Tolerance profile of retinol, retinaldehyde and retinoic acid under maximized and long-term clinical conditions.

BACKGROUND: Topical retinoic acid (RA) causes irritation of the skin. To prevent this side effect, natural precursors of RA have been proposed. The aim of the present study was to compare the local tolerance profiles of retinol (ROL), retinaldehyde (RAL) and RA. METHODS: ROL, RAL and RA were studied using repeated insult patch tests for 14 days (n = 6). Similarly, RAL and RA were assessed in long-term clinical use for 44 weeks (n = 355). Clinical scoring on irritation, measurement of transepidermal water loss (barrier function) and laser Doppler blood flow perfusion units (irritation) were performed. RESULTS: Under maximized conditions, an equally low irritation potential for ROL and RAL and a more pronounced irritant effect with RA could be demonstrated clinically (p < 0.05 in the intergroup analysis). Furthermore, RAL and RA induced more scaling than ROL (p < 0.05), and ROL and RA tended to induce more burning/pruritus than RAL (nonsignificant). The TEWL values were low with ROL and high with RAL and RA (nonsignificant, intergroup analysis). The laser Doppler measurements confirmed pro-irritating effects of RA and the nonirritating effects of ROL and RAL (p = 0. 001, intergroup analysis). The long-term clinical study showed that the study population developed a high frequency of erythema (44% of the population), scaling (35%) and burning/pruritus (29%) with RA in the first 4 weeks of treatment, whereas these 3 parameters were significantly less frequent with RAL (p < 0.0001 in the intergroup analysis). CONCLUSION: The natural retinoids ROL and RAL do have a good tolerance profile, in contrast with the irritating potential of RA.