These two sentences get treated as one, and they are not.
Not permanent is a statement about time. Radiesse is calcium hydroxylapatite microspheres in a gel carrier. The carrier is absorbed over the first couple of months. The microspheres are progressively broken down into calcium and phosphate ions. The collagen laid down around them turns over like any other collagen. The result fades.
Not reversible is a statement about control. For hyaluronic acid fillers there is an off-switch. Hyaluronidase breaks the product down, often within minutes to hours. It is the standard emergency treatment when hyaluronic acid filler blocks an artery. Radiesse has no equivalent.
So the accurate summary is simple. It will go, but not when you want it to. If you are unhappy at week six, the main lever available is waiting.
No trial has measured "how long Radiesse lasts.": What exists is a set of separate trials in separate areas. Each stopped at its own timepoint. Each used its own rating scale. Those scales are manufacturer-developed and product-specific. The numbers below cannot be lined up against each other, or against another brand's numbers.
Nasolabial folds: The original approval rested on a split-face trial in 117 people, comparing Radiesse on one fold with a collagen filler on the other. At 6 months, 78.6% of the Radiesse-treated folds scored at least one point better than the control-treated fold on the same face.
The long-term extension of that trial is the most-quoted and most-misread Radiesse number: 102 of the original 117 patients were followed for up to 39 months after their last injection. 40% of folds assessed at 30 months or later were graded "improved" or better on a global aesthetic scale. The authors noted that improvement was "substantially less at this later time point." That is not the same as "Radiesse lasts 30 months." Sixty per cent of folds were not still improved. And "improved" on that scale is a low bar. The same follow-up reported no nodules, granulomas or infections in those 102 patients over three years.
Jawline: The trial behind the 2021 jawline approval found 75.6% responders at week 12 versus 8.8% in the untreated control group. By week 48, about 67.3% of the initial responders still met the definition of treatment success. The study's final timepoint was week 60.
Décolleté: The trial behind the 2026 approval found 71.2% responders at week 24 versus 6.3% control on the wrinkle scale at rest. The last timepoint was week 84.
Hands: There is no comparable duration figure to quote here. There is a safety figure that belongs beside any hand treatment. The label states that Radiesse "may cause nodules, bumps or lumps in the dorsum of the hand (12% reported this event) and can last up to a 1 year."
You will see "up to two years" attached to Radiesse. It is worth knowing exactly what it is.
It is a manufacturer statement: Merz's press release announcing the décolleté approval says results "can last up to two years." The studied window in that same approval ends at 84 weeks — roughly 19 months — and the same material concedes that safety beyond 84 weeks in the décolleté is not established. The claim reaches past the data behind it.
There is a separate, more cautious statement in the review literature: a 2024 review of filler degradation describes semi-permanent fillers, including calcium hydroxylapatite, as usually persisting up to two years and, in individual patients, more than five. Note what that sentence is about — persistence of material, not persistence of the cosmetic result. The two are constantly confused, and material that is still detectable is not the same as a visible correction.
The label itself makes no duration claim: It says only that safety beyond three years in the face and two years in the hand has not been investigated. The FDA's consumer page on dermal fillers puts calcium hydroxylapatite at approximately 18 months.
Hyaluronidase is an enzyme, and enzymes are specific. It cleaves the chemical bond that holds hyaluronic acid chains together. Radiesse contains no hyaluronic acid. Its microspheres are an inorganic calcium-phosphate mineral and its carrier is carboxymethylcellulose. Neither is a substrate for the enzyme.
This has been tested rather than merely assumed. A 2026 benchtop study pushed Radiesse into a simulated branching microvascular network and then tried to clear it. Hyaluronidase "produced a minimal effect on occlusion," while cellulase — an enzyme that digests cellulose — cleared the network within minutes. Three things have to be said in the same breath: this was a benchtop simulation and not a patient or animal study, one of its authors works for Merz, and cellulase is not an established clinical treatment and there is no approved cellulase product for this use. It is an interesting mechanistic result, not an antidote.
Sodium thiosulfate, or STS, comes up constantly in online discussion of dissolving Radiesse. The published picture is genuinely mixed and worth laying out plainly.
For it: A 2018 proof-of-concept study on 12 cadaveric porcine skin samples reported that intralesional sodium thiosulfate, alone or with topical sodium metabisulfite, completely dissolved calcium hydroxylapatite at 24 hours. There is also a published case report of intralesional sodium thiosulfate used to dissolve facial nodules.
Against: The largest preclinical study, combining in vitro work with injections in farm pigs and read out by 3D camera, micro-CT, CT, histopathology and electron microscopy, concluded: "We did not obtain any indications of CaHA degradation by STS, either in vitro or in vivo." Apparent shrinkage was attributed to the product being dispersed rather than dissolved. The same study reported necrosis and haemorrhage after sodium thiosulfate treatment. A separate cadaveric and in vitro study of filler inside facial artery segments found calcium hydroxylapatite still present after 24 hours regardless of sodium thiosulfate concentration, and it also failed when 300 units of hyaluronidase were added.
The conflict of interest runs the other way here, and readers should know it: The negative study was authored by Merz R&D and Merz North America staff — a manufacturer has a commercial interest in not having an independent reversal agent validated for its own product. The honest position is that both sides of this literature exist, neither is clean, and sodium thiosulfate is not an established reversal agent for Radiesse. It is certainly not a rescue treatment in an emergency.
Taken from what has been published, roughly in order of how invasive they are:
There is no approved reversal agent for calcium hydroxylapatite: The Radiesse label names none, and the review literature lists that absence as a known limitation of the material.
Every filler carries the same rare, serious risk: the product entering an artery and blocking blood flow, which can cause skin death and, when the vessels involved connect to the eye, loss of vision.
With a hyaluronic acid filler, there is something to give: High-dose hyaluronidase is the mainstay of published emergency protocols. With Radiesse there is not. That does not make an occlusion more likely with Radiesse — nothing published establishes that — but it changes what can be done in the hours afterwards.
For context on how often calcium hydroxylapatite appears in the vision-loss literature: a 2019 review of 48 cases of vision change after filler recorded calcium hydroxylapatite in 5 of them, with hyaluronic acid in 39. A 2024 review covering 365 newly identified cases recorded calcium hydroxylapatite in 1.6% and hyaluronic acid in 79.6%. Those are counts of published case reports with no denominator. They track how much of each material is being injected far more than they track comparative risk, and they cannot be turned into a percentage risk for any individual. What the 2024 review can tell you is the outcome pattern: of 318 cases with a documented outcome, 6.0% recovered completely, 25.8% improved partially, and 68.2% did not recover.
Go to an emergency department or call 911 if, during or after any filler injection, you have:
Sudden vision change, double vision or loss of vision in either eye
Sudden severe pain, or pain out of proportion to the procedure
Skin that turns white (blanching) and then dusky, mottled or net-like
Weakness, drooping, slurred speech or any other stroke-like symptom
These can be signs that filler has entered an artery. Time matters in hours. Published guidance treats suspected retinal artery occlusion as a same-hour emergency and suspected skin occlusion as a same-day emergency, though that window comes from a 1980 primate experiment and from expert consensus, not from trials in filler patients.
Do not wait to see whether it settles, and do not use pain as your gate. In a survey of 52 experienced injectors worldwide, pain was mild or absent in 47% of the intravascular injections they reported.
Radiesse raises the stakes here: Hyaluronic acid fillers can be broken down with hyaluronidase in an emergency. Calcium hydroxylapatite cannot, and there is no approved reversal agent for it, so the response is supportive care rather than dissolving the product.
Swelling keeps growing after the first few days instead of settling
A lump appears and does not settle
Redness spreads outward from the injection site
Tenderness, redness or a firm lump turns up weeks or months later — a delayed inflammatory reaction or an infection is possible
The product looks as though it has moved from where it was placed
None of these is an emergency in the same sense as the list above, but all of them are reasons to be seen rather than to wait it out.