Skin Boosters in Daegu — What Has to Be Settled Before the Product Name

The conclusion, first

“Skin booster” is not a kind of product but a word that gathers several different branches under one name. In a 2023 press release the MFDS wrote the expression as “so-called ‘skin booster’”, and the things actually sold under that name are divided across three different regulatory systems: medical devices · medicinal products · human tissue. The academic side is no different — the only paper that attempted to define the term was retracted in 2025 (insufficient peer review and use of images without attribution), and a further piece by the same author in 2026 records that “definitions and clinical evidence remain heterogeneous.” A study in which 15 experts are building a consensus by the Delphi method is currently under way, and its registration documents say this — “differences exist in product selection, injection depth, dose, treatment interval, technique, indications and safety monitoring”, and “most of the available data comes from fragmentary clinical studies or from manufacturer-led guidance.” So choosing the product name first puts the order backwards. We settle five things first, and then choose among whatever candidates remain.

The order we actually work through in consultation

1. We first work out what the problem is — there are cases where a booster is not the answer

The booster classes mostly aim at the density and texture of the dermis. So if the main problem is hollowed volume or structure that has descended, the direction is a different one. It is genuinely not uncommon for a booster to drop out of consideration at this stage. The fork with the device classes is set out in should I have a booster, or a lifting treatment.

2. We settle whether it can be reversed — this is the axis that divides things most

With a material that cannot be reversed we look more strictly at the thickness of the data. An attempt that is permissible with a reversible material is not permissible with one that cannot be reversed.

3. We look at how many people the human data covers — the order of magnitude differs by class

Even when they are all called “skin boosters,” the thickness of the evidence differs by orders of magnitude. Randomised trials pitting products directly against each other on human faces can be counted on the fingers, and the largest of them was retracted.

The full map of the data class by class is in the things called collagen boosters, and the comparison taking GOURI as the reference point is in GOURI and the other skin boosters.

4. We check where the number of sessions came from

“Three sessions at four-week intervals” recurs as it travels from class to class, but the source differs for each class.

The “three sessions” for cross-linked hyaluronic acid has actually been tested, and the result is equivocal. In a trial randomising 53 people to a three-session arm and a two-session arm with 18 months of follow-up, on the investigator’s assessment the two-session arm was equal or higher at 6, 12 and 15 months (82.6% against 87.5% at six months, 86.4% against 87.0% at 12 months). The confidence intervals overlap heavily at every time point, and no between-group statistical test is posted with the registered results. We tell you this as it is.

5. We read out the approved wording exactly as it stands

The approved statement of intended use in Korea records these classes broadly as “temporary improvement of facial wrinkles in adults through physical restoration.” That is, “improved skin texture,” “pore reduction” and “improved tone” are not expressions that fall within the approved scope. And for the hADM class there is no heading of “approved intended use” at all — because it is human tissue, it is not subject to medical device or medicinal product approval. Even so, advertising copy sometimes attaches “pores” and “skin texture” to it. The detail is in what we check when you are looking into Re2O (hADM) in Daegu.

6. Only then do we look at the product

Once this much is settled, the candidates usually come down to one or two. We keep several boosters not in order to say “we have them all” but because the answer diverges at those five points. A product-by-product comparison table is in the collagen booster comparison.

What we do not do

• We do not say “this product is the best.” Most of the trials that pitted classes against each other on human faces ended in “no difference,” and for the remaining combinations there is no trial at all.
• We do not present a selection algorithm as though it were evidence. The 2025 expert consensus that ranked treatments for skin quality itself recorded that “comparative data on these procedures is very scarce, so the ranking rests mainly on the panel’s experience and preference,” and it received manufacturer funding.
• We do not present anything outside the approved scope as an effect. The approved wording is “temporary improvement of wrinkles.”
• We do not present reviews or before-and-after photographs as evidence of effect.
• We do not ask for a same-day decision. With a material that cannot be reversed, bringing a decision forward shifts the risk onto the patient.
• We distinguish between “we could not find any” and “there is none.” We do not disparage other clinics’ treatments.

Five questions you can ask at any clinic

1. “Is this a medical device, a medicinal product, or human tissue?” All three are mixed together under the name “skin booster,” and the kind of number that can be checked is different for each.
2. “Can this be reversed?” The only thing that can be reversed with hyaluronidase is cross-linked hyaluronic acid.
3. “In my case, is the problem density or volume?” If a product is settled on without that distinction, there is room for the direction to be wrong.
4. “Why this number of sessions?” The source differs by class. The only class with the number of sessions written into an approval document is PLLA.
5. “Is there a trial comparing this product with that one in people?” For most combinations the honest answer is “there is none.”

In summary — what is confirmed, what is inferred, what we could not confirm

Where this article was written

• Miso Clinic, Daegu · 4F Bombom Building, 125 Dongdeok-ro, Jung-gu, Daegu, Republic of Korea (near Kyungpook National University Hospital station)
• Director Lee Chi-Hak · Tel. 053-428-2700
• This column was written by the director and reviewed in September 2026.

This article was written not to recommend a particular treatment but to set out the extent of the data needed for a decision. Judgement about an individual condition requires examination, and this article does not substitute for medical care. The consultation procedure with GOURI as the reference point is in if you are having a GOURI consultation in Daegu, and which skin it suits is in which skin GOURI is suited to.

Frequently asked questions

Who wrote this

References

1. Absence of a regulatory definition — the MFDS press release of 5 January 2023 writes the expression as “so-called ‘skin booster’”. The things distributed under that name divide across three regulatory systems: medical devices · medicinal products · human tissue. We did not reach the full original text of each product’s approved intended use.
2. Absence of an academic definition — retraction — Yi KH et al., “Skin boosters: Definitions and varied classifications”, Skin Research and Technology 2024;30(3):e13627 → retraction notice Skin Research and Technology 2025;31(11):e70297. The stated grounds are peer review insufficient to meet the journal’s standards and images used without attribution. A follow-up piece by the same author in 2026 (Journal of Cosmetic Dermatology 2026;25(6)) also records that “definitions and clinical evidence remain heterogeneous.” A letter of rebuttal arguing that the term is a registered trademark and cannot be used as a generic descriptor is published in the same journal (2024;30(6):e13827).
3. An admission that there is not yet a consensus — the registration documents for the Delphi consensus study under way (15 experts, due to complete in 2027) record that “differences exist in product selection, injection depth, dose, treatment interval, technique, indications and safety monitoring” and that “most of the available data comes from fragmentary clinical studies or from manufacturer-led guidance.”
4. PCL against PN — Jeong GJ et al., Journal of Cosmetic Dermatology 2020;19(7):1593–1599. Randomised · split-face · blinded, 30 enrolled · 29 analysed, crow’s feet, 12 weeks. No significant between-group difference on any measure. The 218-person follow-up trial of the same two products (Journal of Cosmetic Dermatology 2025;24(1):e16576) was retracted in 2026 — at the authors’ request, on the grounds that the composition of the products used had been recorded incorrectly.
5. PDLLA against PLLA — Park JY et al., Skin Research and Technology 2026, DOI 10.1111/srt.70324. Multicentre · randomised · split-face · evaluator-blinded, 33 people, three sessions at four-week intervals, 24 weeks of follow-up. The degree of improvement was similar for the two products throughout, and the conclusion is non-inferiority. We did not reach the absolute figures.
6. PLLA against CaHA — we could not verify a randomised trial pitting clinical outcomes against each other. What exists is a 21-person gene expression comparison (Dermatologic Surgery 2024;50(11S):S166–S171), and the same data was published again in another journal under a stronger title (Journal of Drugs in Dermatology 2025;24(1):34–40). Both the research funding and the corresponding author’s affiliation are with one of the manufacturers.
7. hADM comparison trials — Lee YI et al., International Journal of Molecular Sciences 2026;27(5):2193 (randomised · split-face · double-blind, 20 people, 20 weeks, hyaluronic acid control). Zhang C et al., Aesthetic Plastic Surgery 2026;50:3710–3719, PMID 41381953 (multicentre · double-blind · randomised non-inferiority, 202 allocated · 175 completed, control an approved cross-linked collagen filler). Early swelling and pain were significantly more frequent on the hADM side (p < 0.05), and satisfaction at six weeks favoured the control arm.
8. An admission within a selection algorithm — Kerscher M et al., Journal of Cosmetic Dermatology 2025;24(Suppl 4):e70359. A panel of 12. The authors themselves wrote that “comparative data on these procedures is very scarce, so the ranking rests mainly on the panel’s experience and preference.” It is manufacturer-funded, and most of the panel are in a financial relationship with industry.
9. Reversal — CaHA — Yankova M et al., Aesthetic Surgery Journal 2021;41(5):NP226–NP236. CaHA injected into human cadaveric facial arteries and immersed for 24 hours. Neither sodium thiosulfate alone nor a mixture of sodium thiosulfate and hyaluronidase dissolved the CaHA. The authors’ conclusion is that “no suitable antidote is yet available.”
10. Reversal — PCL — Wu L, Journal of Cosmetic Dermatology 2025;24:e70201. In a controlled in vitro test, collagenase alone turned the gel into a solution within five minutes, while hyaluronidase · lidocaine · triamcinolone produced no reaction. The case material is 3 patients · 10 nodules. The subject is microsphere PCL, and it is not the same product as liquid-form GOURI.
11. Sources for the number and interval of sessions — for PLLA, the approved instructions for use (“1–4 sessions, usually 3”, “minimum three-week interval”). For PN, the review admits that its three sessions at four-week intervals is based on the authors’ clinical experience. The default in the PCL manufacturer’s official FAQ is one session. The basis for the CaHA dilution ratio is a 20-volunteer pilot. For PDLLA and hADM we could not find a primary source.
12. The only trial that tested “three sessions” directly — a randomised comparison of a three-session arm against a two-session arm in cross-linked hyaluronic acid (53 people, 18 months of follow-up). On the investigator’s assessment the improvement rates were 82.6% against 87.5% at six months, 86.4% against 87.0% at 12 months and 95.2% against 95.5% at 15 months, the two-session arm being equal or higher. The confidence intervals overlap heavily at every time point, and no between-group statistical test is posted with the registered results.
13. Pointing the other way — within the same class of cross-linked hyaluronic acid, the addition of glycerol alone made the reduction in pore volume 24.2% greater (randomised · double-blind · split-face, 29 completers, p = 0.038). The only multi-class comparison finding that combining beats a single agent is a study in 168 Syrian hamsters. A case of posterior ischaemic optic neuropathy after skin booster injection has been reported (Journal of Neuro-Ophthalmology 2025;45(1):e60–e62).
14. The approved wording — across these classes in Korea it is “temporary improvement of facial wrinkles in adults through physical restoration.” Because the hADM class is human tissue, it is not subject to medical device or medicinal product approval, so there is no heading of “approved intended use” at all. This point was raised publicly at the National Assembly audit in October 2025.
15. The regulatory content of this article is as of 6 September 2026.