Carbon dioxide outperformed erbium on acne scars, and the patient pays for it in downtime

Design

Systematic review and meta-analysis of comparative clinical studies, searched to April 2026

Population

130 analysed participants across five studies with atrophic acne scars

Primary outcome

Clinical response of more than 50% improvement

Effect

Erbium vs carbon dioxide RR 0.69 (95% CI 0.49-0.97, I-squared 45.9%); randomised-only RR 0.76 (0.54-1.08); pain MD 1.86 (1.33-2.39); downtime MD 3.67 days (1.90-5.44)

Five comparative studies with 130 analysed participants were pooled, four of them contributing to the efficacy comparison, harmonised around a single outcome: more than 50% clinical improvement in atrophic acne scarring. Erbium-based fractional lasers - both Er:YAG and erbium-glass - were less likely than fractional carbon dioxide to reach that threshold, with a relative risk of 0.69 (95% CI 0.49-0.97, I-squared 45.9%). Restricting to randomised studies preserved the direction but not the significance: RR 0.76 (95% CI 0.54-1.08).

The tolerability comparison went the other way and was more consistent. Carbon dioxide produced more treatment-related pain (mean difference 1.86, 95% CI 1.33-2.39) and 3.67 more days of downtime (95% CI 1.90-5.44). Post-inflammatory hyperpigmentation did not differ significantly (OR 1.75, 95% CI 0.67-4.57), nor did acneiform events (OR 0.73, 95% CI 0.12-4.25).

The consultation this informs is a trade, and the meta-analysis is honest that it is one. A patient who can absorb four extra days away from work and wants fewer sessions is a reasonable carbon dioxide candidate; someone who cannot is not being short-changed by erbium, whose disadvantage narrowed to non-significance in the randomised evidence. The pigmentation comparison is the one to treat carefully in Indian practice - a confidence interval from 0.67 to 4.57 in 130 patients does not establish that the two devices are equivalent for darker skin.

• Quote downtime in days, not sessions - 3.67 extra days is the difference a patient actually plans around
• Do not tell a patient the two devices carry equal pigmentation risk; the data cannot support that
• Discuss pain expectations before the first pass, particularly for carbon dioxide
• Where work absence is the binding constraint, erbium is a defensible choice, not a compromise
• Test spot and photograph before starting a course, whichever device you use

Why it matters

The choice between these two devices has been made on habit and availability; this gives it an actual trade-off to weigh.

Don't overread it

The efficacy advantage lost significance when only randomised studies were pooled.

The statistics, in plain English

The primary efficacy result, RR 0.69 with an upper bound of 0.97, only just excludes the null, and when the analysis was restricted to randomised studies the interval widened to include it (0.54 to 1.08). That pattern - significance that depends on including non-randomised studies - means the efficacy difference is suggestive rather than established. The safety comparisons are different in kind: intervals from 0.67 to 4.57 for hyperpigmentation and 0.12 to 4.25 for acneiform eruption are so wide that they exclude almost nothing. Absence of a significant difference in 130 patients is not evidence of no difference.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.