Kaplan R J, Daman L, Rosenberg E W, Feigenbaum S
Arch Dermatol. 1978 Jan;114(1):60-2.
In a double-blind study, topically applied caffeine 30%-hydrocortisone 0.5% in hydrophilic ointment was compared to betamethasone valerate 0.1% cream and to hydrocortisone 0.5% in hydrophilic ointment. Eighty-three patients were evaluated over a three-week period for pruritus, erythema, scaling, lichenification, excoriation, oozing, and global impression. The betamethasone and caffeine-hydrocortisone groups performed significantly better than the hydrocortisone group on three of the seven scales: lichenification, excoriation, and global impression. Also, the betamethasone group differed significantly from the hydrocortisone group on six of the seven scales, but did not differ significantly from the caffeine-hydrocortisone group on any scale. It is suggested that caffeine is effective because it elevates local levels of cyclic adenosine-3',5'-monophosphate by inhibiting phosphodiesterase.