A reader emailed me last year. Fitzpatrick V, mid-thirties, on tretinoin 0.05% nightly for nine weeks. She wanted to discuss why her hyperpigmentation was worse, not better.
This is a familiar conversation. Tretinoin is the gold-standard retinoid for hyperpigmentation, and tretinoin is one of the more common iatrogenic causes of hyperpigmentation in darker skin tones. Both statements are true, and the contradiction is the whole story.
Most published coverage hands SoC patients a version of “go slow” and stops there. Going slow is the right advice. What it doesn’t explain is why slow, what slow actually means by week, and the specific signs you’ve crossed the line from therapeutic exposure to net-negative exposure. The mechanism explains all three.
What “irritation-induced PIH” actually is
Tretinoin’s therapeutic action is the same in every skin tone: it binds retinoic acid receptors in keratinocytes, normalizes follicular hyperkeratinization, and over months, accelerates dermal-epidermal junction turnover. The hyperpigmentation benefit comes from that turnover plus a direct downregulation of tyrosinase expression in melanocytes.
The problem is the side effect pathway. Tretinoin causes a low-grade inflammatory response that, in Fitzpatrick I-III skin, presents as visible erythema and tolerable scaling. In Fitzpatrick IV-VI skin, the same inflammation triggers a parallel pathway: PAR-2 (protease-activated receptor 2) activation on keratinocytes, which accelerates melanosome transfer from melanocytes. Compound that with the upregulated α-MSH (alpha-melanocyte-stimulating hormone) signaling that follows any sustained dermal irritation, and you have melanocyte hyperactivation in exactly the skin types that have the most melanocyte machinery to begin with.
Halder and Richards summarized this pathway in 2004 (Am J Clin Dermatol) and the model hasn’t changed. The clinical consequence has: SoC patients started on retinoids without dose-modification protocols develop irritation-PIH that takes 6-12 months to fade — meaning a treatment course aimed at lightening pigment over 16 weeks instead leaves patients darker for a year.
The concentration question
The 0.025% / 0.05% / 0.1% strength options are presented in pharmacy guidance as interchangeable starting points. They aren’t.
For Fitzpatrick IV-VI skin, 0.025% is the only defensible starting concentration. Grimes and colleagues’ 2002 work (Cutis) used 0.05% in darker skin and reported a meaningfully higher PIH risk than the same trial design at 0.025% — and importantly, the efficacy difference between the two concentrations was modest. You don’t get linear potency for the linear irritation cost.
The escalation path is conservative. 0.025% three times weekly for the first 4 weeks. If tolerance is established — no visible erythema 24 hours post-application, no scaling at the nasolabial folds — move to every-other-night. Stay there for another 4 weeks. Only at week 8-12 does daily 0.025% become reasonable. Concentration escalation to 0.05% should be deferred to month 6 at minimum, and isn’t necessary for most therapeutic goals.
This is slower than every published protocol I’ve read. Every published protocol I’ve read includes patient populations that drop out, and the drop-outs are disproportionately SoC patients whose visible PIH made them stop. A protocol that nobody finishes isn’t a protocol that works.
The buffering protocols that work
Three techniques meaningfully reduce irritation without compromising therapeutic exposure.
Short-contact therapy. Apply tretinoin, leave on for 30-60 minutes, wash off with a non-foaming cleanser. The keratinocyte uptake of retinoic acid happens within the first 20-30 minutes; sustained skin contact past that point delivers cumulative irritation without proportional benefit. Short-contact gets you most of the efficacy at maybe 40% of the cumulative irritation.
Sandwich method. Moisturizer, wait 15 minutes, tretinoin, wait 15 minutes, moisturizer again. The buffering reduces transepidermal water loss during the exposure window, which is the main driver of the irritation cascade. Ceramide-forward moisturizers (CeraVe PM, Vanicream, La Roche-Posay Toleriane) outperform humectant-only options here.
Every-third-night ramping. Most dermatology guidance says every other night to start. For Fitzpatrick V-VI, every-third-night for 4-6 weeks before stepping to every-other-night handles the inflammation budget better. The total weekly dose matters less than peak-day inflammation, which is what triggers melanocyte response.
When to stop and switch
The clinical heuristic I use: if PIH appears or visibly worsens within the first 4 weeks of tretinoin exposure, the protocol has crossed the inflammation tipping point. Lowering frequency rarely recovers; stopping and stepping down to a milder retinoid (adapalene 0.1% gel, retinaldehyde 0.05-0.1%) does.
This is the moment most coverage fails SoC patients. “Push through the irritation” is reasonable advice for Fitzpatrick I-III. It’s actively harmful in IV-VI, where the irritation phase is also a PIH-induction phase. You’re not purging through to clear skin. You’re potentially creating 6-12 months of additional pigment to treat.
Adapalene as a stepdown, not a substitute
Adapalene 0.1% gel has a meaningfully lower inflammation profile than tretinoin at equivalent efficacy for acne. For pigmentation it’s less effective. The receptor binding profile is different — adapalene is RAR-β selective, while tretinoin binds RAR-α, RAR-β, and RAR-γ.
This makes adapalene the right pivot when tretinoin’s irritation has exceeded benefit, not the right starting point if pigmentation is the primary goal. The exception: pigmentation that’s predominantly post-acne (acne PIH specifically) often responds adequately to adapalene without ever needing tretinoin, because the underlying acne control is doing most of the work.
What a 16-week protocol actually looks like
Weeks 1-4: Tretinoin 0.025%, three nights weekly, with sandwich method and short-contact (45 minutes). Daily ceramide moisturizer, mineral SPF every morning. Photograph baseline at week 0.
Weeks 5-8: If tolerated — no erythema, no PIH worsening, no scaling — move to every-other-night, same concentration, same technique. Add 4% niacinamide AM serum for melanosome-transfer inhibition support.
Weeks 9-12: Every-other-night transitions to nightly. Short-contact ramps up to 60-90 minutes. Photograph progress.
Weeks 13-16: Nightly tretinoin, full contact (sleep with it on), but only if there’s been zero visible irritation through week 12. This is the conservative path. Many patients stop at every-other-night indefinitely and still get the therapeutic benefit they wanted.
The 16-week mark is when meaningful pigment improvement should be visible. Photograph and compare. If improvement is present and no PIH has appeared, the protocol is working as designed.
What mainstream coverage gets wrong
The published consumer guidance falls into two camps. Camp 1 says tretinoin is safe for all skin tones, just use sunscreen. Camp 2 says darker skin tones should avoid retinoids and use gentler alternatives. Neither is right.
Tretinoin works in SoC patients. It works better than most alternatives for actual pigmentation outcomes when used correctly. The barrier to that outcome is the protocol, not the molecule. Most SoC retinoid failures I see clinically are protocol failures — too fast, too high, too aggressive — not molecule failures. Fix the protocol and the molecule does what it’s supposed to.
FAQ
*Is tretinoin safe for dark skin?*
Yes, with concentration and frequency modifications. The 0.025% starting concentration with a slow 12-16 week titration is well-supported for Fitzpatrick IV-VI patients. The risk is irritation-induced PIH, which is preventable with the right titration, not a contraindication to the molecule itself.
*What concentration of tretinoin should I start with if I’m Fitzpatrick V?*
0.025%, three times weekly, with short-contact and sandwich technique. Higher concentrations and faster escalation increase PIH risk without proportional efficacy gain.
*Why does tretinoin sometimes make my hyperpigmentation worse?*
The inflammation tretinoin causes triggers PAR-2 activation and α-MSH signaling in melanocytes, accelerating melanosome transfer. In Fitzpatrick IV-VI skin, that pathway often outpaces tretinoin’s direct tyrosinase downregulation in the first 8-12 weeks. The visible PIH usually appears at weeks 3-6.
*How long until I see results with tretinoin on darker skin tones?*
Meaningful pigment improvement at 12-16 weeks on a slow titration. Faster timelines published in dermatology trials typically used Fitzpatrick I-III cohorts and don’t translate directly.
References
- Halder RM, Richards GM. Topical agents used in the management of hyperpigmentation. Am J Clin Dermatol. 2004. PubMed.
- Grimes PE. A four-month randomized, double-blind evaluation of the efficacy of 0.05% tretinoin cream in the treatment of melasma in darker phototypes. Cutis. 2002. PubMed.
- Davis EC, Callender VD. Postinflammatory hyperpigmentation: a review of the epidemiology, clinical features, and treatment options in skin of color. J Clin Aesthet Dermatol. 2010. PubMed.
Khabir Uddin is the founder of Elelaf Journal. This article is editorial, not medical advice. For active hyperpigmentation or refractory cases, see a board-certified dermatologist with experience in skin of color.