How effective and safe is isotretinoin, taken in a tablet for people with acne? We reviewed the evidence about the effect of isotretinoin when compared either to itself at a different dose, to placebo (an identical but inactive treatment), or to other systemic (oral or injected medicines that work throughout the entire body) or topical (applied to the outside of the body) therapies. Eligible participants had to have been diagnosed with acne by a doctor.
Acne is a persistent inflammatory disease that can affect more than 80% of teenagers. Acne (including blackheads, whiteheads, and pimples) mostly appears on the face, but can also appear on the back and chest. Mental health problems, depression, and suicidal thoughts have been associated with acne. Isotretinoin, a currently widely used therapy derived from vitamin A, transformed acne treatment. However, it may cause adverse effects and has been associated with still uncertain psychiatric events and inflammatory bowel disease.
We searched the medical literature up to July 2017 and included 31 studies, involving 3836 dermatology outpatients worldwide. There were twice as many males than females; their ages ranged from 12 to 55 years old. Acne severity ranged from mild to severe, although most participants had severe acne.
The pharmaceutical industry funded 12 included studies.
We found studies that compared oral isotretinoin versus placebo or other treatments such as antibiotics. In addition, different doses, regimens (course of medical treatment), or formulations of oral isotretinoin were assessed, as well as oral isotretinoin with the addition of topical agents.
Three studies compared oral isotretinoin versus any oral antibiotic plus any topical agent given to participants with moderate or severe acne for between 20 to 24 weeks. Their outcomes were measured straight after treatment stopped.
There was no difference between therapies in decreasing the number of inflamed lesions (an area of an organ or tissue that has been damaged by disease or trauma). In one participant, isotretinoin led to the development of Stevens-Johnson syndrome (a serious disease where skin reacts severely, often in response to medication); there were no serious side effects in the other group. However, we are uncertain of these results because they were based on very low-quality evidence.
When assessed by a doctor, the severity of acne may be slightly improved by isotretinoin, but it may cause more side effects such as inflamed lips, dry skin, or nausea (low-quality evidence).
Fourteen studies compared different doses/courses of oral isotretinoin between 12 to 32 weeks. Participants had mainly severe or moderate acne.
Two studies, each comparing three different doses of isotretinoin at 20 weeks, found a greater improvement (measured by inflammatory lesion counts) with the higher dose (low-quality evidence). A third study showed that continuous (daily) low dose and continuous (daily) conventional dose may improve acne more than intermittent therapy, measured at 24 weeks (low-quality evidence). Conventional dose isotretinoin reduced inflammatory lesion counts more than low dose, but this was based on very low-quality evidence, indicating uncertainty.
During treatment (from 12 to 32 weeks) or follow-up after end of treatment (up to 48 weeks), no serious side effects occurred in 14 studies analysing different doses of isotretinoin (low-quality evidence). Doctor-measured severity of acne was not assessed in this comparison. Less serious side effects, including skin dryness, hair loss, and itching, were assessed in 13 studies, but we are uncertain if there were any differences between groups (low- to very low-quality evidence, where assessed).
No study reported birth defects.
Quality of the evidence
The overall quality of evidence for all of our key outcomes was low, due to serious limitations of study design and the limited amount of data. Thus, the identified clinical trials neither support nor challenge the established place of oral isotretinoin in acne treatment.
Evidence was low-quality for most assessed outcomes.
We did not find any clear evidence from RCTs that isotretinoin improves acne severity compared with standard oral antibiotic and topical treatment when assessed by a decrease in total inflammatory lesion count, but it may slightly improve physician-assessed acne severity. Only one serious adverse event was reported in the isotretinoin group, which means we are uncertain of the risk of serious adverse effects; however, isotretinoin may result in increased minor adverse effects.
Heterogeneity in the studies comparing different regimens, doses, or formulations of oral isotretinoin meant we were unable to undertake meta-analysis. Daily treatment may be more effective than treatment for one week each month. None of the randomised studies in this comparison reported serious adverse effects, or measured improvement in acne severity assessed by physician's global evaluation. We are uncertain if there is a difference in number of minor adverse effects, such as skin dryness, between doses/regimens.
Evidence quality was lessened due to imprecision and attrition bias. Further studies should ensure clearly reported long- and short-term standardised assessment of improvement in total inflammatory lesion counts, participant-reported outcomes, and safety. Oral isotretinoin is a well-established treatment for severe acne, and for acne that has not responded to oral antibiotics plus topical agents. The clinical trial evidence for oral isotretinoin conducted around 30 years ago was low quality. Further trials are needed to evaluate different dose/regimens of oral isotretinoin in acne of all severities.
Acne vulgaris, a chronic inflammatory disease of the pilosebaceous unit associated with socialisation and mental health problems, may affect more than 80% of teenagers. Isotretinoin is widely recognised as a very effective treatment for severe acne; however, it may cause adverse effects.
To assess efficacy and safety of oral isotretinoin for acne vulgaris.
We searched the following databases up to July 2017: the Cochrane Skin Group Specialised Register, CENTRAL, MEDLINE, Embase, PsycINFO and LILACS. We updated this search in March 2018, but these results have not yet been incorporated in the review. We also searched five trial registries, checked the reference lists of retrieved studies for further references to relevant trials, and handsearched dermatology conference proceedings. A separate search for adverse effects of oral isotretinoin was undertaken in MEDLINE and Embase up to September 2013.
Randomised clinical trials (RCTs) of oral isotretinoin in participants with clinically diagnosed acne compared against placebo, any other systemic or topical active therapy, and itself in different formulation, doses, regimens, or course duration.
We used standard methodological procedures expected by Cochrane.
We included 31 RCTs, involving 3836 participants (12 to 55 years) with mild to severe acne. There were twice as many male participants as females.
Most studies were undertaken in Asia, Europe, and North America. Outcomes were generally measured between eight to 32 weeks (mean 19.7) of therapy.
Assessed comparisons included oral isotretinoin versus placebo or other treatments such as antibiotics. In addition, different doses, regimens, or formulations of oral isotretinoin were assessed, as well as oral isotretinoin with the addition of topical agents.
Pharmaceutical companies funded 12 included trials. All, except three studies, had high risk of bias in at least one domain. Attrition bias was high in 20 trials, selective reporting bias was high in 12 trials, and performance bias was high in 11 trials.
Oral isotretinoin compared with oral antibiotics plus topical agents
These studies included participants with moderate or severe acne and assessed outcomes immediately after 20 to 24 weeks of treatment (short-term). Three studies (400 participants) showed no evidence that isotretinoin decreases trial investigator-assessed inflammatory lesion count more than antibiotics (RR 1.01 95% CI 0.96 to 1.06), with only one serious adverse effect found, which was Stevens-Johnson syndrome in the isotretinoin group (RR 3.00, 95% CI 0.12 to 72.98). However, we are uncertain about these results as they were based on very low-quality evidence.
Isotretinoin may slightly improve (by 15%) acne severity, assessed by physician's global evaluation (RR 1.15, 95% CI 1.00 to 1.32; 351 participants; 2 studies), but resulted in more less serious adverse effects (67% higher risk) (RR 1.67, 95% CI 1.42 to 1.98; 351 participants; 2 studies), such as dry lips/skin, cheilitis, vomiting, nausea (both outcomes, low-quality evidence).
Different doses/therapeutic regimens of oral isotretinoin
For our primary efficacy outcome, we found three RCTs, but heterogeneity precluded meta-analysis. One study (154 participants) reported 79%, 80% and 84% decrease in total inflammatory lesion count after 20 weeks of 0.05, 0.1, or 0.2 mg/kg/d of oral isotretinoin for severe acne (low-quality evidence). Another trial (150 participants, severe acne) compared 0.1, 0.5, and 1 mg/kg/d oral isotretinoin for 20 weeks and, respectively, 58%, 80% and 90% of participants achieved 95% decrease in total inflammatory lesion count. One 24-week RCT of participants with moderate acne compared isotretinoin at (a) continuous low dose (0.25 to 0.4 mg/kg/day), (b) continuous conventional dose (0.5 to 0.7 mg/kg/day), and (c) intermittent regimen (0.5 to 0.7 mg/kg/day, for one week in a month). Continuous low dose (MD 3.72 lesions; 95% CI 2.13 to 5.31; 40 participants; one study) and conventional dose (MD 3.87 lesions; 95% CI 2.31 to 5.43; 40 participants; one study) had a greater decrease in inflammatory lesion counts compared to intermittent treatment (all outcomes, low-quality evidence).
Fourteen RCTs (906 participants, severe and moderate acne) reported that no serious adverse events were observed when comparing different doses/therapeutic regimens of oral isotretinoin during treatment (from 12 to 32 weeks) or follow-up after end of treatment (up to 48 weeks). Thirteen RCTs (858 participants) analysed frequency of less serious adverse effects, which included skin dryness, hair loss, and itching, but heterogeneity regarding the assessment of the outcome precluded data pooling; hence, there is uncertainty about the results (low- to very low-quality evidence, where assessed).
Improvement in acne severity, assessed by physician's global evaluation, was not measured for this comparison.
None of the included RCTs reported birth defects, but oral isotretinoin is contraindicated during pregnancy due to known teratogenic effects.