A New Comment Pushes Back on the Reassuring Isotretinoin Height Study

A new comment published in the Journal of the American Academy of Dermatology takes aim at a Mayo Clinic-led study that found isotretinoin, the acne drug long flagged for stunting growth in teenagers, does not measurably shorten how tall patients end up. The original paper, led by Hyun-Min Seo and colleagues and built on the Rochester Epidemiology Project, tracked 226 adolescents who started isotretinoin before age 15 against 1,179 peers treated with oral antibiotics. At 18, the isotretinoin group stood just 0.67 cm shorter on average, a gap the study's own confidence interval (-2.21 to 0.87 cm) couldn't distinguish from noise.

The wrinkle sits in a different number. Height velocity, how fast these teenagers were growing month to month, dropped by 0.12 cm per month after starting the drug (P=.005), and the slowdown compared to their pre-treatment pace was sharper still: -0.31 cm per month (P=.011). Final height came out fine; the growth curve along the way did not, and that gap between a reassuring endpoint and a worrying trend line is what a comment on the paper is now pressing on.

The concern predates this study by six decades. Pease first described hypervitaminosis A causing premature closure of growth plates in 1962, and isotretinoin, a vitamin-A derivative, has since accumulated its own case reports of epiphyseal arrest, skeletal hyperostosis, and calcified ligaments, mostly in patients on the much higher, longer courses used for neuroblastoma rather than acne. The FDA has logged 41 worldwide reports of premature epiphyseal closure in isotretinoin patients under 18, at doses ranging from 0.5 mg/kg/day sustained over months to 3.5 mg/kg/day over years, the toxicology backdrop that has kept dermatologists cautious about prescribing the drug to kids who haven't finished growing.

A single-institution retrospective cohort of 226 treated patients, drawn overwhelmingly from a non-Hispanic white population in one health system, has limited statistical room to catch a modest but real height deficit, and a snapshot taken at age 18 misses any patient who hadn't yet reached final height. Neither the original study nor a parallel analysis of the TriNetX database that reached a similar conclusion collected skeletal or bone-age data, even though epiphyseal closure, not centimeters measured at 18, is the actual mechanism the six decades of case reports describe.