SGRT Forum

Hi

In the clinical data of chest wall FB and DIBH setup images we have noticed small displacement errors between vertebra and sternum mostly in LNG and sometimes in VRT direction. This is not necessarily seen as pitch on the patient surface and in those cases it is understandable that AlignRT can not detect these kind of position errors with oppsite-t shaped ROI and RTD pitch shows zero. Do you think large ROI including breasts is better for FB setup to better detect the correct posture of the bony structures in the entire chest wall, why? Any technical aspects or results of the vertebra accuracy with chest wall ROIs?

So chest wall can shift without visible pitch (0 degree) in LNG and even in VRT in contrast to vertebra. If we are interested in accuracy of vertebra and its residual errors, it is important that this kind of error in patient posture is detected better in allready FB setup. In DIBH this thing may have influence on realized breath hold level daily variation in the images if chest wall surface has daily variation in FB setup in contrast to vertebra, and therefore in some units users rather trust acquired couch vrt than FB surface in vertical. Have you noticed this and how have you solved this?

Best regards Marko

Hi Marko,

I believe that it is easier to detect (and have a more reliable value of) the pitch if a larger part of the ROI covers the lateral of the patient i.e. the side of the chest wall. We use an ROI that more or less covers the target area.

To your concern regarding the vertebra and sternum. In kV-images we have also seen some disagreement in the longitudinal direction, both for patients treated in FB and DIBH. When using the vertebra as a primary matching structure (in kV-images) we quite frequently performed couch shifts in longitudinal direction and the corresponding MV-(verification)images showed that the LNG-shift (compared to the DRR) was unnecessary and even inaccurate.

Today we use the sternum and the distance between the chest wall and vertebras as primary structures when matching kV-images and the vertebras alone have lower priority. The MV's correspond better and we don´t have the same longitudinal displacement systemetics.

We haven't looked into the couch vrt vs. the surface vertical.

Kind regards Annika

Hi Marko,

We do not use an inverse-T ROI, our ROI for both Free Breathing and DIBH breast patients covers the affected breast but also comes quite across the midline (sternum). I agree with Annika, with the larger ROI the system has a better picture of any pitch.
In regards to setting table height, at our centre we decided against this.  We treat our breast patients on an inclined breast board and allow a 5mm tolerance in where they are longitudinally on the board.  If a table height was set then in this case, it would influence the depth of the breath hold. In addition, the nice thing about always using 2 surfaces that were acquired on the same day (DICOM from CT), is that it makes the set up more robust even when breast swelling occurs.  This would not be the case if the table height was set as a surrogate for free breathing surface.



the only other suggestion I could make is that if there is a vertebral/sternal mismatch then to observe the patient breathing in.  It could be possible that in an attempt to reach the breath hold she has lifted her back.

Hopefully this is helpful,
Amanda

Thank you ladies

You gave useful information. Errors has been noticed between the named structures,  indexed breast board was used and ROI covering the breast for the setup was preferred, (kV/kV match was done as the compromise between the sternum and ribs).  However,  at the beginning of this year AlignRT CBCT- study was published and there was concluded that pitch- and longitudinal problems in the patient position occurs also with the suggested ROI which covers the breast (Radiotherapy setup displacement in breast patients: 3D surface imaging experience, 2018). So if we keep the focus on FB setup now,  it seems that based on published (n=10) and unpublished (n=50) results both deformed breast and rigid part of the chest wall surface can be shifted in SI direction in contrast to vertebra without AlignRT RTDs seeing it. This kind of problem is daily visible in kV- images with laser setup and not much can be done for it. With SGRT there was supposed to be a potential to notice and decrease this kind of error and it is in our interest to solve this as effectively as possible. The solution is not so simple that the kV-image was acquired at certain breathing phase, because systematic error is found between vertebra and sternum in SI direction. So join the discussion as a goal to decrease this typical error in the field of RT with SGRT!

Best regards Marko

To potentially end this topic I would like to say that during this spring we will routinely create two/three different isocenter ROIs for FB breast/chest wall patients for clinical study. We continue with the opposite T- ROI which covers small medial part of the breast laterally, but after couch shifts are done based on kV/kV online match (and before vrt- surface for the treatment is taken) we take a snapshots in the three first fractions of those "competing" other ROIs to see how much RTDs differs between ROIs and which agrees best with the presented online matches. I suppose that this is not any new behavior for the users worldwide to adjust the ROIs in the first fractions but this kind of data with the real patients is minimal. Of course way to think the accuracy of the treatment is different between units and thereby the ROI that is suitable for some units ideology argues with the other units ideology.