What is different about the hyaluronic acid used in skin boosters

The most reliable design in this field is to divide the left and right of the same person's face, put the product on one side and normal saline on the other, and keep both the patient and the evaluator blind to which is which. That trial exists.

Duteil L et al., Journal of Clinical and Aesthetic Dermatology 2023;16(2):29 — randomised · double-blind · placebo-controlled · split-face, 30 enrolled · 28 completed (mean age 57 ± 7). A non-cross-linked hyaluronic acid preparation at 10 mg/mL, 3 sessions 21 days apart, 3 mL per session into one side of the face with a 32G needle. Normal saline on the other side.

That the proportion judged improved by the evaluators was 82% versus 0% is the figure that stands out most in this trial. Because it means that in the same face, the same person, the same number of sessions, only one side got better.

There is data from a cross-linked product as well. Gubanova EI et al., Journal of Drugs in Dermatology 2015;14(3):288 put a stabilised hyaluronic acid gel and normal saline into the two hands of the same person (30 subjects, mean age 53, 3 sessions) and followed them out to 12 months, with the improvement maintained to the 12-month point (p<0.05). The adverse events were all mild.

Hyaluronic acid is originally a long chain that dissolves in water. As it is, it spreads and disappears quickly even when put into the body. So bridges are built between the chains to weave them into a net and make a gel — that is cross-linking.

The most widely used cross-linker is BDDE (1,4-butanediol diglycidyl ether). The reaction is this — under basic conditions the epoxide groups at both ends of BDDE react with the most accessible primary alcohol on the hyaluronic acid chain to form an ether bond.

There is one important point here. Cross-linking does not touch the glycosidic bonds of hyaluronic acid. So the body's original pathways for breaking hyaluronic acid down (the HYAL1 · HYAL2 enzymes, free radical oxidation) remain fully alive. A cross-linked gel is broken down by the body in the end too — it is only broken down more slowly.

The number called degree of modification

The degree of modification (MoD) is the ratio of BDDE residues to hyaluronic acid disaccharide units. It is measured by 400 MHz nuclear magnetic resonance (NMR).

Raising the degree of modification from 5% to 20% raises the elastic modulus proportionally and reduces the degree to which it takes up water and swells. Linear hyaluronic acid that has not been cross-linked does not swell. — Pluda S et al., Gels 2023;9(9):733

Degree-of-modification numbers can only be compared with one another within the same paper. Even for the same product the value changes with the method of measurement — the degree of modification of Restylane is reported as 3% in one source and 1.1% in another. Numbers from different papers must not be set side by side in one table.

Faivre J et al., Aesthetic Surgery Journal 2024;44(6):NP402 measured 16 commercial hyaluronic acid products under the same conditions. Being values from within the same paper, they can be compared with one another.

Looking at the table, degree of modification and firmness do not run alongside each other. Restylane has the lowest degree of modification at 1.1% but the highest firmness at 792 Pa, and Belotero has a degree of modification of 4.7% but the lowest firmness at 44 Pa. What determines the physical properties is not how much cross-linking was done but by what method the gel was made.

The authors of the same paper wrote this — “Manufacturing technology was the most important factor influencing the reversibility of the product.” Hyaluronic acid concentration, chain length and degree of modification each showed no clear correlation on their own.

This is a place where misunderstanding often arises in the consulting room, so we will set it out precisely.

A good many of the products sold as skin boosters are cross-linked products.

• Restylane Skinbooster Vital — as confirmed directly by authors connected with the manufacturer in a letter to a journal, it is a cross-linked small particle, high molecular weight hyaluronic acid (SB-NASHA technology).
• Juvéderm Volite (VYC-12) — it is a Vycross cross-linked product.
• By contrast, products such as Profhilo · M-HA10 · Redensity 1 are non-cross-linked (or stabilised by heat without a chemical cross-linker).

And they are not divided in Korean regulation either. A skin booster approved as a medical device is class 4, exactly like a filler. This is because the class is set not by whether it is cross-linked but by the risk of putting it into the body and having it stay there.

“Skin booster” is not a classification that exists in regulation but a word used in the market. The European medical device regulation and the US FDA likewise do not give this product group a separate category and classify them all as Class III.

The measurements also come out in the opposite direction to the received wisdom

Finke A et al., Plastic and Aesthetic Research 2024 compared three products under the same conditions.

The non-cross-linked preparation was firmer and held out longer. In other words the schema of “cross-linked = firm and long-lasting, non-cross-linked = thin and quickly gone” does not hold depending on the product. Whether something is cross-linked is too coarse a yardstick to use as a criterion for choosing a product.

Even with the same product, change the depth you put it into and the result changes. There is a study that verified this in the same patients.

Kim J, Archives of Aesthetic Plastic Surgery 2014;20(2):97 — 150 subjects · 1,000 injections in total (mean age 51.2), a stabilised hyaluronic acid preparation. It compared the case of placing it within the dermis (at a depth of about 1 mm, 0.001 cc per injection) with the case of placing it subcutaneously with a cannula in the same patients a year later.

These are the results 4 weeks after the intradermal injection.

The author's conclusion is clear — where it was placed in the dermis the texture and the thickness of the skin changed together, and where it was placed subcutaneously it merely replaced volume and the texture did not improve.

So a skin booster is a procedure whose result is determined as much by “where and in what amounts you divide it” as by “what you put in”. It means the result can differ with the same product, and put the other way round it also means there is a great deal of room to design.

Varying the layer according to firmness is the general recommendation — products of high firmness into the deep fat layer · on the periosteum, intermediate ones into the deep dermis, low ones within the dermis. However, this is the recommendation of a review and not the result of a trial that randomly assigned subjects by layer for comparison.

Hyaluronic acid preparations have one property that the other classes do not. It is that they can be dissolved and reversed with an enzyme called hyaluronidase. Classes such as polycaprolactone or calcium hydroxylapatite cannot be removed immediately with an enzyme.

This carries great weight for safety. Because it means that not only when you are unhappy with the result, but also in the rare emergency of a blocked vessel, there is a means available.

How well it dissolves differs from product to product. These are the values measured by Faivre 2024, mentioned above, under the same conditions (recombinant human hyaluronidase 150 U/mL, administered every 5 minutes, stopping when the firmness fell to 30 Pa or below).

Between the product that dissolved most easily and the one that dissolved least there was a 9.5-fold difference. And that order does not match the order of degree of modification — the Restylane products with the highest firmness (792–807 Pa) dissolved with 200 µL, whereas Volbella, with a firmness of 220 Pa, required 300 µL.

What matters in practice is this — if you have a record of which product was put in, you can estimate how much will be needed when it has to be reversed. It is one of the reasons we check the product and the dose with you before the procedure.

The “1,500 units · high-dose pulsed” regimen commonly cited in emergency protocols for vascular occlusion is a guideline based on expert experience and consensus, and the guideline authors themselves wrote that “there are no standard concentrations or doses described in the literature”. Even so, partial or complete recovery is reported in 84.2% of hyaluronic acid-related occlusions where hyaluronidase was used, and a delay beyond 5 days was correlated with permanent damage. If you feel that something is wrong, contacting us straight away is the most important thing.

We are asked about cross-linked products whether the cross-linker does not remain in the body. There is data from actual measurement.

Fidalgo J et al., Medical Devices (Auckland) 2018;11:367 — when 8 autoclaved hyaluronic acid hydrogels were measured by liquid chromatography-mass spectrometry, residual BDDE was from under 1 ppb to 2.46 ppb. A ppb is one thousandth of a ppm.

The control standard current in the industry is under 2 ppm. The measured values above are at one thousandth of that standard.

Once the reaction is over, BDDE exists in four states — bound at both ends, bound at one end only, inactivated by reacting with water, and left unreacted. Of these, hydrolysed BDDE has shown no toxicity or genotoxicity even at concentrations far higher than those in commercial products. Its metabolites 1,4-butanediol and glycerol are also reported to have no-observed-adverse-effect levels in mice of 100 mg/kg/day and 2,000 mg/kg/day respectively.

One thing to know for reference — the number “2 ppm” is not a threshold calculated toxicologically but a value that was the limit of detection at the time, which hardened as it passed through regulatory review. The original literature says as much. Today's analytical technology measures down to concentrations far below that.

What is verified

• Skin boosters have had an effect verified in a double-blind trial divided against a placebo — hydration +35% versus −10%, elasticity −27%, evaluator judgement of improvement 82% versus 0% (p<0.001).
• Cross-linking is a chemical reaction that weaves the chains together, and the body's pathways for breaking it down remain fully alive.
• Degree of modification and firmness do not run alongside each other — the manufacturing technology determines the physical properties.
• A good many of the products sold as skin boosters are cross-linked products, and in Korea they are managed as class 4, the same as fillers.
• Which layer you put it into changes the result — put in the dermis, the texture and the thickness change; put subcutaneously, only volume is replaced (150 subjects, 1,000 injections).
• It can be reversed — but the amount of enzyme required differs between products by as much as 9.5-fold.
• Residual cross-linker was in ppb on measurement — one thousandth of the control standard.

What is not yet verified

• Biopsy data showing that a non-cross-linked preparation increases collagen in human skin we did not find. The original source now cited for “hyaluronic acid makes collagen” (11 subjects, biopsies at 4 weeks · 13 weeks, P<.05) is a study carried out with cross-linked hyaluronic acid, and the mechanism was that the gel physically stretched the dermis and stimulated the fibroblasts.
• The “3 sessions at 3 to 4 week intervals” schedule is a number that came from the protocol each study adopted. A trial that divided subjects at random for comparison against another schedule we were not able to verify. The recommendations by product are inconsistent too — 2 sessions 4 weeks apart, 2 sessions a month apart, 3 sessions 4 weeks apart.
• The statement “non-cross-linked disappears in a few days to a few weeks” appears often in the literature but in many cases with no source attached. The actual observational data are a report (15 subjects) that it was not detectable on ultrasound at week 3, and at that point the increase in dermal thickness was still maintained.

What we say in the consulting room is short — rather than choosing by cross-linked or non-cross-linked, it is better to decide together which layer to put what into and in what amounts. The numbers needed for that decision are all written above.

How boosters and fillers differ is written up further in What is the difference between fillers and boosters, and the number of sessions and how long it lasts in How many sessions do I need and how long does it last.

Written and reviewed by Lee Chi-Hak, MD, medical director of Miso Clinic in Daegu, South Korea. Every study cited above is given together with its design, its size and the limitations the authors themselves recorded, and where we could not find data, we have said that we could not find any.