The formulas currently utilized for noninvasive evaluation of right ventricular systolic pressure (RVSP) include right ventricular-right atrial pressure gradient (RV-RAG) and right atrial pressure (RAP). The former is expressed by trans-tricuspid systolic flow velocity, the latter is generally assumed. We recently observed that ultrasound estimation of RAP through inferior vena cava collapsibility index (CI) may help in the choice of the more appropriate formula for the evaluation of RVSP. However, these traditional methods (method A:RV-RAG + 10; method B:RV-RAG x 1.1 + 14) have limitations, particularly when RAP is low. The present study was undertaken to improve noninvasive estimation of RVSP through new formulas based on CI prediction of RAP. One hundred and four patients, in whom tricuspid regurgitation was adequately documented with CW-Doppler, were included in this study. They were classified into 3 groups: Group 1 with CI > 45%, Group 2 with CI < or = 35%, Group 3 with CI 35-45%. RVSP was evaluated by 3 different methods: A, B, and C. Method C was based on CI, assigning 6, 16, or 9 mmHg to RAP (respectively, the mean values in the 3 groups of our previous study). Results indicate that method C improves noninvasive estimation of RVSP in Group 1 and Group 2, with respect to other methods, with reduction of the SEE and of the mean difference of the t-test between hemodynamic and echographic values. In Group 3, Doppler estimation by method A and C, and catheter measurements are comparable, whereas method B significantly overestimates the actual value.(ABSTRACT TRUNCATED AT 250 WORDS)