The aim of the present work was to study the reliability of conductance microcatheter volumetric measurements as compared to magnetic resonance imaging (MRI) in the same set of mice. Mice left ventricular (LV) volumes were monitored under basal conditions and in a hypertrophy model induced by transverse aortic constriction (TAC). Cardiac function was evaluated in isoflurane anesthetized mice (n = 8) by MRI followed by 1.4 F Millar microtip catheter measurements. The second group of mice with TAC-induced cardiac hypertrophy was studied eight weeks after surgery. Reliability of 3D-reconstructed MRI data was confirmed by comparison with autopsy masses (autopsy LV mass = 73.6 +/- 3.4 mg; MRI LV mass = 76.9 +/- 3.7 mg). Conduction catheter was found to greatly underestimate end-diastolic and end-systolic volumes and thus stroke volume as well as cardiac output in control mice (MRI: EDV = 79 +/- 8 microl, ESV = 27+/-9 microl, SV = 51 +/- 9 microl, CO = 25 +/- 6 ml/min; Catheter: EDV = 28 +/- 5 microl, ESV = 8 +/- 4 microl, SV = 19 +/- 4 microl, CO = 10 +/- 2 ml/min). However, values for ejection fraction showed no significant differences between the two methods. In the hypertrophy model, stroke volume and cardiac output were increased when measured with MRI (SV: +19 +/- 20%; CO: +28 +/- 27%), whereas catheter data showed opposite directional changes (SV: -22 +/- 37%; CO: -31 +/- 37%). Ejection fraction was found to be reduced only in catheter measurements (-31 +/- 26%). In summary, our data demonstrate that absolute volumetric values are strikingly underestimated by conduction catheter measurements and that even detection of directional changes with this method may not always be feasible.