What dose constraint, if any, do you use for the diaphragm when treating liver or lung tumors with SBRT?

I treated a HCC patient with a subdiaphragmatic tumor with SBRT 54 Gy in 3 fractions. He developed right posterior chest wall pain that radiated to his right shoulder. On autopsy, he was found to have a path CR in his tumor and necrosis of the adjacent diaphragm muscle. The chest wall appeared grossly and pathologically normal. I attempted to do a dose-volume parameter analysis but quickly realized that the diaphragm doesn't lend itself to dose-volume analysis. It's a mobile structure and a chal...

We do not use a specific constraint for the diaphragm when treating lung tumors with SBRT near it and to the best of my knowledge, such a constraint has not been included in any protocols. We have not seen what I would have designated as a diaphragm-specific toxicity. We have treated many lower lobe tumors abutting the diaphragm and I do explain to patients that they will likely develop some degree of pleural inflammation and effusion but I don't consider that as diaphragm-specific.

I am not aware of any constraints for diaphragm. The central part of the diaphragm is made of tendon, so it should have a higher tolerance to ablative radiation. The closest cousin is the chest wall which is made up of ribs and muscles.

I agree with @Dr. First Last. The more important issue is the GI track under the diaphragm which tolerates SBRT poorly. Due to motion, we need to be very careful about the high dose delivered to the GI track. There was a report about stomach ulceration after 50 Gy in 10 fx (max dose of 50 Gy) delivered to stomach due to tx of the left lower lobe tumor.