Background: Allergy to mercury as a cause of oral lichenoid lesions (OLL) remains controversial. Some authors reported high frequency of sensitization to mercury and beneficial effect from removal of amalgam fillings in such patients, whereas others state that this procedure affects favorably all OLL, whether patients are sensitized to inorganic mercury or not. Objective: Our purpose was to determine the frequency of sensitization to metal salts in 194 patients (patients with OLL partly adjacent to amalgam fillings: 19, oral lichen planus (OLP) without close contact to amalgam: 42, other oral diseases: 28, oral complaints: 46, control group: 59). We further studied the histologic changes of biopsy specimens from positive patch tests to metal salts, and investigated the effect of removal of amalgam in OLL, to clarify whether it is possible to identify patients who will benefit from this procedure. Methods: Patch testing was performed with the German standard series, a dental prosthesis series, and a metal salt series including gold, mercury, and palladium salts as well as other salts of metals used in dental restorations. Late readings (10 and 17 days after application of the patch tests) were performed in all patients. Results: Of 19 patients with OLL adjacent to amalgam fillings, 15 (78.9%) were sensitized to inorganic mercury (INM), significantly more than those with OLL not adjacent to amalgam, other oral diseases or complaints, and the control group. In 5 of 15 (33.3%) of the patients with OLL, a positive patch test to INM was observed only at D10 or D17. Amalgam was removed in 18 patients with OLL (sensitization to INM: 15), and in 11 patients with OLP (sensitization to INM: 2). After removal, the lesions of 13 of 15 of the INM-sensitized patients with OLL (86.7%) and 2 with OLP healed or improved significantly, but this was not observed with the INM negative patients. Frequency of sensitization to gold sodium thiosulfate (GST) and palladium chloride 1% pet (PDC) was high in all groups. This was partly because readings were performed late. Lesions of 2 patients with allergic contact stomatitis caused by gold and 1 caused by palladium healed completely after removal of these restorations. Histologically, lichenoid changes were observed in 14 of 36 biopsy specimens of positive patch tests from INM (9/21), GST (2/10), and PDC (3/5) in all patient groups, mainly in persistent patch tests at D10 or D17. This was not observed in 12 biopsy specimens taken from persistent patch tests from other substances, including nickel sulfate. Conclusion: Our results suggest that sensitization to mercury is an important cause of OLL, whether all lesions or only a part of them are adjacent to amalgam fillings. Sensitization to GST may reflect true gold allergy and should be considered as a cause of oral diseases in some patients. Sensitization to PDC is frequent but has yet only little clinical relevance. Patch tests may be positive only at D10 or D17. This suggests the importance of additional readings of GST, PDC, and mercury salts at this time. (J Am Acad Dermatol 1999;41:422-30.)