News › Delayed inflammatory reactions after hyaluronic acid fillers: how common are they? A meta-analysis of 242 studies

12 October 2026

Delayed inflammatory reactions after hyaluronic acid fillers: how common are they? A meta-analysis of 242 studies

Most adverse effects after a filler treatment appear immediately or within the first few days. Delayed inflammatory reactions are different: they occur weeks, months or years after the injection – as nodules, swelling, redness or induration – are hard to predict and can be protracted. How often they occur has hardly been quantified so far. A systematic review with meta-analysis has now pooled the published studies on this question.

Study design

The authors searched PubMed and Embase up to 3 December 2024 for studies on the safety of hyaluronic acid fillers in aesthetic treatments of the face. The protocol was registered in advance (PROSPERO), and reporting follows PRISMA.

Definition: A delayed inflammatory reaction was an event occurring two weeks or more after the injection with nodules, redness, swelling, induration or oedema; granulomas and foreign body reactions were included. Nodules without signs of inflammation that needed no treatment were not counted.
Meta-analysis of incidence: 242 prospective and retrospective studies with at least four weeks of follow-up – 30,907 treated patients, around 23,800 patient-years in total. 87% of the studies were prospective; median follow-up was 9.9 months (1 to 60 months).
Description of cases: In addition, 71 publications were evaluated, mainly case reports and post-marketing surveillance data. They were not included in the calculation of incidence.
Statistics: Random-effects model; study quality was assessed with the Joanna Briggs Institute tools but was not an exclusion criterion.

What did the meta-analysis find?

Incidence: 32 delayed inflammatory reactions occurred in the 242 studies. This gives a rate of 0.0005 events per patient-year (95% confidence interval 0.0002 to 0.0013) – about one event per 2,000 treated patients with one year of follow-up.
Onset: In 238 individually described cases, the reaction began after a median of 3 months. The range was 2 weeks to 12 years.
Patients affected: 97.5% were women; the median age was 48 years (22 to 77).
Course: The median time to resolution was 2.9 months – one day in one case, six years in the longest. Some reactions recurred and required prolonged treatment.
Treatment: Most often hyaluronidase, corticosteroids (mostly systemic, sometimes injected into the lesion) and antibiotics, less often surgical excision. In several cases the injection was given under ultrasound guidance. One case of suspected iatrogenic Cushing's syndrome after prolonged prednisolone treatment is described.
Histology: It was available in 22% of cases and mostly showed granulomatous inflammation or a foreign body reaction.
Differences between fillers: Broken down by crosslinking technology, the rates ranged from 0.0006 to 0.0044 per patient-year; for several technologies no event was reported in the studies.

Assessment: what the figures mean – and what they do not

More of a lower limit: The studies followed patients for a median of just under ten months, but the reactions began up to twelve years after the injection. Most known cases come from case reports, which were not included in the rate. The authors themselves consider the true incidence to be underestimated.
The definition determines the figure: An earlier review with a broader definition arrived at 1.1 affected patients per 100 patient-years – roughly twenty times as many.
The comparison of technologies carries little weight: It rests on a total of 32 events. The confidence intervals are correspondingly wide; in one subgroup they range from 0.0006 to 0.0314. Moreover, the comparison is not between randomised groups but between studies with different regions, volumes and durations.
“No event” is not evidence of “no risk”: The largest subgroup without an event comprises around 1,800 patient-years. At the average rate, just under one event would have been expected there – zero events are therefore also compatible with an average risk.
Conflicts of interest: All eight authors work as consultants, speakers or employees for the same filler manufacturer, which paid the publication costs. The discussion highlights a technology for which no event was reported. That does not make the figures wrong, but it belongs in the appraisal.
Level of evidence: The journal classifies the paper as level of evidence 3.

Conclusion for practice

Delayed inflammatory reactions after hyaluronic acid fillers are rare – in studies about one event per 2,000 treated patients per year, in reality probably more frequent. They begin after a median of three months but can still occur years later and persist for months. Whether the risk depends on the crosslinking of the filler cannot be answered reliably with these data.

In our view, three things follow from this: patient information should also mention reactions that occur only months after the treatment. The product and batch belong in every record so that a delayed reaction can be traced. And whoever treats should remain reachable for patients long after the injection. How to recognise complications and treat them in a structured way is the subject of our course Complication management in aesthetic medicine.

Frequently asked questions

How common are delayed inflammatory reactions after hyaluronic acid fillers?

The meta-analysis of 242 studies calculates 0.0005 events per patient-year – about one reaction per 2,000 treated patients in one year of follow-up. Because the studies followed patients for a median of just under ten months and case reports were not included in the rate, the figure should be read as a lower limit.

When do delayed inflammatory reactions occur?

In the 238 individual cases evaluated, the median onset was three months after the treatment; the range was two weeks to twelve years.

How does a delayed inflammatory reaction present?

Palpable nodules, swelling, redness and induration were described most often, sometimes with tenderness. The paper counted events from two weeks after the injection as delayed reactions.

How were the reactions treated in the studies?

Often with hyaluronidase, corticosteroids (mostly systemic, sometimes injected into the lesion) and antibiotics; in individual cases the material was removed surgically. The paper does not derive a uniform standard of treatment from this.

Does the risk depend on the filler used?

The subgroups by crosslinking technology showed different rates. However, the analysis rests on only 32 events in total, the confidence intervals are wide, and all authors work for one filler manufacturer. A ranking of products cannot be derived reliably from it.

Reference: Seo J, Kadouch J, Pavicic T, Corduff N, Li Q, Trindade de Almeida A, et al. Incidence and Characteristics of Delayed Inflammatory Reactions Secondary to Hyaluronic Acid Filler Injections: A Systematic Review With Meta-Analysis. Aesthet Surg J Open Forum. 2026;8:ojag165. doi: 10.1093/asjof/ojag165. PMID 42781409. Open access (CC BY 4.0): PMC13599015.