased by nine percentage points; intermediate ranges remained stable. This pattern was similar across laterality and RNI (Fig. 1, Table 1).
Dose to the ipsilateral lung and heart remained acceptable overall. For left-sided WBI without RNI, median lung doses increased slightly, while doses were unchanged for left-sided WBI + RNI and for right-sided tumours. Heart doses showed slight increases for left-sided cases and remained stable for right-sided tumours (Table 1).
For right-sided tumours, doses to the heart and lung were also calculated separately for patients with or without RNI. As no significant differences were observed (data not shown), all right-sided tumours are presented together.
Overall, target coverage exceeded V95%_CTVp > 99% in all patients across the cohorts. For patients treated with lumpectomy, coverage remained stable for right-sided tumours and left-sided WBI without RNI. In left-sided WBI + RNI, V95%_CTVp increased slightly but significantly from 99.2% (IQR, 98.3–99.5) to 99.8% (99.5–99.9; Table 1). For mastectomy, V95%_CTVp remained stable across cohorts and laterality, consistently ≥ 99.3%.
Coverage of CTVn was consistently high across all targeted lymph node volumes. All nodal levels achieved the intended median V90%_CTVn of ≥ 98%, except for Level 1, which reached 97%. Cohort comparisons showed no significant differences (Table 1).

Discussion
Overall, the introduction of the instructional video led to larger lung volumes, reflected by a shift toward more patients with high volumes and fewer with low volumes, and, in addition, to an increased proportion of eligible patients undergoing DIBH. Doses to organs of interest remained relatively unchanged, with only minor differences observed, unlikely to affect late effects. Although more patients underwent DIBH, indicators of gating quality remained favourable [1], [13]. Lung volumes increased, the gating window remained narrow (2 mm), and replanning due to inadequate gating was rare.
Median age of patients who successfully underwent respiratory gating increased after implementation of the instructional video. This may suggest that older patients benefit from the opportunity for preparation and home rehearsal, allowing more time to familiarise themselves with the breathing technique.
From clinical experience, improved gating makes it technically easier to produce a high-quality radiotherapy plan, potentially reducing planning time (not measured). The intervention does not lengthen patient information in the hospital setting and may even reduce the time required to achieve stable gating during the planning CT and daily treatment [14]. Minor variations in doses to organs of interest may be explained by relatively low absolute dose levels.
Knowledge regarding patient education on respiratory gating remains limited, although patients express a need for information and support regarding gating and DIBH [11]. Targeted education appears to improve patient confidence and performance, and in at least one study, coaching before CT simulation reduced setup time and dose to organs of interest [14].
The present study provides a real-world dataset comprising 476 patients treated over two six-month periods. By including all eligible patients and providing comprehensive quantification of ipsilateral lung volume, planned doses to target volumes (CTVp and, where indicated, CTVn), and organs of interest, the study offers insights into daily clinical practice. All contouring was performed according to the ESTRO guideline [15]. High consistency was expected due to the single-centre setting and a small, consistent group of RTTs responsible for delineation and treatment planning. The findings reflect routine workflows. The intervention was implemented as an adjunct to standard education without requiring additional clinical resources, as patients accessed the video at home. This approach supports integration and generalisability. The video link is provided in the manuscript without copyright restrictions; users should evaluate its benefits and limitations in their own clinic before routine adoption.
Several limitations should be considered. First, the comparison involves non-randomised cohorts, where residual confounding cannot be excluded. Second, dose metrics reflect national practice, with prescriptions to the CTV rather than the PTV, which may limit direct comparability with studies that prescribe to the PTV. Third, patient-reported perspectives on the information material were not collected. Although qualitative evidence indicates an interest in video-based information, this was not formally assessed within the present cohort [11]. Fourth, metrics evaluating whether improved gating translated into more efficient and shorter planning were not included. Finally, we assume that larger lung volumes reflect a better DIBH outcome, and that differences between cohorts are attributable to improved DIBH performance rather than anatomical variation.

Conclusion
Overall, the implementation of instructional gating videos led to increased lung volumes and an increased proportion of gated patients, without affecting the rate of replanning from DIBH to no-DIBH. Target coverage was maintained, and doses to organs of interest remained low, suggesting potential benefits for treatment efficiency and planning accuracy.

Declaration of generative AI and AI-assisted technologies in the manuscript preparation process
During the preparation of this manuscript, the author(s) used AI tools/Chat GPT to support language refinement, including suggestions for wording, synonyms, and reducing word count. The author(s) reviewed and edited all content and take(s) full responsibility for the published work.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.