Leah Muhm Lin MD, Abstract Co-Author: Nothing to Disclose

Lauren L. Ihde MD, Abstract Co-Author: Nothing to Disclose

Steven Cen PhD, Abstract Co-Author: Nothing to Disclose

Alison Wilcox MD, Presenter: Speaker, Toshiba Corporation

Gerard K. Nguyen MD, Abstract Co-Author: Nothing to Disclose

Michelle Bussinguer, Abstract Co-Author: Nothing to Disclose

Christopher Lee MD, Abstract Co-Author: Nothing to Disclose

To establish a prediction model for pulmonary hypertension (PH) using CT measurements correlated to mean pulmonary artery pressure (mPAP).

189 patients who had received chest CT and right heart catheterization (RHC) within twelve months were evaluated. 153 patients had also received transthoracic echocardiogram. The following CT measurements were made: main pulmonary artery diameter (mPAD), main pulmonary artery area, main pulmonary artery to ascending aorta diameter (mPAD/AAD) ratio, left pulmonary artery (LPA) diameter, LPA area, right pulmonary artery (RPA) diameter, RPA area, and lower lobe segmental artery to bronchus ratios. Spearman correlations were utilized to test relationships between CT measurements and mPAP. Receiver operating characteristic (ROC) curve analysis was performed to assess whether a composite index of mPAD and mPAD/AAD ratio, as well as echocardiography-derived right ventricular systolic pressure (RVSP), could improve prediction of PH. Sensitivity and specificity of various cutoff points were calculated.

Among the various CT markers, mPAD and mPAD/AAD ratio had the highest correlation with mPAP (R2 = 0.63 and 0.64, respectively; p < .001). RSVP demonstrated even higher correlation (R2 = 0.69). mPAD cutoff of 32.1 mm was found to be 62.4% sensitive and 87.5% specific for PH. mPAD/AAD ratio cutoff of 0.93 was 81.6% sensitive and 73.4% specific for PH. The accuracies of mPAD, mPAD/AAD ratio, composite of mPAD and mPAD/AAD, and composite of mPAD, mPAD/AAD ratio, and RVSP in predicting PH, as calculated utilizing the areas under the ROC curves, were 0.83, 0.84, 0.86, and 0.90 respectively.

mPAD and mPAD/AAD ratio showed the highest correlation with mPAP. Area measurements, left and right pulmonary artery measurements, and lower lobe segmental level measurements demonstrated weaker correlations. In our patient population, a composite of mPAD and mPAD/AAD ratio was not associated with increased accuracy compared to mPAD/AAD ratio alone.

CT measurements of mPAD and mPAD/AAD ratio correlate well with pulmonary artery pressures. The specific threshold for mPAD will vary depending on the individual’s preference for sensitivity and specificity. A composite index of CT and echocardiography is the most accurate method to stratify patients for RHC.

Lin, L, Ihde, L, Cen, S, Wilcox, A, Nguyen, G, Bussinguer, M, Lee, C, Which Is the Most Accurate? Evaluation of CT Measurements in Predicting Pulmonary Hypertension.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14005804.html