“Radiesse side effects” is searched around 1,200 times a month in the US, with “Radiesse nodules”, “complications” and “gone wrong” close behind (clickstream estimate, October 2026). This page answers from two kinds of record, read on 15 October 2026: the adverse-event tables Merz posted for four of its own trials, and the reports FDA's device database has received since mid-2023. Both are partial, in different ways, and the limits are stated beside the numbers. It is not advice about treatment; anyone weighing Radiesse, or worried after it, is best served by a qualified clinician. What Radiesse is, and how it compares with Sculptra, is on our Sculptra vs Radiesse page.
The risks FDA lists for every filler
Radiesse is calcium hydroxylapatite microspheres suspended in a gel, approved by FDA as a dermal filler. FDA's dermal filler page lists the risks of approved uses for the whole class (fda.gov, content current as of 6 July 2023): bruising, redness, swelling, pain and itching; delayed swelling after illness, vaccination or dental work; nodules or granulomas that may need injections, antibiotics or surgery; infection; open or draining wounds; allergic reaction; and necrosis. It adds that injection into a blood vessel is unlikely but can cause serious and possibly permanent complications — tissue death, vision abnormalities including blindness, and stroke.
What four trials listed
Each of these trials posted adverse-event tables to ClinicalTrials.gov, listing every event above a stated threshold and every serious event. All four were sponsored by Merz, Radiesse's maker.
| Trial | Where, how many | Follow-up | Events listed most often |
|---|---|---|---|
| Hands safety study (NCT02904096) | US, 256 | 24 months | bruising 39/130 and 37/126; swelling 32 and 27; redness 20 and 8; nodule 17 and 14; pain 11 and 9; haematoma 7 and 1 |
| Jawline, 2021 approval (NCT03583359) | 180 enrolled | 60 weeks | injection-site mass 13/122; bruising 11/122; pain 8/122 |
| Nasolabial folds, split-face vs Restylane (NCT04647721) | China, 134 | about 52 weeks | induration 15/120 folds (Restylane side 7/120); nodule 12/120 (Restylane 4/120) |
| Nasolabial folds in people of colour (NCT01012388) | 100 | not stated | redness 16; swelling 12; bruising 7 (2% threshold) |
The hand study split participants into two groups by how much fat loss their hands showed, and the first figure in each pair is the group with the most severe loss. In the split-face trial each participant had Radiesse on one fold and Restylane, a hyaluronic-acid filler, on the other, so the registry counts folds, not people; that design is the closest thing in this set to a like-for-like comparison, and on it Radiesse's side recorded about three times as many nodules. The jawline trial's control group, treated later, recorded injection-site mass in 6 of 53.
“Nodule”, “induration” and “injection-site mass” are different codes for related findings — a firm area or lump at the injection site — and trials do not use them consistently, which is one reason figures from different trials are hard to line up.
The serious events
Serious events in these tables are listed whether or not anyone thought the filler caused them, and most are what any group of adults produces over a year or two: heart attacks, cancers, joint replacements, an accident. Two entries need stating plainly. In the hand study, one participant in the severe-loss group had a serious event coded “blindness unilateral”, and one a transient ischaemic attack. The registry does not say whether either was judged related to treatment, so the entry cannot be read as a filler-caused blindness without the investigators' assessment — which the public record does not give — and it cannot be dismissed either. In the jawline trial the one serious event in the treated group was erysipelas, a skin infection. The split-face trial recorded no serious event affecting either treated fold.
What FDA's device reports describe since mid-2023
MAUDE, FDA's device-report database, holds reports from manufacturers, clinicians and patients. Counting reports naming Radiesse received from 1 June 2023 to 15 October 2026 gives 435, of which 432 were classed as injury reports.
| Patient problem coded | Reports, Jun 2023 – 15 Oct 2026 |
|---|---|
| No clinical signs, symptoms or conditions | 180 |
| Obstruction / occlusion | 175 |
| Nodule | 69 |
| Subcutaneous nodule | 37 |
| Swelling / oedema | 35 |
| Necrosis | 30 |
| Deposits | 23 |
| Implant pain | 16 |
| Skin infection | 16 |
| Erythema | 13 |
| Cellulitis | 12 |
A report can carry several codes. The large “no clinical signs” group is a coding artefact the database does not explain; it sits beside reports that are classed as injuries, so it should not be read as 180 harmless reports. What stands out is obstruction or occlusion: 175 reports in a little over three years, against 7 for Sculptra over the same window. In filler reports that code usually describes product blocking a blood vessel, which is the mechanism behind necrosis and vision loss, and necrosis appears in 30 of these reports. The reason Radiesse's reports lean this way is not something MAUDE can show — differences in where it is injected, who reports, and how manufacturers code are all possible.
Yearly report counts for Radiesse have run above 100 since 2021: 110, 136, 124, 111 and 132 for 2021 to 2025, and 126 in 2026 by mid-October. There is no denominator: MAUDE does not know how many syringes were used, so none of these is a rate and a rising count can simply mean wider use. And a report is an allegation, which may be incomplete, duplicated or unrelated to the device.
Why a Radiesse lump is managed rather than dissolved
Hyaluronic-acid fillers can be broken down with an enzyme, hyaluronidase. Calcium hydroxylapatite is a mineral, and its carrier gel is absorbed while the microspheres remain, so there is no equivalent enzyme for the particles. That is why FDA's list mentions injections, antibiotics and surgery for persistent lumps, and it is the practical difference our Sculptra vs Radiesse page sets out under reversibility. Diluted Radiesse, injected for skin quality rather than volume, has its own trial and safety table, on our hyperdilute Radiesse page.
A PubMed search on 15 October 2026 for Radiesse or calcium hydroxylapatite with nodules or granulomas returned 71 records, and with blindness, vascular occlusion or necrosis 38, most of them case reports and reviews.
What the record does not show
It does not show how often serious events happen per treatment, because the trials are too small to count rare events and MAUDE has no denominator. It does not show how long lumps lasted, because the posted tables give totals. It does not show which injection areas carry most of the vessel-blockage reports, because the counts above were not split by area. And it does not show results for uses outside the approved indications. It does show that across Merz's own trials the events recorded were mostly bruising, swelling and lumps, and that FDA's reports since 2023 are dominated by a code that describes the filler's rarest and most serious risk. Our page on biostimulators places Radiesse among the other collagen-stimulating injectables.
Sources and dates
Opened 15 October 2026: ClinicalTrials.gov posted adverse-event tables for NCT02904096, NCT03583359, NCT04647721 and NCT01012388; FDA MAUDE via the openFDA device event endpoint, device.brand_name:radiesse, counts by patient problem, event type and date received; FDA's dermal fillers page (content current as of 6 July 2023); openFDA PMA records for P050052, including the 2021 jawline supplement (S129); PubMed searches as stated above. Corrections go to the contact page.