Acne: Study Finds No Increased IBD Risk With Isotretinoin or Tetracyclines

Christine Kilgore

February 06, 2023

Isotretinoin was not associated with a 1-year risk of incident inflammatory bowel disease (IBD) in a large population-based cohort study that also found no significant association of oral tetracycline-class antibiotics with IBD – and a small but statistically significant association of acne itself with the inflammatory disorders that make up IBD.

Dr John Barbieri

For the study, senior author John S. Barbieri, MD, MBA, of the department of dermatology, at Brigham and Women's Hospital, Boston, and his colleagues used data from the TriNetX global research platform, which mines patient-level electronic medical record data from dozens of health care organizations, mainly in the United States. The network includes over 106 million patients. They looked at four cohorts: Patients without acne; those with acne but no current or prior use of systemic medications; those with acne managed with isotretinoin (and no prior use of oral tetracycline-class antibiotics); and those with acne managed with oral tetracycline-class antibiotics (and no exposure to isotretinoin).

For the acne cohorts, the investigators captured first encounters with a diagnosis of acne and first prescriptions of interest. And studywide, they used propensity score matching to balance cohorts for age, sex, race, ethnicity, and combined oral contraceptive use.

"These data should provide more reassurance to patients and prescribers that isotretinoin does not appear to result in a meaningfully increased risk of inflammatory bowel disease," they wrote in the study, published online in the Journal of the American Academy of Dermatology.

"These are important findings as isotretinoin is a valuable treatment for acne that can result in a durable remission of disease activity, prevent acne scarring, and reduce our overreliance on oral antibiotics for acne," they added.

Dr Jonathan Weiss

Indeed, dermatologist Jonathan S. Weiss, MD, who was not involved in the research and was asked to comment on the study, said that the findings "are reassuring given the large numbers of patients evaluated and treated." The smallest cohort – the isotretinoin group – had over 11,000 patients, and the other cohorts had over 100,000 patients each, he said in an interview.

"At this point, I'm not sure we need any other immediate information to feel comfortable using isotretinoin with respect to a potential to cause IBD, but it would be nice to see some longitudinal follow-up data for longer-term reassurance," added Dr. Weiss, who practices in Snellville, Georgia, and is on the board of the directors of the American Acne and Rosacea Society.

The findings: Risk with acne

To assess the potential association between acne and IBD, the researchers identified more than 350,000 patients with acne managed without systemic medications, and propensity score matched them with patients who did not have acne. Altogether, their mean age was 22; 32.1% were male, and 59.6% were White.

Compared with the controls who did not have acne, they found a statistically significant association between acne and risk of incident IBD (odds ratio, 1.42; 95% confidence interval, 1.23-1.65) and an absolute risk difference of .04%. Separated into Crohn's disease (CD) and ulcerative colitis (UC), ORs were 1.56 and 1.62, respectively.


To assess the association of oral tetracycline use and IBD, they compared more than 144,000 patients whose acne was managed with antibiotics with patients whose acne was managed without systemic medications. The patients had a mean age of 24.4; 34.7% were male, and 68.2% were White.

Compared with the patients who were not on systemic medications, there were no significant associations among those on oral tetracyclines, with an OR for incident IBD of 1 (95% CI, 0.82-1.22), an OR for incident CD of 1.09 (95% CI, 0.86-1.38), and an OR for UC of 0.78 (95% CI, 0.61-1.00).


To evaluate the association of isotretinoin and IBD, the researchers compared more than 11,000 patients treated with isotretinoin with two matched groups: patients with acne managed without systemic medications, and patients with acne managed with oral tetracyclines. The latter comparison was made to minimize potential confounding by acne severity. These patients had a mean age of 21.1; 49.5% were male, and 75.3% were White.

In the first comparison, compared with patients not treated with systemic medications, the OR for 1-year incidence of IBD among patients treated with isotretinoin was 1.29 (95% CI, 0.64-2.59), with an absolute risk difference of .036%. The ORs for CD and UC were 1.00 (95% CI, 0.45-2.23) and 1.27 (95% CI, .58-2.80), respectively.

And compared with the antibiotic-managed group, the OR for incident IBD among those on isotretinoin was 1.13 (95% CI, 0.57-2.21), with an absolute risk difference of .018%. The OR for CD was 1.00 (95% CI, 0.45-2.23). The OR for UC could not be accurately estimated because of an insufficient number of events in the tetracycline-treated group.

"Challenging" area of research

Researching acne treatments and the potential risk of IBD has been a methodologically "challenging topic to study" because of possible confounding and surveillance bias depending on study designs, Dr. Barbieri, director of the Brigham and Women's Advanced Acne Therapeutics Clinic, said in an interview.

Studies that have identified a potential association between isotretinoin and IBD often have not adequately controlled for prior antibiotic exposure, for instance. And other studies, including a retrospective cohort study also published recently in JAAD using the same TriNetX database, have found 6-month isotretinoin-related risks of IBD but no increased risk at 1 year or more of follow-up – a finding that suggests a role of surveillance bias, Dr. Barbieri said.


The follow-up period of 1 year in their new study was chosen to minimize the risk of such bias. "Since patients on isotretinoin are seen more often, and since there are historical concerns about isotretinoin and IBD, patients on isotretinoin may be more likely to be screened earlier and thus could be diagnosed sooner than those not on [the medication]," he said.

He and his coauthors considered similar potential bias in designing the no-acne cohort, choosing patients who had routine primary care visits without abnormal findings in order to "reduce potential for bias due to frequency of interaction with the health care system," they noted in their paper. (Patients had no prior encounters for acne and no history of acne treatments.)

Antibiotics, acne itself

Research on antibiotic use for acne and risk of IBD is scant, and the few studies that have been published show conflicting findings, Dr. Barbieri noted. In the meantime, studies and meta-analyses in the general medical literature – not involving acne – have identified an association between lifetime oral antibiotic exposure and IBD, he said.

While the results of the new study "are reassuring that oral tetracycline-class exposure for acne may not be associated with a significant absolute risk of inflammatory bowel disease, given the potential for antibiotic resistance and other antibiotic-associated complications, it remains important to be judicious" with their use in acne management, he and his coauthors wrote in the study.

The potential association between antibiotics for acne and IBD needs further study, preferably with longer follow-up duration, Dr. Barbieri said in the interview, but researchers are challenged by the lack of datasets with high-quality longitudinal data "beyond a few years of follow-up."

The extent to which acne itself is associated with IBD is another area ripe for more research. Thus far, it seems that IBD and acne – and other chronic inflammatory skin diseases such as psoriasis – involve similar pathogenic pathways. "We know that in IBD Th17 and TNF immunologic pathways are important, so it's not surprising that there may be associations," he said.

In their paper, Dr. Barbieri and his coauthors emphasize, however, that the absolute risk difference between acne and IBD is small. It's "unlikely that population level screening is warranted among patients with acne," they wrote.

A second new study

The other study, also published recently in JAAD, used the same TriNetX research platform to identify approximately 77,000 patients with acne starting isotretinoin and matched them with patients starting oral antibiotics.

The investigators, Khalaf Kridin MD, PhD, and Ralf J. Ludwig, MD, of the Lübeck Institute of Experimental Dermatology, University of Lübeck (Germany), found that the lifetime risks (greater than 6 months) for patients on isotretinoin were not significantly elevated, compared with those on oral antibiotics for either CD (hazard ratio 1.05; 95% CI, 0.89-1.24, P = .583) or UC (HR, 1.13; 95% CI, 0.95-1.34; P = .162) They also looked at the risk of irritable bowel syndrome (IBS) and found a lower lifetime risk in the isotretinoin group.

In the short term, during the first 6 months after drug initiation, there was a significant, but slight increase in UC in the isotretinoin group. But this risk decreased to the level of the antibiotic group with longer follow up. "The absolute incidence rates [of IBD] and the risk difference of UC within the first 6 months are of limited clinical significance," they wrote.

It may be, Dr. Weiss said in commenting on this study, "that isotretinoin unmasks an already-existing genetic tendency to UC early on in the course of treatment, but that it does not truly cause an increased incidence of any type of IBD."

Both studies, said Dr. Barbieri, "add to an extensive body of literature that supports that isotretinoin is not associated with IBD."

Dr. Barbieri had no disclosures for the study, for which Matthew T. Taylor served as first author. Coauthor Shawn Kwatra, MD, disclosed that he is an advisory board member/consultant for numerous pharmaceutical companies and has served as an investigator for several. Both are supported by the National Institute of Arthritis and Musculoskeletal and Skin Diseases. The other authors had no disclosures. Dr. Kridin and Dr. Ludwig had no disclosures for their study. Dr. Weiss had no disclosures.

This article originally appeared on, part of the Medscape Professional Network.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.