the FDA-panic posts run hot and most of them recycle the same three screenshots. here’s what actually moved my understanding, after months of trying to figure out what’s verifiable from the patient side vs what isn’t. n=1 on the usefulness, ymmv. 1. FDA’s own compounding pages (503A vs 503B). boring but load-bearing. the distinction most threads skip: only 503B outsourcing facilities are required to set BUDs based on actual stability testing for that specific formulation. 503A BUDs lean on general USP convention. that’s why the same vial gets a 28-day number in one place and a longer one elsewhere. the category isn’t the tell, the paperwork behind the BUD is. 2. USP <797> summary docs on multi-dose vial BUD. worth reading once just to see that the 28-day number most people quote traces back to general multi-dose guidance, not anything tirzepatide-specific. once you see that, half the “my pharmacy is sketchy” posts read differently. 3. any pharmacy’s COA, read critically. a certificate of analysis is a release-time document. it certifies identity and potency at the point they tested the lot, not how your reconstituted vial behaves in your fridge three weeks later, and you can’t match it against the specific lot already in you. useful, just not the proof people treat it as. 4. the SURMOUNT trial PK supplements (NEJM/JAMA). for grounding what steady-state actually looks like so you stop attributing every wobble to your vial. the gap none of these close: you can ask a pharmacy which lot release path a batch took, and the answer tells you something real about their QC philosophy. but the lot you already injected gives you no way to verify the answer. the most procedurally informed patients still hit that wall the second the dose is in them. i log lot, fill date, and source per vial alongside CGM so drift has somewhere to point. the dose-day check-in reminder is the dumb little thing that keeps my week count honest. what sources did i miss? specifically looking for anything with real reconstituted-vial stability data, not release-time numbers.
The SURMOUNT PK supplements are worth reading, but “stop attributing every wobble to your vial” is harder advice to follow if you’re seven years post-RYGB. Altered gastric emptying and proximal bowel shortening mean your absorption profile wasn’t in that trial cohort in any meaningful way, so you don’t actually have a stable PK baseline to compare against. Which means the vial-drift question and the anatomy question become genuinely hard to separate, and the SURMOUNT data doesn’t resolve that the way it would for someone with intact anatomy
you’re right that “you don’t actually have a stable PK baseline to compare against,” and that’s the part that flips for me: if SURMOUNT can’t be your reference, your own logged baseline is the only one you’ve got, which makes holding the vial fixed and tracking drift against yourself more load-bearing post-RYGB, not less. fair point that anatomy and vial-drift get genuinely hard to separate there, but that’s an argument for tighter per-vial logging against your personal baseline, not for giving up on isolating the variable.
the stability data that exists for GLP-1 class compounds (not tirzepatide specifically, that’s basically absent publicly) suggests degradation isn’t a single linear pathway. oxidation vs hydrolytic breakdown produce different profiles, and a vial that tests clean at release can behave differently depending on how many temperature excursions happened between pharmacy and your fridge. worth logging cold-chain gaps alongside lot number, bc “week 3 of vial A” and “week 3 of vial B” aren’t the same variable even if your tracking treats them that way.
Batch-to-batch logging is doing more work than attribution alone, and I’d push back on “the lot you already injected gives you no way to verify.” True in isolation, but batch number on the stopper, reconstitution date, draw count, and CGM trends mapped together across multiple vials is pattern detection; it surfaces drift signals before you’re fully blind to what changed. On your specific ask: reconstituted-vial stability data for compounded tirz doesn’t publicly exist. That’s a literature gap, not a reading gap. Keeping batch and glucose data in the same daily log (the CareClinic check-in does this without extra friction) is what makes the pattern visible across vials rather than just within one.
point 3 is where i’d actually push, but in the opposite direction from where most people take it. you’ve got the COA pegged correctly as a release-time document, but there’s a layer under that most patients never see: the COA you get is usually for the bulk API, not the finished reconstituted product. identity, potency, related substances tested on the powder before it ever became your vial. so it’s not just “release-time vs three weeks in your fridge,” it’s “a different physical thing than what’s in the syringe.” bulk-API CoA and finished-product CoA are categorically different assertions and the patient-facing one is almost always the former. on the reconstituted-vial stability data you’re asking for: i’ve looked and i don’t think the clean public dataset you want exists, at least not at a resolution that survives the way people cite it. the stuff that floats around gets stretched into “you might be pulling 5-6mg by week four” claims that the actual data doesn’t license. so the honest answer is probably that the wall you hit at injection isn’t the only wall, the upstream stability literature is thinner than the confident posts imply. the confound i’d add to your per-vial log, since you’re already tracking lot and fill date: within-vial draw number. as a multi-dose vial draws down, the air-to-liquid headspace ratio climbs and interface area per remaining mg climbs with it, which is exactly where aggregation seeds. so dose 4 out of a single vial isn’t the same chemistry as dose 1 even on an identical fridge log. that’s separate from the cold-chain wobble and worth its own column. the dose-day check-in habit is doing more work than it gets credit for, fwiw. i log injection day and reconstituted-vial age in CareClinic next to fasting glucose so when something drifts i can at least see whether it tracks the weekly cadence or the vial age, and those two clocks pull apart more than i expected. doesn’t close your verification gap, nothing patient-side does, but it tells you which clock the wobble is riding.
this is an excellent summary, you’ve hit the exact points that are doing the real work. the gap is even wider than just the stability data, though. a couple of confounds i’ve been tracking alongside the ones you listed: 1. the CoA.
the one most patients see is for the bulk API powder, not the finished, sterile-filled product. a finished-product CoA is a different document, testing for identity, potency, and related substances on the thing you’re actually injecting. most 503a’s won’t provide it. 2.
cold chain history. a compliant fridge log at home doesn’t account for accumulated thermal stress during transit. a vial that sat at 20C for six hours in a delivery truck in july has a different stability profile than one that didn’t, but visual inspection won’t show it. 3.
within-vial chemistry. this is the one that gets at your stability question directly. dose 4 out of a multi-dose vial isn’t the same chemistry as dose 1, even with
the upstream “specific formulation” stability testing for 503B still doesn’t close the reconstitution consistency gap on the patient side, bc how it’s actually mixed by us is another variable before fridge stability even comes into play