Single point of failure

I'm going to be working out some life-or-death engineering details, and I'm going to shove them down the medical profession's collective throat. It isn't exactly lucrative, but its the kind of live-or-die engineering that should already have been dealt with long before-hand. Here's the background story.

My Uncle Ron has needed a respirator to stay alive for the better part of two years. Sunday he needed the air bottle swapped out for a full one. The bottle got stuck and wouldn't release properly. My aunt could not get a new bottle in before the old one went from "warning-low" to "actually empty" The result was hypoxia-induced cardiac arrest.

I bet there is an inexpensive way that an extra piece of hardware could have worked around the single point of failure (the dispenser).