GOURI — from the day of the injection to months later

What we confirmed while making this table — what exists in the peer-reviewed literature on liquid-form PCL is five papers in all. Two rat experiments, one human pilot (30 people · 6 months), one case series (2 people · 3 months) and one case report (1 person). The largest number in humans is 30, and the longest follow-up is 6 months. We think that fact itself should be the starting point for judging this procedure.

Microsphere polycaprolactone (the Ellansé family) and the liquid form differ fundamentally in formulation. That difference is covered in detail in it is the same polycaprolactone, so what is different between the liquid form and the microsphere-based form, and here we touch on only the part that affects the passage of time.

According to the original text of the manufacturer's patent, the composition is a block copolymer of polycaprolactone, a hydrophobic polymer (molecular weight 2,000–25,000 g/mol preferred), and methoxy polyethylene glycol, a hydrophilic polymer, and the copolymer concentration in the aqueous solution of the worked example is 25% by weight. The remainder is water. The patent also explains the mechanism of action directly — once inside the body the structure that had been stably dispersed in water collapses, and the matrix structure formed by the polymers binding to one another induces collagen.

So you should not read the raised look immediately after the procedure as the result. That is the water that was put in plus the swelling from the injection, and it disappears within a day or two. Nor does its disappearing mean the procedure did not work — with this product it is what comes after that is the beginning.

We do note, though. The “6 hours” is rat data and cannot be transposed to humans as it stands. We did not find data measuring the time for the carrier to be absorbed in humans. And there is no guarantee that the composition in the patent is the same as the approved composition of the product on the market — the consumer page gives “PCL 21%”, a different figure from the worked example in the patent.

Two people is not an incidence. You should not read it as anything more than “in 2 of 2, swelling lasting under 48 hours”. And we have not copied across the figures for microsphere PCL here (persistent swelling 4.5% · bruising 2.7% · lumps 0.45%, in 1,111 cases) — because that is data on a product of a different formulation. Those figures are set out in nodules that form after a procedure and in the difference between the liquid form and the microsphere form.

When swelling comes 1–2 weeks later

The manufacturer's guidance says that early swelling occurs within 72 hours and delayed swelling may appear 1–2 weeks later. Searches worrying about swelling in this period do indeed come up in several languages. However, we did not find published material reporting the frequency or duration of this delayed swelling. We set it down noting that it is the manufacturer's guidance.

Needle or cannula — which bruises less

For liquid PCL they have never been compared. The two published studies simply used a 31G needle and a 25G cannula respectively; there is no study comparing the two under the same conditions. “A cannula reduces bruising” is carried across from the received view in the general filler literature. We do not think that received view is wrong, but we do note that it has not been verified for this product. The general difference between needles and cannulas is set out in is Rejuran injected with a needle or a cannula.

So our answer is this. What you see over a few days is not effect but swelling. The earliest time point at which a change is recorded in the literature is 4 weeks, and the earliest at which increased collagen was confirmed in tissue is 6 weeks, in animals. The manufacturer advises “first change within 1–2 weeks”, but we did not find published measurements supporting this. How to set the point at which to judge a procedure is written about in more detail in when results actually start to appear, and when you should judge them.

For reference, there is human biopsy data for microsphere PCL — in a study in which 0.5 cc was injected once into the dermis of 13 people with biopsies at 2 weeks · 1 year · 4 years, temple skin thickness at 1 year had increased by 26.74% ± 9.26% (1,412 → 1,781 µm, p < 0.001 — this percentage is the mean of the individual increases and so differs slightly from the ratio of the two mean values), and fibroblasts · giant cells · new capillaries · new collagen were observed around the PCL particles. Being data on a product of a different formulation, it cannot be carried across to the liquid form as it stands.

Reading (2) and (3) as if they were (1) is the most common misunderstanding. The time it takes for a material to degrade in the body and the period over which you look better in the mirror are different stories. There is no material connecting the two ideas. The sentence “PCL degrades over 6–12 months, so the effect lasts that long too” is a leap of logic.

The manufacturer's consumer page also contains the statement that “phase 1 and phase 2 clinical trials have been completed and over 200 people took part”. We did not find the paper · abstract · trial registration number for that 200-person study anywhere. This does not mean it does not exist; it means there is no way to confirm it publicly.

If you have seen the figure “1 year” or “2 years”, that is usually data on microsphere PCL (Ellansé) — on that side there is a 24-month randomised trial and a 4-year biopsy. It is the same polycaprolactone but a product of a different formulation, and the data cannot be carried across between them.

Within the range we searched, we did not find a single case reporting a nodule or granuloma with liquid PCL. On the contrary, there are sources that expressly wrote that there were none.

• Case series (2 people) — “There was no bruising, visible or palpable lumps, discolouration, signs of vascular occlusion, granuloma/nodules or infection”
• Rat experiment (12 weeks) — the injected bolus dispersed within 6 hours and there was no granuloma formation thereafter, with no residual material or foreign-body reaction during the 12-week study period

But you should not read this as “nodules do not occur”. This product is distributed in 45–55 countries, and the number of people observed in the published literature is 33. The number observed is far too small relative to the number exposed. In this situation, “there are no reports” is not evidence of rarity but means there is no data, and calculating an incidence is itself impossible. This is the principle we state repeatedly on this site — “it has not been reported” and “it does not occur” are different things.

For microsphere PCL, case reports of delayed-onset nodules do in fact exist — a case treated under ultrasound guidance, a case of delayed hypersensitivity after COVID-19 vaccination and others have been published, and in a series of 1,111 cases lumps were reported at 0.45%. The manufacturer and conference presentations put forward the mechanistic logic that “because there are no particles, the risk of nodules formed by particles clumping is low”. It is a biologically reasonable hypothesis, but it has not been verified as an incidence in humans.

By design the target is the dermis to the dermal-subcutaneous junction. Unlike microsphere products, which create volume in the subcutaneous layer, the liquid form is made to spread widely through a shallow layer. Being a homogeneous aqueous solution with no particles, the needle does not block and there is no question of the suspension being uneven, is the explanation given in the development paper.

There is, however, a point where the numbers do not agree with one another. The case series used 0.2 mL per point for a total of 2 mL, while the manufacturer advises one 1 mL syringe over 10 points for the whole face, which comes to 0.1 mL per point. A twofold difference. And not one paper states the spacing between points in centimetres. It is more accurate to regard this as an area where there is as yet no standard.

Since the search “GOURI DEXLEVO” comes up, we will set this out. They are not two different procedures but the relationship between a company and a product.

• DEXLEVO — a Korean medical device company founded in 2013
• GOURI — the brand name of the liquid-form PCL product that company makes
• In every market we checked (Europe · Australia · Singapore · Taiwan · Korea) the brand name is the same, GOURI — the name does not differ from country to country
• There is a separate topical ampoule product under the same brand. It is a different item from the injectable, so please do not confuse the two

On the regulatory history, press reporting indicates that it first obtained the European CE mark · Australian · Singaporean certifications, and obtained Korean MFDS approval during 2025. Taiwan was in May 2026. An article from March 2025 refers to it only as “CE-certified, for export”, so the structure appears to be that the export-only approval came first and the domestic one was attached later. How domestic and export-only approvals differ is set out in what is different between domestic and export-only.

We will write this honestly. In writing this piece we were unable to obtain the original text of GOURI's Korean marketing approval number · item name · intended use. Both the MFDS electronic civil application window and the integrated information system block automated lookups, and access to the manufacturer's Korean site was blocked as well. Nor did a web search turn up a page stating the approval number. We confirmed from press reporting that a domestic approval exists, but we were not able to confirm the original text of the number.

So this is what you can do.

• Look at the product box. Article 20 of the Medical Devices Act places on the manufacturer · importer a duty to write the approval number on the container or outer packaging. If the product came through normal distribution, the number is on the box.
• You may also ask the clinic. “Please tell me the product's approval number” is an entirely normal verification step.
• Once you have the number, look it up with no spaces on the MFDS medical device information portal and check that the “Export only” field in the detailed information reads “No”. The whole method of checking is in what one line of approval number can and cannot tell you.

Searching by brand name generally returns nothing. That is because the “product name” field in the MFDS database is empty for many items — including perfectly normal domestic products. Always look it up by approval number.

What we most want to say in this piece is neither “do not have it” nor “have it”. It is that this is a procedure you look at from a month later rather than the day after, and that the period the material we now have actually watched is 6 months. Start knowing those two things and there is less chance of expectation and outcome diverging. And if you have been told at some clinic that “it lasts a year”, you are entitled to ask where that number comes from.

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.