The aim of this study was to evaluate the clinical relevance of amylase level monitoring as an objective measure in diagnosis and assessment of treatment response in bulimia nervosa. Thirty-three subjects who fulfilled DSM-111-R criteria for bulimia nervosa had serum levels of total and salivary amylase monitored during an 8-week treatment trial. At the beginning of treatment, the average total amylase level was within the upper limits of normal, whereas average salivary amylase levels were abnormally high. During the course of treatment, there was a significant reduction in the average salivary isoenzyme to within the normal range. Significant reductions in amylase levels were recorded in patients with good treatment outcome, but not in those with poor outcome. Amylase levels were not significantly correlated with severity of bulimic symptoms. These results do not justify the use of amylase assays as a routine diagnostic or monitoring test, but isoenzyme monitoring may provide useful clinical information in selected cases.