To determine the early diagnostic parameters of the left ventricular diastolic filling defect in the uncontrolled adult onset diabetes mellitus, 86 diabetics were evaluated from the left ventricular inflow velocity pattern using pulsed Doppler echocardiography compared with normal 21 subjects. The diabetics were divided into 3 groups according to the presence or absence of background diabetic retinopathy(RE or coexisting cardiovascular diseases(group I : 34 cases without RE, group II : 24 cases with RE, group III : 28 cases with CVD). RE was thought to be having microangiopathy, but the cases with persistant massive proteinuria were excluded in this study. The left ventricular inflow velocity patterns were recorded from the apical approach. Peak velocity of the rapid filling phase(PFVE), that in the atrial systole(PFVA), E/A ratio, acceleration time(AT), deceleration time(DT), acceleration rate(ATR) and deceleration rate(DTR) were measured in the left ventricular inflow patterns. The results were ; 1) PFVE, AT and ATR in group I(56.26+/-12.21mm/sec, 70.91+/-14.98msec, 858.5+/-247mm/sec2), group II(51.91+/-14.35mm/sec, 64.84+/-14.98msec, 855.7+/-248.5mm/sec2) and group III(50.07+/-12.45mm/sec, 67.59+/-17.46msec, 817.5+/-266.8mm/sec2) were not significantly changed(p>0.05) compared to the controls(50.24+/-8.24mm/sec, 66.19+/-10.98msec, 784.5+/-221mm/sec2). 2) PFVA and DT in group I(51.21+/-14.86mm/sec, 156.2+/-23.42msec) and group II(64.26+/-13.93mm/sec, 64.84+/-14.13msec) were significantly increased(p<0.05, p<0.0001) compared to the controls(43.43+/-9.1mm/sec, 140.05+/-25.22msec). But those were not significantly different between group II and group III(p>0.05). 3) E/A ratio was significantly decreased in group II(0.82+/-0.21) and III(0.75+/-0.23) compared to the controls(1.19+/-0.25, p<0.0001) or group I(1.16+/-0.41, p<0.0001). But not significantly different between group II and group III(p>0.05) was noted. 4) DTR was also decreased in group II(311.9+/-95mm/sec2, p<0.05) and group III(297.7+/-125.8mm/sec2, p<0.05) compared to the controls(370.2+/-88mm/sec2) and group I(379.8+/-126mm/sec2). In conclusion, left ventricular diastolic filling defect in adult onset diabetics could be determined by using a various parameters of the pulsed Doppler echocardiography, which were closely related with diabetic retinopathy(RE). And DT and PFVA could be used as good parameters for early determination of the left ventricular diastolic filling defect in diabetics even without microangiopathy.