Around 1635, Peter Paul Rubens painted The Three Graces. Three nude women stand close together, and their flesh dimples, folds and gathers the way flesh does. Rubens did not smooth them into marble. He painted bodies that had been lived in.

Patients who ask me about stretch mark treatment often want the opposite of what Rubens painted. Stretch marks, called striae distensae in medical language, are thin linear bands that usually appear on the abdomen, breasts, buttocks and thighs, most often during pregnancy, weight gain or a growth spurt (Sun et al., 2024). Estimates of how common they are differ: a 2026 review reports an overall incidence of 56% (95% CI, 52% to 59%) (Wu & Wang, 2026), while a 2024 meta-analysis states that about 90% of pregnant women develop them (Aktoz & Yilmaz, 2024).

My argument is that stretch mark treatment can improve the marks, but the two best-studied devices, microneedle radiofrequency and fractional CO2 laser, look more alike than different on results, so the real choice between them is which side effect you would rather risk: pain or pigment change. Three papers carry the case: a 2026 review in the Journal of Cosmetic Dermatology, a 2024 systematic review and meta-analysis of microneedling in Aesthetic Plastic Surgery, and a 2024 systematic review and meta-analysis comparing the two devices in Lasers in Medical Science.

What are stretch marks, and why are they hard to treat?

Stretch marks are scars in the dermis, the deeper layer of skin, that form when skin is stretched faster than it can adapt. They begin as smooth or slightly raised pink or red bands (striae rubrae) and, over several months, become thin, wrinkled, pale bands (striae albae) (Sun et al., 2024). The 2026 review describes broken collagen fibers and fewer elastic fibers in the dermis, and names mechanical tension, hormonal change, genetic susceptibility and inflammation as contributors (Wu & Wang, 2026).

That is why surface treatments struggle. Topical tretinoin works better on early red marks than on white ones and can cause irritation such as redness and peeling (Wu & Wang, 2026). The same review notes that stretch marks can carry a psychological cost, including anxiety and depression (Wu & Wang, 2026).

Does microneedle radiofrequency beat fractional CO2 laser?

The two meta-analyses reach different answers on effectiveness, and the difference is informative. The 2024 microneedling meta-analysis pooled 11 controlled studies (6 randomized and 5 non-randomized) found in four databases through September 2023. In the subgroup of four studies comparing microneedle radiofrequency (heat delivered through tiny needles) with lasers, clinical improvement favored radiofrequency (standardized mean difference, SMD, 0.57; 95% CI, 0.20 to 0.94; P = 0.003; I² = 0%) (Sun et al., 2024). An SMD puts different scoring scales on one footing. Patient satisfaction between radiofrequency and laser was comparable (P = 0.26) (Sun et al., 2024).

The 2024 meta-analysis of the two devices asked a narrower question. It found only three prospective cohort studies comparing microneedle radiofrequency directly with fractional CO2 laser, and it counted a lesion as improved when improvement exceeded 50%. On that yardstick, professional assessment showed no significant difference (odds ratio, OR, 1.27; 95% CI, 0.49 to 3.31), and neither did patients' own assessment (OR 0.72; 95% CI, 0.13 to 3.90) (Aktoz & Yilmaz, 2024). The comparisons were small, with 10 to 17 patients each, and two of the three treated different areas of the same patient (Aktoz & Yilmaz, 2024).

The gap likely reflects what was measured and what the comparators were. Sun and colleagues compared continuous scores, and their four radiofrequency studies used three different lasers as comparators: fractional CO2, Er:YAG and 1565 nm non-ablative fractional (Sun et al., 2024). Aktoz and Yilmaz looked only at CO2 laser and used the 50% cutoff (Aktoz & Yilmaz, 2024). The 2026 review reports both findings and describes the two approaches as having similar effects on collagen remodeling in white stretch marks (Wu & Wang, 2026).

Which treatment hurts more, and which changes skin color more?

Microneedling hurts more than laser, and fractional CO2 laser causes more pigment change. In the microneedling meta-analysis, dyspigmentation (mostly post-inflammatory hyperpigmentation, PIH, the brown discoloration that can follow skin injury) was less frequent with microneedling than with other treatments (OR 0.14; 95% CI, 0.05 to 0.41; P = 0.0003). Procedure pain was significantly greater than with laser (P < 0.00001), based on three studies with complete pain data, all with laser comparators (Sun et al., 2024). The device-specific analysis agrees on pigment: PIH was less frequent after microneedle radiofrequency than after fractional CO2 laser (OR 0.24; 95% CI, 0.08 to 0.70) (Aktoz & Yilmaz, 2024).

Serious harm was not reported. No severe or long-lasting adverse effect appeared in the eligible trials, mild effects included pain, transient redness, swelling, itching, pinpoint bleeding and scaling, and no participant withdrew because of intolerable pain (Sun et al., 2024). One participant did stop at a fractional laser site because of PIH, and another withdrew consent for redness on the radiofrequency side, which resolved after one pulsed dye laser session (Sun et al., 2024).

Skin type matters here. Where trials reported it, participants were mostly Fitzpatrick skin phototypes III to V (Sun et al., 2024), and the 2026 review adds that the risk of pigmentation after laser is higher in darker skin (Wu & Wang, 2026).

What about plain microneedling, PRP and other options?

Plain microneedling without radiofrequency looked comparable to laser and better than older non-laser options. Against lasers, the difference was not significant (P = 0.35; I² = 54%, moderate heterogeneity; two studies). Against non-laser treatments such as microdermabrasion and a liquid nitrogen cryoroller, microneedling improved clinical scores more (SMD 1.01; 95% CI, 0.51 to 1.51; three studies) (Sun et al., 2024).

The 2026 review describes combination approaches as the direction of the field. It reports that fractional CO2 laser combined with platelet-rich plasma (PRP) produced 50% to 75% improvement in most patients in the studies it summarized, and that 1565 nm non-ablative laser plus a topical beta-glucan improved atrophy scores at 12 weeks compared with single treatments in a randomized trial (Wu & Wang, 2026). Those are individual small studies, and the review itself calls for larger trials and standardized outcome measures (Wu & Wang, 2026).

The strongest objection

Some will say the answer is already clear: radiofrequency microneedling scored better in one meta-analysis and caused less pigment change in both, so choose it. That is partly true, and pigment is a real reason to lean that way in darker skin. But the efficacy result is not consistent across the two analyses (Aktoz & Yilmaz, 2024; Sun et al., 2024).

The trials were small, with no more than 40 participants in each, and most followed patients for only one to three months (Sun et al., 2024). The 2026 review notes that most studies observe patients for 3 to 6 months, which makes recurrence impossible to judge (Wu & Wang, 2026). In the microneedling analysis, four of the five non-randomized studies carried moderate risk of bias and one carried high risk, and the included populations came mainly from Asia and Africa (Sun et al., 2024). The device comparison did not test for publication bias because too few studies were available (Aktoz & Yilmaz, 2024).

How should you choose a stretch mark treatment?

Start with the stage of the marks and your skin type. Early red marks and old white ones behave differently, and the 2026 review states that the effects of age, skin type and stage on outcomes have not been fully clarified (Wu & Wang, 2026). If your skin tends to darken after injury, the pigment data favor microneedle radiofrequency over fractional CO2 laser (Aktoz & Yilmaz, 2024; Sun et al., 2024).

Then ask how success is defined. The device meta-analysis counted improvement above 50% as a positive result, a threshold the authors chose (Aktoz & Yilmaz, 2024). None of the three papers I read reports complete removal as an outcome.

Finally, ask what happens after the last session. With follow-up this short, no one can yet tell you how long the improvement lasts (Sun et al., 2024; Wu & Wang, 2026).

The Clinical Bottom Line

Stretch mark treatment with microneedle radiofrequency or fractional CO2 laser can improve the marks, and the two look broadly similar on effectiveness. A 2024 meta-analysis found no significant difference in the share of lesions improved by more than 50% (Aktoz & Yilmaz, 2024), a second favored radiofrequency on clinical scores (Sun et al., 2024), and both found less pigment change with radiofrequency, at the price of more pain (Aktoz & Yilmaz, 2024; Sun et al., 2024).

The caution is about everything the evidence cannot yet say. The studies are small, short and heterogeneous, the review that frames them calls for standardized measurement and long-term follow-up, and the choice among devices still depends on the individual (Wu & Wang, 2026).

Rubens painted the Graces with their folds and dimples intact. A good consultation does something similar: it says what can be softened, what it may cost you in pain or pigment, and what the evidence cannot yet promise.

For clinical assessments and consultation in Seoul: itsdrbock.com

References

Aktoz, F., & Yilmaz, N. (2024). Comparing fractional microneedle radiofrequency and fractional CO2 laser for striae distensae treatment: A systematic review and meta-analysis. Lasers in Medical Science, 39, Article 271. https://doi.org/10.1007/s10103-024-04231-8

Sun, X., Jia, X., & Huang, L. (2024). Microneedling therapy for striae distensae: Systematic review and meta-analysis. Aesthetic Plastic Surgery, 48(15), 2915–2926. https://doi.org/10.1007/s00266-024-03954-x

Wu, Y., & Wang, H. (2026). Advances in the treatment of striae distensae. Journal of Cosmetic Dermatology, 25(1), Article e70683. https://doi.org/10.1111/jocd.70683