If this was recently there was an API bug that caused the readings that were marked as correct for some Kanji to be incorrectly returned when the items were pulled in batch by the API. It affected Smouldering Durdles for sure but probably also affected Tsurukame since pulling items in batch is the most efficient way to get the information for an app like that.
If you have an option to reset your database in Tsurukame and repull the subjects that should fix the issue if this was the cause. And if this was the cause you’re likely to run into this issue again with some other select Kanji if they’ve already been downloaded.