Accutane (Isotretinoin) Side Effects: Incidence Rates Across Clinical Trials

Print or save as PDFOpens the printable lab tracker and your print dialog. No sign-up.

Which side effects come with an incidence figure?

The Accutane label reports trial percentages for several blood-lipid changes. It states that marked elevations of serum triglycerides were reported in approximately 25% of patients receiving Accutane in clinical trials, that approximately 15% developed a decrease in high-density lipoproteins, and that about 7% showed an increase in cholesterol levels [1]. The same section says these effects on triglycerides, HDL and cholesterol were reversible when the drug was stopped, and that triglyceride elevations above 800 mg/dL have been reported in treated patients [1].

The Absorica and Absorica LD label handles the common reactions differently. It lists its most common adverse reactions as those occurring at an incidence of 5% or greater: dry lips, dry skin, back pain, dry eye, arthralgia, epistaxis, headache, nasopharyngitis, chapped lips, dermatitis, increased creatine kinase, cheilitis, musculoskeletal discomfort, upper respiratory tract infection and reduced visual acuity [3]. That threshold belongs to the Absorica and Absorica LD products; the label does not give a separate percentage for each reaction in that group [3].

The older-format isotretinoin label reports that about 16% of patients in a clinical trial developed musculoskeletal symptoms, including arthralgia [1]. Both labels note that cheilitis and hypertriglyceridemia are dose related [3][1].

Laboratory changes: the best-quantified data

A retrospective cohort in a Northern California managed-care plan followed 13,772 patients aged 13 to 50 who took oral isotretinoin between March 1995 and September 2002 [2]. Among patients whose values were normal at baseline, the cumulative incidence of a new abnormality during treatment was 44% for triglycerides, 31% for total cholesterol and 11% for transaminases [2]. Moderate to severe lipid and transaminase abnormalities were generally transient and reversible, and new hematologic abnormalities were uncommon [2].

A 2016 meta-analysis in JAMA Dermatology pooled 26 studies and 1,574 patients and reported mean on-treatment values rather than rates of abnormality: triglycerides 119.98 mg/dL (99% CI 98.58 to 141.39), total cholesterol 184.74 mg/dL (99% CI 178.17 to 191.31), LDL cholesterol 109.23 mg/dL and HDL cholesterol 42.80 mg/dL [4]. Those are average levels across treated patients, not the share of patients who crossed a threshold.

Other study data on side-effect frequency

The 1984 multicenter dose-response study enrolled 150 patients with treatment-resistant nodulocystic acne into three equal groups at 0.1, 0.5 and 1.0 mg/kg/day, double-blind [5]. All three doses produced a highly significant clinical response with no significant difference in response between doses, and the investigators reported only minor differences in clinical side effects and laboratory abnormalities across the groups [5]. Retreatment was needed in 42% of the 0.1 mg/kg/day group [5].

A much smaller early study treated 14 patients with treatment-resistant cystic and conglobate acne for four months at an average of 2.0 mg/kg/day. Clinical toxicity was limited to the skin and mucous membranes in most patients, was dose dependent and reversed rapidly after the drug was stopped [6].

A 2026 single-centre randomized trial compared cumulative doses of 120 and 150 mg/kg. It reported that adverse events were mild and comparable between the two groups, but did not report rates for individual reactions; 12-month relapse was 26.7% at 120 mg/kg versus 32.3% at 150 mg/kg (p = 0.619) [7].

An international survey of the last 100 isotretinoin patients of each of 12 dermatologists (1,000 patients) found mucocutaneous side effects occurred frequently but were manageable, while severe systemic side effects were rarely problematic, at 2% [8].

Psychiatric events: the trial data

A 2017 systematic review and meta-analysis identified 31 studies of isotretinoin in acne [9]. In controlled studies, the change in depression scores from baseline did not differ significantly between patients on isotretinoin and those on an alternative treatment (standardized mean difference -0.334, 95% CI -0.680 to 0.011) [9]. The prevalence of depression after treatment declined (relative risk 0.588, 95% CI 0.382 to 0.904), and mean depression scores fell from baseline (standardized mean difference -0.335, 95% CI -0.498 to -0.172) [9]. A limitation the authors noted: no randomized controlled trials were included [9].

The Accutane label warns that the drug may cause depression, psychosis and, rarely, suicidal ideation, suicide attempts, suicide and aggressive or violent behaviors, and that patients should be assessed for these at each visit [1]. The Absorica and Absorica LD label carries the same warning [3]. Of the patients reporting depression, some reported that it subsided with discontinuation and recurred with reinstitution of treatment [3][1].

Inflammatory bowel disease

A population-based case-control study linking the University of Manitoba IBD database to the provincial prescription registry found that 1.2% of people with IBD had used isotretinoin before diagnosis, statistically similar to 1.1% of matched controls, with no difference between Crohn's disease and ulcerative colitis [10]. The authors concluded isotretinoin is not likely to cause chronic IBD, while noting anecdotes of acute colitis [10]. The Accutane label still lists inflammatory bowel disease among its adverse reactions and warnings [1], and the Absorica and Absorica LD label says to discontinue treatment for abdominal pain, rectal bleeding or severe diarrhea [3].

Why many reactions have no incidence figure

The Absorica and Absorica LD label says its listed reactions were identified in clinical trials or postmarketing reports, and that because some were reported voluntarily from a population of uncertain size, it is not always possible to reliably estimate their frequency or establish a causal relationship to drug exposure [3]. The FDA's adverse event dashboard makes a similar point about report data generally: the reports cannot be used to estimate incidence, and a report does not establish that the drug caused the event [11].

Birth defects: a boxed warning without a percentage

The Accutane boxed warning does not state a malformation rate. It says there is an extremely high risk that life-threatening birth defects will result if pregnancy occurs while taking any amount of Accutane, even for short periods, and that potentially any exposed fetus can be affected [1]. It lists documented malformations of the face, eyes, ears, skull, central nervous system, cardiovascular system, thymus and parathyroid glands, notes reported IQ scores below 85 with or without other abnormalities, and notes an increased risk of spontaneous abortion and premature births [1].

A 2001 study of exposure trends estimated that about 2.5 per 1,000 reproductive-aged women were exposed to Accutane in 1999 [12]. Of 14 women with recent exposed pregnancies who were interviewed, eight used no contraception at the time and 13 did not use two forms of contraception [12]. The iPLEDGE program has existed since 2005 and was approved as a REMS in 2010; patients must be enrolled to receive isotretinoin [13].