I know the point has already been made abundantly clear in earlier posts & forgive a green member finding his feet here...sorry for the lengthy ramble that follows - I got a bit carried away.
I don't think this risk can be overstated - must rate among the more important HR info available. The info provided by those with first-hand experience is for the most part, and in my authoritative-sounding but naïve and humble view, excellent.
BDZs and opiates/oids do appear to have a more than additive respiratory depressant effect, and I don't know for sure but from experience/partly-educated guess, suspect that tolerance to either does not necessarily occur (certainly not predictably) in the parts of the brainstem that mediate respiratory drive in response to changes in O2/CO2 levels.
If I were a regular user of the combination, I would stop combining, or at least certainly err on the low-dose/low-potency/short-acting benzo side and/or engage in a little forethought (which I fully acknowledge is not a strong point of mine); we are likely to be at our least judicious re: this in the midst of opiate-associated sleeplessness, which I have experienced (I thought a bit idiosyncratically, but I'd like to know if others have found it to be common). I presume this is one scenario that would lead to use of BDZ following opiate, or coadministration of both.
My near-miss story involved the brilliant idea to follow oxycontin IV with lorazepam 4 mg (oral) and a small handful of cyproheptadine (4mg ea I think) tablets (also oral) to simultaneously treat the generalised itching and insomnia from oxyc. I was found not breathing with a deep purple-blue tinge to my skin, and was unrousable... but fortuitously for me quickly dispatched to intensive care, suffering hypoxia > respiratory failure, hypotension and renal failure. I was benzo tolerant, relatively opioid naïve, and it is anyone's guess to what extent the cyproheptadine contributed (it is sedating, and presumably contributed to some extent to respiratory depression). Previously I would not really have noticed that dose of lorazepam, because of BDZ tolerance and likely cross-tolerance from alcohol.
I was lucky not to have experienced a (obvious/noticeable

) hypoxic brain injury. My renal function, etc. eventually returned to normal, but reflecting on it, I think there was a fine line between that outcome, one involving permanent care and dialysis, etc., and one involving death.
I apologise if this sort of 'cautionary tale, based on when I...' causes irritation, or seems excessively self-indulgent. I don't know if reading such a post would have changed what I did. On reflection though, it really struck me how completely unexpected the BDZ/alcohol + opioid combo - associated respiratory depression was, because I had assumed my tolerance to alcohol and benzodiazepines, and oral BDZ use, would have made this a non-issue.
The literature, like that posted by SJP, reflects pretty clearly that it is the addition of benzodiazepines that contributes the most to unintended opiate ODs with a fatal outcome. It wouldn't surprise me if this, over time, has included a significant number habitual/experienced users of both (here in Australia temazepam + heroin/opioid both IV is or at least was, pretty common).
I would probably never again combine benzodiazepines and opiates, and certainly wouldn't administer benzodiazepines IV.
Sorry if already well known, but I'm not sure that we do recognise that benzos differ a lot not only in half-life but how rapidly they enter the CNS and there is probably differences in how they are distributed - so some may be preferentially incorporated into areas of the brainstem, others into other cortical/subcortical areas - Xanax/alprazolam for example differs from diazepam here. Temazepam, diazepam and alprazolam enter the CNS fairly quickly depending on the route, lorazepam less so but how quickly it enters the CNS appears to depend on serum levels (and therefore dose). I guess my point is that even if you could assume a degree of tolerance and attempt to work out when levels are likely to peak, etc. relative to opiate peak, there are so many other variables that I don't think it could be done reliably for HR. Temazepam is considered a relatively low-potency BDZ with short half-life, widely prescribed, considered relatively benign right, but it is likely to enter the CNS rapidly on iv admin and again, more-than-additively, interact with an opiate to cause rapid, likely unexpected and potentially fatal respiratory depression. I'll finish my ramble there.