The reproducibility and agreement of the estimates of stroke volume (SV), cardiac output (CO) and total peripheral resistance (TPR) by transthoracic impedance cardiography (ZCG) and M-mode echocardiography (ECHO) were analyzed before and after the placebo-controlled administration of ascending doses of isosorbide dinitrate and nicorandil in 12 healthy subjects. There was no biostatistical agreement between the two methods in estimating cardiovascular function either before or after dosing (ZCG estimated substantially larger SV, CO and lower TPR). But, ZCG and ECHO estimated about similar overall treatment related changes (across treatments and periods) and reached substantially better agreement when the values were expressed as ratio of the baseline before dosing. Such improvement did not occur when the data were expressed as arithmetic difference from baseline. In spite of the improvement of agreement by expressing the data as ratio of baseline, the coefficients of reproducibility between the two methods (circa 25% of baseline) remained too large to judge the methods interchangeable.