When 0.75% metronidazole gel was administered to rosacea patients during the clinical trials only infrequent, irritant-type local adverse events were attributable to the drug. No correlations were made between these events and patient-demographic, drug-disease, or drug-drug interactions. No systemic adverse events or abnormal clinical laboratory data were attributable to metronidazole gel. Since topically applied 0.75% metronidazole gel was shown to be minimally absorbed, systemic events are minimised. The adverse event occurrence rates were of 1.8% for irritation, dryness or redness effect and of 0.9% for watery (tearing) eyes or burning effect. All individual events occurred in less than 2% of patients. Additionally, results from patch testing in humans showed no evidence of irritancy, sensitisation, phototoxicity or photoallergic reactions when topical metronidazole was applied to non-facial areas of normal volunteers.