Juan,
None of my other images from other data result in this DBE behavior.
I re-created the same image that results in this DBE behavior from the same debayered data (which followed batch conversion, calibration and cosmetics correction - with the same master flats, biases and darks), and do not see the same behavior with DBE in other words, I re-star aligned and integrated). Keep in mind that the original file was also fine initially. It was at some point in playing around with the image in PI that the behavior suddenly began.
Let me know if there's anything else I can provide if you plan to look into this more, but the new image is fine so I plan to delete the old problematic image and move on.
Thanks for following up on this problem.
Larry