Comparison of intravenous digital subtraction cineangiocardiography with conventional contrast ventriculography for the determination of the left ventricular volume at rest and during exercise.

Left ventricular volumes were determined by means of digital subtraction cineangiocardiography (DSA) which was performed in the right anterior oblique projection after contrast agent injection into the superior vena cava. Monoplane end-diastolic (EDV), end-systolic volumes (ESV), and ejection fraction (EF) were calculated using the 'area-length' method and were compared with the same parameters obtained by conventional left ventricular cineangiocardiography. A first group of 20 patients was studied at rest and a second group of 10 patients during bicycle exercise at a work load of 64 watts during 2 min, by DSA and conventional cineangiocardiography. Three different subtraction modes were evaluated: (1) mask mode subtraction (MMS), (2) time interval difference (TID) method and (3) a combination of MMS and TID called MMS + TID method. With the MMS method good correlations were obtained for EDV, ESV and EF at rest (r greater than 0.91) and during exercise (r greater than 0.91). The TID method showed only moderate correlations for patients at rest (r greater than 0.86) and during exercise (r greater than 0.79). Similar results as with MMS were achieved by the combined method (MMS + TID) at rest (r greater than 0.91) and during exercise (r greater than 0.91). Interobserver variability indicated a high reproducibility for all methods except for TID during exercise. It is concluded that DSA is an accurate technique for left ventricular volume determination not only at rest but also during exercise. The best results are obtained with MMS or MMS + TID methods, while left ventricular contour detection is easier and more convenient with MMS + TID.