Abstract

A three phase study was designed to define further the sensitivity and specificity of symptomatic salicylate ototoxicity (primarily tinnitus) for serum salicylate concentrations. In phase one 260 patients with osteoarthritis and 112 with rheumatoid arthritis, none taking salicylates, were interviewed about their ear symptoms. Their responses were not significantly different from those of 134 salicylate treated patients with rheumatoid arthritis previously reported. In the second phase 56 patients who were taking salicylates, and who volunteered the complaint of tinnitus, had serum salicylate concentrations measured while symptomatic, and 30 (54%) had concentrations less than 1.3 mmol/l. Few tolerated an upward salicylate dose adjustment. For phase three, 94 patients were found to have a salicylate concentration above 2.2 mmol/l on one or more occasion, and these subjects were interviewed. Fifty two patients (55%) had no tinnitus, and tinnitus correlated with the blood salicylate concentration in only 28 (30%). Audiological evaluation of most of the symptomatic patients was carried out, and results were abnormal in the majority, even in those patients not reporting tinnitus. Symptomatic salicylate ototoxicity is too nonspecific and too insensitive to be a useful indicator of serum salicylate concentration.