Twitter’s back-end systems run on a distributed cluster of decomposed services. There are often cases where data sets will need to resynchronize between instances and data centers. This means, all the user data moves between server farms all the time, and some data servers may update later than others.
TLDR; services like Circleboom makes requests to Twitter via their official APIs (programmer access interfaces) and Twitter fulfills the requests successfully. Yet, they may show some of the destroyed data for a while.