From the Journals

Review looks at effectiveness of isotretinoin for acne

View on the News

Cochrane Review process is ‘fatally flawed’

I think dermatologists would agree with the assertions of the Cochrane Review that there is limited evidence according to today’s standards of quality evidence. The Cochrane Review process examines available data, ranking it by the quality of the data. The paper by Costa et al. shows where the Cochrane Review process is fatally flawed. It assumes that, where the highest quality data is lacking, the medication in question must be ineffective or that this at least casts doubt on its efficacy.

Isotretinoin was approved for use in acne in the 1980s. If you look at one initial report by Peck et al. examining the efficacy of isotretinoin versus placebo in patients with very severe nodulocystic acne, the isotretinoin group was so effective that half of the patients in the placebo arm broke code because of a 57% worsening of their acne and went to the treatment arm. Overall, there was a 95% improvement in the isotretinoin-treated group at the end of treatment (J Am Acad Dermatol. 1982 Apr;6[4 Pt 2 Suppl]:735-45). This study truly reflects how effective isotretinoin is in the treatment of severe acne, and any dermatologist who uses isotretinoin knows that for a fact. I suspect that, if someone applied the Cochrane Review process to insulin, they also would question its efficacy.

I agree there are questionable data on high versus standard versus low dosing. For most patients, standard dosing of 1.0 mg/kg per day over 20-24 weeks will result in complete and sustained clearance. There are always outliers, however, and alternative dosing may be needed – either higher cumulative dosing to achieve clearance or longer lower dosing to manage side effects.

The authors clearly have limited experience in treating acne, or the aim of the article was simply a review of available data with no consideration of actual clinical usage. However, if patients see this report, it could undermine dermatologists’ efforts to care for their patients with acne. It is rather irresponsible to publish these types of reviews, in which the efficacy of the medication in question was established long ago and is clearly without question, and it really discredits the integrity of the Cochrane Review.

Dr. Andrea L. Zaenglein

Dr. Andrea L. Zaenglein

This report will endure as an example of where bad data can result in bad conclusions.

Andrea L. Zaenglein, MD , is professor of dermatology and pediatric dermatology at Pennsylvania State University, Hershey, Penn., and is a cochair of the American Academy of Dermatology’s most recent acne management guidelines ( J Am Acad Dermatol. 2016 May;74[5]:945-73.e33 ). She reports being a consultant for Sun Pharmaceutical Industries.



There is limited evidence indicating that oral isotretinoin is effective in treating people with acne, compared with oral antibiotics or topical treatments, when based on reductions in inflammatory lesion counts, according to a recent Cochrane Review.

“The current recommendation of clinical guidelines that oral isotretinoin should be the first-line treatment for moderate to severe acne unresponsive to previous therapies, being more effective than the use of oral antibiotics plus topical agents, underpins current dermatological practice,” Caroline S. Costa, MD, from the Emergency Medicine and Evidence-Based Medicine department at Universidade Federal de São Paulo and her colleagues wrote in their review. “The recommendation cannot be fully supported with certainty by the finding of this review; however, neither does this review challenge it.”

Dr. Costa and her colleagues identified 31 randomized, controlled trials in the Cochrane Skin Group Specialised Register, CENTRAL, MEDLINE, Embase, PsycINFO, and LILACS databases up to July 2017, which altogether included 3,836 participants with mild to severe acne who were aged 12-55 years. The patients received treatment with oral isotretinoin versus placebo or other treatments, such as antibiotics.

In 3 studies that altogether included 400 participants and compared isotretinoin with oral antibiotics plus topical treatments, there was no significant difference with isotretinoin in decreases in the investigator-assessed inflammatory lesion count in participants with moderate to severe acne after 20-24 weeks (risk ratio, 1.01; 95% confidence interval, 0.96-1.06). One serious side effect – Stevens-Johnson syndrome – was noted in the isotretinoin-treated group (RR, 3.00; 95% CI, 0.12-72.98). These results were based on “very-low-quality evidence,” the authors noted.

Two studies also comparing isotretinoin with oral antibiotics and topical treatments in a total of 351 participants showed a 15% improvement in acne severity (RR, 1.15; 95% CI, 1.00-1.32) as assessed by physician’s global evaluation but with a greater number of adverse events considered less severe (RR, 1.67; 95% CI, 1.42-1.98), such as vomiting, nausea, dry lips and skin, and cheilitis.

With regard to dosing, one study with 154 participants with severe acne showed 0.05 mg/kg per day of isotretinoin reduced inflammatory lesion count by 79% after 20 weeks, while 0.1 mg/kg per day and 0.2 mg/kg per day reduced inflammatory lesion counts by 80% and 84%, respectively, after 20 weeks. A different study of 150 participants with severe acne found a 95% decrease in inflammatory lesion counts for 58% of participants receiving 0.1 mg/kg per day, 80% of participants receiving 0.5 mg/kg per day, and 90% of participants receiving 1 mg/kg per day of oral isotretinoin after 20 weeks.

In a randomized, controlled trial with 40 participants with moderate acne comparing intermittent dosing (0.5-0.7 mg/kg per day for 1 week per month), continuous conventional dosing (0.5-0.7 mg/kg per day), and continuous low dosing (0.25-0.4 mg/kg per day) of oral isotretinoin found greater differences in mean decreased numbers of inflammatory lesions among participants in the continuous low dose (a decrease of 3.72 lesions; 95% CI, 2.13-5.31) and continuous conventional dose (a decrease of 3.87 lesions; 95% CI, 2.31-5.43) groups. The authors noted they were unable to perform a meta-analysis because of study heterogeneity in the three studies examining the primary outcome.

In 14 studies with a total of 906 participants who had moderate and severe acne, the authors noted there were no severe adverse events with different doses and regimens of isotretinoin treatment from 12 weeks to 32 weeks or at follow-up at the end of treatment at up to 48 weeks. There were some adverse events, such as skin dryness, hair loss, and itching in 13 studies with a total of 858 participants.

No birth defects were reported in the studies.


Next Article: