In this study we quantified the motion magnitude of the right diaphragm dome during free breathing and DIBH, and various breathing control strategies supported by non-invasive mechanical ventilation. We investigated such motion control strategies using repeated MR imaging in each of the healthy volunteers. Regularized breathing and prolonged breath-holds prepared with preoxygenation and mechanically induced hypocapnia were investigated as possible alternatives for short DIBHs with air (without any preparation), being the current clinically used standard besides free breathing to minimize organ motion during radiotherapy.

In fifteen healthy volunteers we first assessed the variation of the right diaphragm dome position during multiple repeated DIBHs. Secondly, we investigated the diaphragm motion during RB at 22 brpm compared to free breathing. Finally, we analyzed residual diaphragm motion during PBHs from inhalation and exhalation on MRI.

The median IQR of the diaphragm positions over the five DIBH acquisitions of all volunteers and all sessions in our study was 4.2 mm, which is similar to interfraction variabilities reported in literature [15, 37]. In a study of fifteen pancreatic cancer patients, the diaphragm position variation in cranio-caudal direction between multiple DIBHs were found to have a group mean of 0.5 (SD 2.9) mm as measured on fluoroscopic [15]. This is in line with the 3.3 mm SD between successive breath-holds found during lung stereotactic body radiation therapy (SBRT) using breath-holding assisted with spirometry and repeat CT imaging [37]. Assuming a normal distribution and converting IQR = SD/1.35, these numbers are comparable to what we found in our study. Furthermore, we observed diaphragm displacement variations between DIBHs of up to 23.6 mm, in line with up to 19.9 mm reported by Lens et al. for pancreatic cancer treatment [15]. In patients treated with liver SBRT, it was shown that variations in daily breath-holding can have a large effect on interfractional diaphragm positions with respect to the vertebrae position varying from − 14 to + 15 mm [38]. his implies a risk of tumor misses when treating patients with repeated DIBHs. Finally, it should be noted that within a DIBH of 60 s the diaphragm drifts in the cranial direction. Holland et al. observed diaphragm drifts of up to 0.6 mm/s during DIBH, and Lens et al. showed that the diaphragm may move about 10 mm in cranio-caudal direction within one minute, from which 3.2 mm motion takes place in the first 10 s of the DIBH [14, 16].

MRI has been used previously to demonstrate how rapid shallow breathing with mechanical ventilation reduces breathing amplitudes with respect to FB [22, 24, 39]. Our mechanically induced rapid shallow breathing at 22 brpm, also resulted in significantly smaller peak-to-peak amplitudes compared to FB. We measured amplitudes of 11.3 mm and 16.6 mm in the respective two MRI sessions, which is comparable with the mean respiratory amplitudes of 9.4 mm and 10.5 mm at 20 brpm, and 8.0 mm and 8.6 at 25 brpm, respectively as measured on MRI in ten healthy volunteers [24]. Furthermore, in that study mean amplitude reductions of 56% and 62% for 20 and 25 brpm, respectively were reported. Van Ooteghem et al. analyzed shallow-controlled breathing at 30 brpm showing mean amplitude reductions of 36% compared to volume-controlled breathing, and 4% compared to spontaneous breathing [23]. We found a median relative amplitude reduction of 39% during RB at a frequency of 22 brpm compared to FB.

Previously, single PBH from inhalation (> 5 min) has been demonstrated to be feasible in healthy volunteers and in breast cancer patients [20,21,22, 34]. However, as MRI data evaluating internal motion during PBHs was not available up to now, our study is the first to quantify residual diaphragm motion during PBHs from inhalation and exhalation using MRI. Conform another report we demonstrated a displacement of the right diaphragm dome in cranial direction during breath-holding [16]. The cranial displacement of the diaphragm is a consequence of the gradual lung deflation caused by gas exchange in the lungs, whereby the uptake of oxygen from the lungs to the blood is not equally compensated by the secretion of carbon dioxide from the blood to the lungs [16, 19]. The linearly fitted displacements of the right diaphragm dome over time showed the median diaphragm drift velocity to be smaller during PIBH (3.0 mm/minute) than during PEBH (4.4 mm/minute). We argue that this is due to the same volume of oxygen being extracted having a greater proportional effect on lung volume at initially smaller lung volumes. In contrast, mean diaphragm motion velocities during DIBH (i.e. 20 s)—also measured on MRI, were reported to be greater during end-inspiration (~ 0.6 mm/s) than during end-expiration (0.15 mm/s) breath-holding [16]. Similar results were reported in a study comparing diaphragm motion magnitude and velocity during 60 s breath-holds with different lung volumes where the motion magnitude in cranial direction was larger during inhalation breath-holds than during exhalation breath-holds [15].

In our study we focused on the quantification of diaphragm motion since this possibly is the structure in the abdomen that moves the most. We limited our measurements to the motion of the top of the right diaphragm dome in cranial-caudal direction only by translations at the level of the pancreas in anterior–posterior direction. Since the curvature of the diaphragm will be different at different lung inflation levels, this introduces additional uncertainties. However, as in previous work [14], we found that the ventral and dorsal region of the diaphragm move differently than the mid diaphragm, suggesting that deformable image registration techniques might yield higher accuracy. Whereas the diaphragm motion is highly correlated with liver motion, other abdominal organs including spleen, pancreas and kidneys might move differently and/or to a lesser extent, and the diaphragm might not be a direct surrogate for abdominal organ (and tumor) motion [15, 40].

Using the investigated breathing control strategies in radiotherapy potentially reduces radiation-associated toxicities by decreasing the margins around the target volume, and sparing healthy tissues. Regularized breathing with mechanical ventilation at 22 brpm reduced the median motion amplitude from 20 to 12.4 mm in our cohort. This would correspond with an ITV reduction of 7.6 mm in cranial-caudal direction. When considering a mid-position approach as investigated by Lens et al., the PTV would be reduced from around 15–11 mm utilizing RB for lung cancer treatment, and 17–13 mm for pancreatic cancer treatment [3].

A limitation of this study is that the healthy volunteers were relatively young (median age 22 years). However, it has been shown that RB is well tolerated by lung, liver and breast cancer patients up to 83 years old and PBH is well tolerated in breast cancer patients up to 74 years old [21, 41]. We therefore do not expect important difficulties when we include patients in clinical studies.

In radiotherapy, typically DIBH durations of 30 s are used. Considering the intra-DIBH diaphragm drift of 3.2 mm displacement in the first 10 s, and 2.8 mm displacement in the following 20 s, this accounts for a 6 mm displacement within a DIBH which is not incorporated in safety margins. On top of this, consecutive voluntary DIBHs vary with regard to volume and amplitude, with a 4 mm IQR variation in our volunteer cohort. At the treatment machine, the inter-DIBH variation can be reduced with the aid of breath-holding tools such as a spirometer with visual feedback to increase lung volume reproducibility, or an active breath-holding control system, which was not available at our MRI experiments. Based on our results, incorporating both inter-DIBH variation and intra-DIBH motion into a margin recipe is not straightforward and requires more research. Furthermore, a PIBH of 10 min would require a 3 cm margin to incorporate the steady drift, which is highly unfavorable. Tumor tracking during PIBH at the linac would be one viable approach to account for this motion. Alternatively, we are investigating how to compensate for the gradual lung deflation during PBH with gradual lung re-inflation.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.


This article is autogenerated using RSS feeds and has not been created or edited by OA JF.

Click here for Source link (