IPL: the first of fifteen procedure pages on this site with **an untreated control group**

A 2020 prospective randomised controlled trial in China. Patients with telangiectasia in late-stage rosacea were recruited; all received anti-mite therapy and were then randomised into two groups.

• IPL group: three 540 nm IPL treatments at four-week intervals
• Control group: no treatment

The same clinician assessed them across two years. Thirty-three were lost to follow-up, leaving 107 in the IPL group and 120 in the control group for final analysis.

The results are clear.

• Effective (≥90% clearance of telangiectasia): IPL 66.36% against control 0%
• Total efficacy (including ≥30% clearance): IPL 95.33% against control 30.83%
• Recurrence rates were also lower in the IPL group.
• Adverse effects: reactions in 9.7% subsided within a week; hyperpigmentation (1.9%) within three months.

Sit with that 0% for a moment. What we have kept pointing out on other procedure pages is that 'better than baseline' cannot be separated from 'better because of the treatment'. This design separates them.

The limits go down as written too.

• The assessor was not blinded. 'The same clinician' assessed, and that clinician would have known who received IPL. Blinding is genuinely hard here, since treatment leaves visible traces.
• Thirty-three were lost to follow-up. Out of 260, that is not a small number.
• All participants tested positive for Demodex and received anti-mite therapy alongside. The result does not transfer outside those conditions.

We have attached the 2019 GRADE systematic review on rosacea alongside, so the judgement does not rest on one trial.

The 2010 trial has the same virtue. Twenty-four Chinese women with photoaging received four IPL treatments at 3–4 week intervals on one side of the face only, with the other side spared as control.

Same person, same span of time, same sun. The only difference is that one side got IPL.

The results:

• Global photoaging scores on the treated side fell from 3.02 to 1.22, while the untreated side remained unchanged.
• 21 of 24 (87.5%) rated their improvement as excellent or good.
• Changes in melanin index and erythema index were significantly larger on the treated side — after the first session, after the fourth, and at three-month follow-up (P < .05).

And the trial goes one step further. After four sessions, skin biopsies were taken.

• Melanin content decreased significantly, and
• collagen fibres obviously increased,
• only on the treated side (P < .05).

This is a rare combination for this site: an untreated control, instrument measurement and histology inside one trial.

The limit is scale. Twenty-four people, three months of follow-up. That the participants had darker phototypes is arguably a strength — IPL also targets melanin, so risk is higher in darker skin, and this was demonstrated under exactly that condition.

This item is not about a result. It is about the name.

IPL is not a particular device but the name of an approach. Which filter is used changes the wavelength range emitted, and pulse duration, energy and cooling differ by device. The two trials above used a 540 nm filter and another setting respectively.

The 2020 study compared short-pulse IPL against pulsed dye laser for facial redness. Trials comparing two devices like this keep appearing precisely because which is better is not settled. We could not confirm its detailed figures from the abstract, so sample size and duration are left empty.

The practical consequence: two treatments given under the name 'IPL' may be quite different conditions, and there is no guarantee that the results above carry over to other settings.

It is the same story this site keeps telling in the ingredient layer — concentration and vehicle matter more than the name. For device procedures, wavelength and settings occupy that slot.