Jean-Auguste-Dominique Ingres painted Madame Moitessier twice, in 1851 and again in 1856. In both portraits her skin is rendered so evenly that no pore interrupts it. It is the kind of surface people carry in their heads when they walk into a clinic and ask about "skin Botox."

Skin Botox, also called intradermal botulinum toxin or microbotox, is not the familiar injection into a facial muscle. Tiny amounts of botulinum toxin type A are placed into the skin itself, in a grid of small points across the cheeks, with the aim of reducing oil and making enlarged pores look smaller. Patients ask me about it in exactly those terms: shine by the afternoon, pores that show in photographs.

My argument is that skin Botox probably does reduce facial oil and may make pores look smaller, but the studies are small, short and inconsistent, and nothing cited here shows that it beats cheaper options. That is a reasonable thing to offer a patient. It is a smaller claim than the name implies. Four papers carry the case: a 2024 meta-analysis, two recent studies in the Journal of Cosmetic Dermatology, and a review of the proposed mechanism.

How does skin Botox work on oil glands? The proposed mechanism

Skin Botox is botulinum toxin type A injected into the dermis, the layer of skin just under the surface, rather than into muscle. The proposed reason it lowers oil is that sebaceous (oil) glands respond to acetylcholine, the same chemical signal the toxin blocks (Rho & Gil, 2021).

In a narrative review that drew on 61 citations, Rho and Gil describe acetylcholine receptors on sebocytes, the cells that make sebum, including α7 nicotinic and M2 muscarinic receptors. In cultured sebocytes, acetylcholine increased lipid synthesis in a dose-dependent way through ERK signaling (Rho & Gil, 2021). From this the authors propose that the toxin may reduce sebum by interfering with cholinergic transmission between the glands and nearby autonomic nerve terminals (Rho & Gil, 2021).

The word to hold on to is proposed. This is a review, not a trial, and it presents the mechanism as a hypothesis built from cell studies and clinical results. The 2026 study discussed below describes it in similar terms: inhibition of acetylcholine release, which plays a role in sebocyte differentiation and gland activity (Dominguez-Salgado et al., 2026).

Does skin Botox reduce oil and enlarged pores? What the pooled data show

Probably yes for oil, and possibly for pores, but the pooled evidence is too thin to call either settled. A 2024 systematic review and meta-analysis in Plastic and Reconstructive Surgery Global Open, the sister journal of Plastic and Reconstructive Surgery, pooled 10 studies: 5 randomized controlled trials and 5 prospective cohort studies, with 153 participants in total (mean age 41 years, 92.1% female) (Rahman et al., 2024). The searches covered January 2008 to March 2023 (Rahman et al., 2024).

The effects are reported as standardized mean differences (SMD), a way of putting different measurement scales on one footing. For sebum production, pooled from 4 studies, the SMD was −1.07 (95% CI, −2.04 to −0.09; P = 0.03), where a negative value means less sebum. For pore size, also from 4 studies, it was −2.34 (95% CI, −3.90 to −0.78; P = 0.003) (Rahman et al., 2024). The review also reported improvement in skin texture (SMD −0.52), wrinkles (−0.57) and erythema (−0.72). Skin hydration did not improve significantly (SMD −0.23; 95% CI, −0.66 to 0.19; P = 0.26) (Rahman et al., 2024).

Two further numbers deserve more attention than the headline. The first is heterogeneity, the degree to which studies disagree with one another. I² was 88% for sebum and 94% for pore size (Rahman et al., 2024). The studies used different toxin products, dilutions and doses, from 15 units into one cheek to 10 units in total per patient (Rahman et al., 2024).

The second is the review's own trial sequential analysis, which estimates how much data would be needed before the result can be trusted. The required information size (roughly, the sample size needed) was 189 for sebum and 2,499 for pore size, and none of the outcomes reached its required size, in a review whose participants totalled 153 (Rahman et al., 2024). Three of the included trials were sponsored by manufacturers. Removing them one at a time did not change the results, and the authors found no significant difference between products or dilutions (SMD −0.29; P = 0.42) (Rahman et al., 2024). The randomized trials were rated at moderate risk of bias (Rahman et al., 2024).

The authors' own conclusion was measured: positive outcomes across several skin-quality attributes, and a need for more high-quality research (Rahman et al., 2024).

What does a recent enlarged-pore study show? Six weeks in 30 patients

In a 2026 prospective study in the Journal of Cosmetic Dermatology, oil and measured pore size both fell at six weeks, in a design with no untreated comparison group (Dominguez-Salgado et al., 2026). Thirty adults aged 18 to 35 with oily skin and enlarged pores were treated at one center in Mexico City (70% women, median age 27 years) (Dominguez-Salgado et al., 2026).

The technique shapes how to read the numbers. Five points were marked on each cheek, and 3 units were injected intradermally at each point with a 30-gauge needle at about 45 degrees, raising a small wheal, for 15 units per cheek in total (Dominguez-Salgado et al., 2026). The product was Dysport, from a 500-unit vial diluted in 7 mL of saline (Dominguez-Salgado et al., 2026). Units are specific to each product, which is one reason doses vary so widely between studies.

Sebum was measured with oil-absorbing sheets. The area of oil on the sheet fell from 291.9 ± 163.69 to 191.6 ± 80.39 mm² on the right cheek (34.4%; P = 0.0016) and from 290.5 ± 171.92 to 187.1 ± 93.22 mm² on the left (35.6%; P = 0.0011) (Dominguez-Salgado et al., 2026). Pore size, imaged with an OCULUS Keratograph 5 M across 12 pores per cheek, fell from 531.3 ± 130.54 to 235.3 ± 60.78 μm on the right (55.7%) and from 554 ± 144.90 to 241.3 ± 77.71 μm on the left (56.5%), both P < 0.0001 (Dominguez-Salgado et al., 2026). Patient-reported satisfaction with appearance (FACE-Q) rose from 50.87 to 70.07, and a skin-quality complaint score (SASS-Q) fell from 2.1 to 1.47 (Dominguez-Salgado et al., 2026).

No serious adverse events were observed. Mild, brief injection-site discomfort and redness resolved on their own (Dominguez-Salgado et al., 2026).

Does the way the toxin is delivered matter? Intradermal injection versus microneedling

In a 2025 split-face randomized trial, intradermal injection and microneedling-assisted delivery of the toxin improved enlarged pores to a similar degree (Iraji et al., 2025). The trial, also in the Journal of Cosmetic Dermatology, enrolled 30 patients (29 women, mean age 34.2 years). One side of each face received 50 units by intradermal injection, and the other side was treated with microneedling followed by 50 units of toxin (Iraji et al., 2025).

At one month, three blinded dermatologists scored improvement on a quartile scale using dermoscopy at 10× magnification and physical examination. There was no significant difference between the two methods for the cheeks or the nose (P > 0.05). For example, cheek dermoscopy scores were 1.2 ± 0.5 and 1.1 ± 0.6 (P = 0.69), and patient satisfaction was 3.8 ± 0.8 against 3.5 ± 0.7 on a five-point scale (P = 0.13) (Iraji et al., 2025). Side effects were small: brief swelling on the injected side that resolved within 2 hours, mild redness on the microneedled side that settled within 24 hours, and no post-inflammatory hyperpigmentation (Iraji et al., 2025).

Read this result carefully. Both sides of every face received botulinum toxin. The trial compares two ways of delivering it, and it cannot say how much of the improvement comes from the toxin itself (Iraji et al., 2025). The authors list their own limits: 30 patients, almost all women, one month of follow-up, subjective scoring, and no objective sebum or 3D imaging measures (Iraji et al., 2025).

Is skin Botox better than retinoids or other oil treatments?

No paper cited here shows that skin Botox is more effective than topical retinoids or other established treatments. Rho and Gil state plainly that there is no evidence of superiority over topical retinoids, and they recommend keeping botulinum toxin as a secondary option in acne management until more trials confirm the early results (Rho & Gil, 2021). In one study they cite, the sebum reduction at one month after abobotulinumtoxinA was comparable to that after oral isotretinoin (Rho & Gil, 2021).

They also note that most trials enrolled only 10 to 25 patients, that the toxin is not labeled for acne, and that repeat treatments are needed (Rho & Gil, 2021). In the studies they cite, the effect of abobotulinumtoxinA on sebum lasted up to six months, while onabotulinumtoxinA effects typically lasted three to four months (Rho & Gil, 2021). Reported side effects were generally mild, mainly temporary stinging and possible bruising, with no allergic reaction, facial palsy or severe paralysis in the studies reviewed (Rho & Gil, 2021).

The strongest objection

Some will say that pore size fell by more than half in the 2026 study, and that the pooled effect on pores is the largest in the meta-analysis, so the case is closed. That is partly true. Those are the numbers, and they point in one direction (Dominguez-Salgado et al., 2026; Rahman et al., 2024).

But each source has limits. The 2026 study had no control group and six weeks of follow-up, made multiple comparisons without adjustment, and analyzed each person's two cheeks as separate measurements. The authors themselves describe it as supportive observational data rather than definitive evidence (Dominguez-Salgado et al., 2026). The pore-size result in the meta-analysis came from four studies with an I² of 94% and a required information size above 2,000 (Rahman et al., 2024). Three included trials were manufacturer-sponsored, and several of the review's authors report ties to toxin manufacturers (Rahman et al., 2024). The one randomized comparison cited here had no untreated side and one month of follow-up (Iraji et al., 2025).

Reassurance and uncertainty are both real here.

What should you ask before skin Botox?

A good consultation about skin Botox starts with what is being treated. The pooled data support less oil and, more tentatively, smaller-looking pores, and they did not show a hydration benefit (Rahman et al., 2024). A patient whose main concern is dryness or dullness should know that.

It is reasonable to ask which product and dose will be used and why, because doses in the literature ranged widely (Dominguez-Salgado et al., 2026; Iraji et al., 2025; Rahman et al., 2024). It is also fair to ask how oil and pores will be judged afterward, and what happens beyond the one-month and six-week follow-up of the recent studies (Dominguez-Salgado et al., 2026; Iraji et al., 2025). Reviews describe effects that last for months and need repeating, with variation between products (Rho & Gil, 2021). Ask how the plan compares with a topical retinoid, since no study here shows an advantage (Rho & Gil, 2021). And expect minor local reactions such as stinging, bruising, brief swelling or redness (Dominguez-Salgado et al., 2026; Iraji et al., 2025; Rho & Gil, 2021).

The Clinical Bottom Line

Skin Botox is a reasonable option for people with oily skin and enlarged pores who understand what the evidence does and does not show. In pooled data, sebum and pore appearance improved (Rahman et al., 2024). A 2026 prospective study reported lower sebum and smaller measured pores at six weeks (Dominguez-Salgado et al., 2026), and reported side effects were mild and brief across the sources (Dominguez-Salgado et al., 2026; Iraji et al., 2025; Rho & Gil, 2021).

The caution is about certainty. The studies are small and short, the pooled analysis did not reach the sample size it called for, and nothing here shows an advantage over retinoids. What we still lack are larger controlled trials with an untreated or placebo comparison, longer follow-up, and consistent dosing across products.

Ingres could paint skin without a single pore. Skin Botox cannot, and a good consultation starts by saying so.

For clinical assessments and consultation in Seoul: itsdrbock.com

References

Dominguez-Salgado, L. A., Ramirez-Hernández, R., & Nava-Castañeda, Á. (2026). Prospective Keratograph 5 M assessment following intradermal botulinum toxin type A for seborrhea and enlarged facial pores. Journal of Cosmetic Dermatology, 25(7), Article e71073. https://doi.org/10.1111/jocd.71073

Iraji, F., Moeini, R., Abedini, M., Mousavi, M. S., Saber, M., Rahimi Barghani, M. R., Sagheb Ray Shirazi, M., & Seyedyousefi, S. (2025). Comparison of intradermal versus microneedling-assisted botulinum A toxin injection for enlarged facial pores: A randomized clinical trial. Journal of Cosmetic Dermatology, 24(6), Article e70114. https://doi.org/10.1111/jocd.70114

Rahman, E., Rao, P., Philipp-Dormston, W. G., Webb, R., Garcia, P. E., Ioannidis, S., Kefalas, N., Kajaia, A., Friederich, L., Yu, N., Wang, K., Parikh, A., Trindade de Almeida, A. R., Carruthers, J., Carruthers, A., Mosahebi, A., Wu, W. T. L., & Goodman, G. (2024). Intradermal botulinum toxin A on skin quality and facial rejuvenation: A systematic review and meta-analysis. Plastic and Reconstructive Surgery Global Open, 12(8), Article e6084. https://doi.org/10.1097/GOX.0000000000006084

Rho, N.-K., & Gil, Y.-C. (2021). Botulinum neurotoxin type A in the treatment of facial seborrhea and acne: Evidence and a proposed mechanism. Toxins, 13(11), Article 817. https://doi.org/10.3390/toxins13110817