Hi Alison
We have been using only DIBH scan for 7 years now with good success. We have planned to test again to acquire both FB and DIBH- scans in the future.
The reason we have been using only DIBH scans is that we have RPM on the CTs and that so often we have to override CT SIM DIBH surface and replace it with “SGRT BODY” DIBH surface in most cases. The reason we have to replace it so often is that on the first fraction LAT- images, before making any matches on the image, spine needs to be inside 1 mm and sternum needs to be inside 2 mm in vertical direction, tight tolerances for the accuracy in other words. If those are not inside named ALs, new DIBH SGRT BODY surface is acquired after corrections done based on online match. This is what happens often. We kind of re- simulate 1) the breath hold level based on the displacement between spine and the sternum, 2) correct isocenter and 3) correct posture of the target and patient on the first fraction with very tight action levels to eliminate systematic errors in the future setups.
The advantage of using only DIBH at the CT is a small dose save of low dose CT, since FB scan is not acquired. The other advantage is easier workflow on the first fraction. The advantage of taking FB scan is minor with our workflow, with the reasons mentioned earlier.
The advantage of taking FB scan on the CT is that we have better understanding at the setup of how much patient needs to take air into lungs. This problem is solved in our current workflow with taking first LAT image and begin to adjust spine and sternum into their correct location and with the aid of that adjust the settings of AlignRT surfaces and couch VRT for the upcoming fractions to firstly create the correct BHL. However, this kind of adjustment is needed often even though FB scan is available. We have tested this earlier.
Some words of our first fractions workflow. We do not use tattoos with the breast patients at all. First we check the initial straightness of the patient on the couch. Then we shift the lasers to the indexed value on the breast board saved on the CT and check that it meets the mamillae level. Same things are done on the first fraction, if FB scan is available. Then we drive the couch to saved couch values (also LNG is calculated to be individually correct) and begin to monitor DIBH CT SIM surface. This is the moment where we are not sure how much patient really needs to take air into lung= the distance between FB surface (FB chest ROI) to DIBH surface (DIBH chest ROI) is unknown, even though the planned VRT should guide the spine to correct location in VRT. This is solved as mentioned after couple minutes when acquiring the first LAT image. Planned couch VRT is not allowed to differ from planned value at the setup (fits to workflows with and without FB scan, most centers do mind if couch VRT daily differs at the FB setup, I suppose). Then we drive the couch to planned VRT (where the spine is in its correct VRT) and try to relax the patient to get AlignRT VRT delta to zero error and send to couch other small couch shits based on AlignRT. These two latest actions are what we can do beginning from the first fraction already if FB scan is available. Now with our current workflow this FB surface setup with AlignRT is possible only beginning from the second fraction of course. So, with acquiring the FB scan, the setup workflow is the same beginning from the first fraction and easier to understand and learn for the new users for example. The other advantages are that we do not need to create FB surface on the first fraction, since it already exists at the AlignRT system. Now with our current workflow the creation of FB setup surface is done after image guidance with taking the gated "this and future fraction" surface under CT SIM or new DIBH surface. Then this is renamed as "free breathin surface" and changed to "setup surface only" after first fraction. This is what most other hospitals do alrady before first fraction at the preparation, since the FB BODY surface is avalaible in the first place.
As mentioned, after seven years of using only DIBH scan, we are planning to add this FB scan into workflow. The main reason is to make setup workflow beginning from the first fraction slightly easier and more compatible with the upcoming fractions. I do not see much advantage of using FB scan for the accuracy of the treatment with our workflow with using tight action levels for the accuracy in the images and thereby stict demandings for the accuracy of the setup surfaces. Note, this our current well planned and accurate workflow may differ from the ideology of the most users, so hopefully there was at least something usefull for you in the text. At least some others have considered to skip FB scan
![Wink ;)](./images/smilies/icon_e_wink.gif)
. I really consider you to use AlignRT in DIBH, it is great, no matter was FB scan used or not.
BR Marko