Background : Electrogastrography (EGG) enables the cutaneous measurement of gastric electric activity. Abnormal gastric slow wave frequencies have been observed in diabetic gastroparesis. The aim of this study was to know the clinical application of EGG, and to determine the relationship between gastric myoelectrical activity and gastric emptying time (GET) in patients with type 2 diabetes mellitus. Methods : This study included 26 type 2 diabetic patients (11 men, 15 women, mean age 53±10 years, disease duration 10.5±6.5 years). They were given solid and liquid diet with 1 mCi (99m)Tc DTPA for measuring GET (Tl/2). They were divided into normal GET group (Tl(2 < 60 minutes) and delayed GET group (Tl/2 > 60 minutes). Gastric myoelec-trical activity was recorded by cutaneous EGG. EGG was recorded for 30 minutes in fasting state and 30 minutes in post-prandial state with same test meal. Several EGG variables including percentages of dominant frequency in the normal range (2 cycle per minute, cpm), bradygastria (0.5~2 cpm), tachy- gastria (4~9cpm), postprandial to preprandial power ratio, and dominant frequency instability coefficient (DFIC) were calculated by fast Fourier transformation. Result: Autonomic nerve function test showed no difference between normal GET group and delayed GET group. EGG values in delayed group did not differ from data in normal group. There were no correlation between b1ood glucose level and GET. In fasting, normal slow wave (r=-0.50, p<0,05) and bradygastria (r=0.50, p<0.05) were correlated with fasting blood glucose. And in postprandial state, normal slow wave (r=0.47, p<0.05) and bradygastria (r=0.84, p<0.05) were correlated significantly with fasting blood glucose. EGG parameters did not correlated with GET Conclusion : EGG may be influenced by blood glucose level, but seems to be less valuable for assessment of gastroparesis in Korean type 2 diabetes mellitus, according to our investigation.