Can facial filler affect bone? What FDA's evidence review actually found

FDA’s 2025 dermal-filler advisory materials identified a safety signal: several medical-device reports and published cases described bone resorption after hyaluronic-acid filler placed near bone, particularly in selected supraperiosteal facial injections. The evidence does not establish a population incidence, prove causality in every case, or show that all fillers, planes, areas, doses, durations, or patients cause bone loss.¹²

This is an evidence-literacy question, not a reason to diagnose bone change from appearance or remove filler automatically. FDA’s document was an executive evidence review prepared for an advisory committee; the panel’s discussion was advisory, not a new class-wide contraindication, recall, safety communication or final labeling order.¹²

Separate signal, association, causation and incidence

FDA’s review described three MDR cases and literature totaling up to roughly 60 reported cases across several publications.¹ Those counts should not be divided by an unknown number of injections to invent a rate. Case publications may overlap, select unusual findings, or lack comparable untreated imaging.

“Filler” hides the variables that matter

For any suspected relationship, preserve:

• exact product, formulation, lot and expiration;
• material class and whether it was FDA approved for the area;
• injection date, amount, dilution and repeat sessions;
• facial site, depth and relationship to periosteum;
• needle or cannula and entry points;
• prior filler, implants, surgery, trauma, dental work and infection;
• baseline and follow-up CT, cone-beam CT or other imaging;
• symptoms, examination and alternative explanations.

A supraperiosteal chin bolus and a subdermal lip treatment are not the same exposure. Hyaluronic acid, calcium hydroxylapatite, poly-L-lactic acid, PMMA and autologous fat should not be pooled into one causal claim without evidence.

Imaging findings need baseline and clinical context

Published chin and midface reports use CT or related imaging to describe cortical remodeling or resorption adjacent to filler.³⁴ Without pretreatment imaging, normal asymmetry, aging, dental or orthodontic change, pressure, prior trauma, infection, surgery, skeletal remodeling, or imaging technique can complicate attribution.

Ask a qualified clinician or radiologist to review the original images, not only screenshots. Relevant questions include:

• Is the finding true bone change or artifact?
• Is filler visible and in the same anatomical location?
• Are prior studies available with comparable technique?
• Are there symptoms, dental findings, inflammation or functional changes?
• Is the observation stable, progressive or clinically consequential?
• Would additional imaging change management or only add radiation?

Serial CT should not be a default response to an uncertain signal. A qualified clinician should define whether additional bone-focused imaging would change management and weigh the radiation and downstream consequences.

Mechanisms remain hypotheses until tested

Proposed explanations include pressure from material near bone, inflammatory response, mechanical forces, product properties, volume, repeated treatment and anatomical susceptibility. Biological plausibility can guide research but does not identify one mechanism in an individual.

Marketing claims can distort both directions. “Filler stimulates healthy bone” requires evidence; so does “filler dissolves your face.” Strong wording should track strong study design.

The advisory panel did not issue a final product verdict

FDA convened the 2025 panel to discuss long-term filler risks and possible regulatory approaches. The 24-hour summary records discussion and recommendations; it is not a final agency order, mandatory scan schedule, recall or class-wide label change.² Any later action should be checked in the current product labeling and FDA communications rather than inferred from the meeting.

Panel members also discussed research and postmarket evidence needs. That matters because passive reports and small imaging series cannot answer incidence, dose-response, susceptible anatomy, natural history after dissolving, or whether a finding changes function. A careful article should preserve those unanswered questions rather than turning the meeting into a headline conclusion.

Future studies need a denominator and baseline

A stronger study would enroll a defined population before injection, record exact product, lot, amount, plane and repeat exposure, use standardized bone-focused imaging only when ethically justified, include an appropriate comparison group, and follow participants long enough to distinguish normal remodeling from treatment-associated change. Blinded image review and prespecified thresholds would reduce hindsight bias.

Research also needs patient-important outcomes. A small imaging change may be asymptomatic; another finding may affect dental support, contour or future surgery. Frequency, magnitude, symptoms and reversibility should not be collapsed into one endpoint.

Registries could help only if injectors reliably record exposure and patients can be followed. This is another reason a lot-specific, plane-specific clinical record matters even when no immediate complication occurs.

Reporting should preserve uncertainty

If a clinician believes a serious injury or device-related problem may be associated with filler, preserve the product identifiers, dates, imaging and alternative explanations and consider the appropriate manufacturer and FDA reporting routes. The report can state suspected association without claiming confirmed causality.

When reading MAUDE or literature later, look for duplicates, missing product names, prior treatments and absent baseline images. The adverse-event report guide explains why a report can be valuable as a signal while remaining unsuitable for calculating risk.

Consultation should include the bone interface when relevant

For proposed supraperiosteal augmentation of chin, jaw, cheek or other facial skeleton, ask the injector to explain why that plane and product are chosen, what amount is planned, whether the use is on-label, and what evidence exists for repeated long-term exposure. The answer should include what is not known.

The cannula-versus-needle guide explains why access tool does not determine every risk. The ultrasound guide covers soft-tissue mapping; soft-tissue ultrasound does not answer the same question as CT or other bone-focused imaging.

Removal is not an evidence-neutral default

Hyaluronidase may break down hyaluronic-acid filler, but the effect on an established bone finding is not automatically known, and dissolving has its own limitations and risks. Other filler materials cannot be treated the same way. A decision should integrate symptoms, product, location, time, imaging, alternatives and the consequences of intervention.

If the issue is incidental and asymptomatic, the consultation should distinguish observation, additional imaging and intervention and state who owns each. If there is pain, infection concern, dental change, neurologic symptom or another urgent feature, cosmetic reassurance alone is not an evaluation.

The decisive question

Ask: “For this exact product, plane, facial area, amount and repeat exposure, what evidence exists about the bone interface, what remains uncertain, and what record would let us evaluate a future finding?” The responsible answer neither dismisses an emerging signal nor converts limited cases into a universal warning.