Background: Embryofetal developmental toxicity associated with oral administration of vitamin A analogs has led to concern about risks from topical application. Objective: This study was conducted to evaluate the potential developmental toxicity of tretinoin emollient cream when applied to the skin of pregnant New Zealand white rabbits during organogenesis (gestational days 7 through 19). Methods: Twenty rabbits each were randomly assigned to a control group (group I) or to receive vehicle (group II) or tretinoin emollient cream topically at dosages of 10 (0.05 mg/kg*, group III) or 100 (0.5 mg/kg^{NO LABEL*The milligrams per kilogram dosage refers to the amount of active ingredient (tretinoin). The 0.05 mg/kg and 0.5 mg/kg groups were treated with 0.005% and 0.05% wt/wt tretinoin emollient cream formulation. The 0.05% tretinoin emollient cream is the Renova clinical formulation. The 10 and 100 times clinical multiples refer to Renova clinical multiples and are based on a 50 kg adult patient's applying 500 mg of 0.05% tretinoin emollient cream formulation daily to yield a clinical dosage of 0.005 mg/kg.} , group IV) times that used clinically in humans. Does and fetuses were examined for tretinoin-induced toxic effects, and maternal plasma tretinoin and metabolite levels were measured. Results: The rate of abortion was increased significantly in does in group IV ( p ≤ 0.01) compared with the control group. Dosage-dependent increases in incidence and severity of skin reactions occurred in groups administered the vehicle and the two dosages of tretinoin. Does in groups III and IV had clinical and necropsy observations that were considered direct or indirect effects of tretinoin administration, persistent weight loss, and reduced feed consumption. Maternal endogenous plasma tretinoin levels were below the lower limit of quantitation of 5 ng/ml and were not significantly altered with treatment. Group IV had significantly reduced mean fetal body weight ( p ≤ 0.01) and a greater frequency of resorptions compared with group I. Although external, visceral, or skeletal alterations occurred at significantly greater levels in group III, they were unrelated to tretinoin administration because the fetal incidences were not dosage dependent, and the litter incidence did not significantly differ from the control group values. Conclusion: Maternally toxic dosages of tretinoin were associated with an increased incidence of abortions and resorptions and reduced fetal body weight, two end points of developmental toxicity. Consistent with the absence of detectable tretinoin plasma levels, however, no changes in fetal morphology were attributable to tretinoin administration. (J Am Acad Dermatol 1997;36:S67-S76.)