week 18, compounded semaglutide (no polysorbate 80), and i want to name a side effect i haven’t seen written up cleanly anywhere. not GI. not the food noise drop. temperature. days 2-3 post-injection i run cold. like hoodie-on-when-nobody-else-in-the-house-is-cold cold. fingertips go white at the tips, the kind of thing where i’m rubbing my hands together at my desk in a 72 degree room. gone by day 5 every time. not every single cycle but often enough that i stopped calling it coincidence around week 14. here’s why i think it’s a real signal and not just “eating less = colder.” the appetite/food noise drop and the cold window don’t line up the same way. food noise bottoms out days 5-6 for me. the cold is earlier, days 2-3, which lands closer to where peak plasma would be on a ~5 day half-life (peak suppression on a once-weekly isn’t at some flat 48h mark, it’s days 2-3 as the drug climbs). so the timing tracks the Cmax window better than it tracks the intake window. different curve. steel-manning the boring explanation first: rapid fat loss + lower thermic intake + maybe some fluid shifts absolutely can make you run cold, and that’s probably part of it. but a once-a-week pulse that resolves by day 5 and comes back on schedule doesn’t read like a slow caloric-deficit thing to me. that’s a steady-state effect, not a 3-day window. the peripheral piece (white fingertips specifically, not just “chilly”) is what i can’t fully place. could be vascular, could be nothing. i’m not going to pretend i know the mechanism, my provider didn’t flag it and i haven’t pushed on it yet. caveats on my window so nobody over-reads this: i’m compounded sema specifically, no polysorbate 80, so the excipient profile is different from pens. i’m also still cycling, and i haven’t ruled out that some of the cold maps to cycle phase rather than shot day. that’s the confound i’m trying to isolate next, logging shot day vs cycle day side by side instead of eyeballing it. and i added strength work at week 15, which changes a lot of downstream stuff. for anyone who gets this: does it line up with your peak window (days 2-3 ish) or is it spread across the whole week for you? and tirz folks, do you see it at all or is this a sema thing? trying to figure out if it’s pharmacokinetic or just me.
the days 2-3 vs days 5-6 split is a genuinely good catch, and tracking the cold against Cmax instead of the intake window is the right read. where I’d push is the “not every cycle” line, because that’s the detail doing the most work and a clean PK story doesn’t actually explain it. if it were purely peak-driven, your Cmax lands in the same window every week as long as dose and absorption hold steady, so it should show up every cycle. when it’s intermittent, the variable I’d check first is vial open life, not the curve itself. oxidation shifts potency, potency shifts your effective Cmax, and that reads as a stability thing rather than a PK one. worth logging cold cycles against vial open date alongside the shot-day vs cycle-day column you already mentioned.
the central piece is the angle i haven’t seen anyone raise here. there’s rodent work, can’t cite it cleanly so take it as “a study i read,” showing central GLP-1 agonism actually drops core temp and dials down brown-fat thermogenesis, not through intake at all, a direct hypothalamic hit on the thermoregulatory setpoint. that would track your Cmax window better than the fat-loss story does, because it’s a drug-pulse effect not a steady-state deficit thing, and it lowers the setpoint rather than just leaving you with less insulation. the “cold days 2-3, gone by day 5” shape fits a central mechanism that rises and falls with plasma drug, which is exactly the curve you’re describing. where it breaks for me: that’s a core-temp story, and white fingertips specifically is peripheral vasoconstriction, a different layer sitting on top. lower core setpoint means the body clamps down on peripheral flow to defend it, so that gets you partway, but white-at-the-tips reads more Raynaud’s-shaped than “generally chilly.” and that’s where the still-cycling caveat you flagged earns its keep, because Raynaud’s-type vasospasm skews hard female and tracks estrogen, so the white tips could be riding cycle phase even if the all-over cold is riding shot day. two different curves landing in overlapping windows and getting read as one. so your shot-day vs cycle-day column might need a third split: the diffuse cold vs the white-fingertip events as separate rows, because i’d bet they don’t move together. if the diffuse cold tracks shot day and the tips track luteal phase, that’s a cleaner story than forcing one mechanism to carry both. fwiw i’m sema not tirz so i can’t speak to whether the central piece even transfers, the receptor occupancy and half-life are different enough that the window might not line up the same. anyone logging this granular, do the peripheral events and the all-over cold actually show up on the same days, or do they split?
the two-curve split is clean and i’d bet you’re right that diffuse cold and white-tip events don’t move together. but the case for “diffuse=shot day, tips=luteal” still treats one variable as constant that probably isn’t: the delivered dose itself. you flagged “not every cycle” and then routed all the variance into cycle phase. on a compounded sema fill with no polysorbate 80, you’re leaning harder on cold chain and BAC quality to hold potency, and if your vial’s been open more than 30 days the effective Cmax is drifting shot to shot. so the cold showing up “often enough but not every time” could be tracking vial open life, not just where you are in your cycle. that matters for your logging plan. you’re about to add a column for shot day vs cycle day, and a third row split for diffuse vs tips. add a fourth: vial open date and days-since-open at each shot. if the diffuse cold weakens as the vial ages, that’s a potency story sitting underneath the central-mechanism story, and it’d look exactly like “not every cycle” if you’re not watching for it. two clocks again, basically. the rodent central-thermoreg thing you cited is plausible and the Cmax shape fits, but a lower-potency shot late in the open window produces a smaller pulse, which reads as “skipped” rather than “weaker.” on the Raynaud’s-tracks-estrogen point, that’s a prescriber question and i won’t pretend to adjudicate the vasospasm mechanism. i can only tell you that on the compounding side, “this effect isn’t consistent shot to shot” is a potency flag before it’s a PK flag, and the open-vial age is the cheapest variable to rule out first. ymmv but it’s free to log.
The white fingertips specifically is the part I keep returning to, because that’s not the same signal as “running cold” and the distinction might actually matter. What you’re describing sounds closer to a vasomotor response than a thermoregulatory one: peripheral vessel constriction rather than core temperature drop. GLP-1 receptors are expressed in vascular smooth muscle, which opens a plausible route for transient peripheral vasoconstriction at Cmax that has nothing to do with caloric intake or the thermic effect of food. I’m on tirz post-sleeve rather than sema so I can’t compare these directly, but your tirz question has some mechanistic legs worth thinking about: GIP receptor activity carries its own vascular dimension, and if this is primarily GLP-1-mediated you might see it blunted or absent on tirz rather than just shifted in timing. Worth raising with your prescriber, not because it sounds alarming, but because a vasomotor pattern and a cold-from-deficit pattern would be framed differently clinically, and if the white-tipping is consistent across cycles they’d probably want to know it’s there. You’ve done good work separating the food noise curve from the temperature curve, that’s the kind of timing detail that actually helps them take it seriously.
the cmax timing logic is solid, but you’ve framed it as pharmacokinetic-vascular vs “eating less = colder,” and there’s a third option that fits your 3-day window better than either. falling leptin signals the hypothalamus to drop thermal setpoint, and that tracks the post-injection pulse and fat mass, not daily intake. separate mechanism from caloric deficit, which is exactly why it doesn’t line up with your food noise bottoming at days 5-6. the white fingertips i can’t place either, reads more vascular to me. ymmv.
the mechanism case is interesting but the white fingertip detail is doing more work than the post credits it for. “could be vascular, could be nothing” undersells it – blanching at the tips specifically tracks Raynaud’s pattern, which can be triggered or amplified by GLP-1 vasoconstriction effects independent of the thermogenic suppression hypothesis. those aren’t the same signal and separating them matters before you bring it to your provider.
Your timing logic is the strong part here, lining the cold window up against the Cmax climb rather than the intake curve is a genuinely careful read, and I’d not argue it. Where I’d push back is “haven’t pushed on it yet.” The white fingertips specifically, going pale at the tips on days 2-3, is not the part to sit on for another four weeks while you isolate the cycle confound. That’s a vascular sign with a name, and it’s exactly the kind of thing a GP wants to hear described in your own words. Bring it to them precisely as you wrote it here. I’m only a pharmacy tech, so I won’t guess at mechanism, but the specificity of what you’ve logged is what makes it worth raising now rather than later.
On tirz since early 2024, post-sleeve, and I don’t see that tight 3-day cold window you’re describing. Worth noting, because if pure GLP-1 agonism produces this and dual agonism doesn’t, that’s a pharmacological signal rather than individual variation. On your cycle-phase logging plan: the free-text journal tied to each injection entry in CareClinic is what made a similar luteal/peak overlap legible for me. Context around the date, not just the date, is where that kind of confound shows up.
the Cmax timing argument is genuinely solid framing, and the food noise curve running later than the cold window is the kind of detail that’s easy to hand-wave but harder to explain away once you’ve logged it across enough cycles. that said, “i haven’t ruled out that some of the cold maps to cycle phase” is underselling how much that variable actually matters here. luteal phase is a documented trigger for raynaud’s-type vascular episodes specifically because of how progesterone affects peripheral vascular tone. if your shot day has been landing in a consistent cycle window for 18 weeks, the white fingertips especially could be almost entirely cycle-driven and just looking pharmacokinetic because the timing overlaps. you’d need to desynchronize shot day from cycle phase to actually pull those apart, and that’s not a trivial test to run on yourself. the general coldness and the white fingertip piece might also be two separate signals running concurrently, one potentially sema-related, one not. collapsing them into a single mechanism makes the Cmax case look cleaner than it maybe is. the weekly summary in CareClinic was actually what showed me my own fatigue pattern was clustering in the same luteal window every month, not distributed evenly, which is when i stopped blaming my vial.
the confound-isolation plan is a good instinct, but i’d pull the white fingertips out from the rest of it. “fingertips go white at the tips” on days 2-3 is a vascular sign, and that’s not something i’d sit on for another cycle while you tease out cycle phase vs shot day. the run-cold-all-over part can wait for the log. the peripheral piece i’d flag to your provider now, in exactly the words you used here. white at the tips, days 2-3, gone by day 5. anecdotal from a pharmacy tech, but that specificity is the part they need to hear, don’t round it off.
edit: typo