The purpose of the present study was to assess the cutaneous response to intradermally administered vancomycin in healthy adults and to determine whether the magnitude of the cutaneous response correlated to the severity of "red man syndrome" (RMS) following intravenous administration of vancomycin to the same subjects. Eleven healthy males were skin tested with intradermally administered histamine and saline controls and intradermally administered vancomycin at different concentrations. Vancomycin caused a dose-dependent area of flare in all subjects. The sigmoidal maximal flare model was used to fit each dose-response curve, and cutaneous responsiveness to vancomycin was quantified by various methods, including the flare area at each dose, maximum flare area (maximal flare), dose required to produce 50% of maximum flare, dose required to produce a flare area of 400 mm2, and the slope of the dose-response curve. One week after skin testing, subjects received an infusion of vancomycin, 15 mg/kg of body weight over 60 min. For the assessment of the severity of RMS, we used previously described methods. Although all subjects experienced erythema from the intravenously administered vancomycin and 10 subjects had pruritus, there was no significant correlation between vancomycin skin test results and the severity of RMS. We conclude that vancomycin skin tests do not predict the severity of RMS. In addition, vancomycin skin tests may be of no benefit for assessing immunoglobulin E-mediated allergy to vancomycin, since all subjects had a positive reaction at concentrations of > or = 10 micrograms/ml.