(patients with no observable tricuspid regurgitation), the strength of correlation of RVSP versus PVO% increased slightly (r 2 = 0.147, plot not shown). We then constructed a multivariate regression model to determine the significance of each predictor variable. The resultant equation was as follows: PVO% = 0.331242 ) 0.001554 SaO 2 % + 0.001882 HR + 0.002051 D-dimer ) 0.111275 Troponin + 0.037913 log(BNP) + 0.004763 RVSP.When we performed a t-test on each coefficient, only the RVSP coefficient was significant (P = 0.0003).We also performed follow-up at 6 months in all survivors (n = 182) and 15 patients reported recurrent deep venous thrombosis (DVT) and three other patients reported recurrent PE (including one with PE and DVT). The three patients with recurrent PE had percentage occlusion values of 95%, 80% and 5%.Our data suggest that the PVO%, a ready index of clot size, correlated poorly with five of six independent variables tested. We found the virtual absence of a correlation between heart rate, SaO 2 %, troponin I and BNP vs. PVO%. We found that RVSP had the tightest correlation with PVO%, although the correlation we observed between RVSP from echocardiography vs. PVO% from CTA was much weaker than correlations previously reported between RVSP and pulmonary vascular resistance, when both parameters were measured with formal pulmonary angiography [6]. PE renders a complex set of stresses to the human body, including global circulatory insufficiency, shear stress and ischemia in the right ventricular muscle, abnormalities in gas exchange, and cardiopulmonary inflammation. Each of the six independent variables clearly measure important host responses to the stress of PE: heart rate reflects baroreceptor unloading; SaO 2 reflects ventilationperfusion mismatch; D-dimer reflects fibrin mass and fibrinolytic activity; troponin and BNP reflect cardiomyocyte damage; and RVSP reflects impedence to RV systolic ejection. Because only three out of 182 survivors reported recurrent PE, we cannot make any inference as to whether clot size predicts PE recurrence. In summary, we believe these data underscore the notion that clot size should not be viewed as the predictor-inchief in assessing PE severity.
Disclosure of Conflict of InterestsThe authors state that they have no conflict of interest.