Databases run as clusters in replication...

Yes? I don't understand your point. Queueing systems are a kind of database with different semantics and performance characteristics.

The answer to what happens if the queue goes down is that it runs highly available.

Most people don't need another kind of database. A few do but not enough to justify an "everyone should do it this way by default" opinion.

Sure, if you're at the "it's not worth maintaining two dbs or paying for any managed offering" stage then you should still use the pattern but implement the work queue in your primary db. It'll be slower and you'll have to diy a little if there aren't ready made libs to do it but you still get all the major benefits.