Combining CT measurements assists pulmonary hypertension prediction

Mar 13 2015

By Eleanor McDermid, Senior medwireNews Reporter

Research suggests that combining pulmonary artery (PA) and ventricular measurements improves the chances of identifying pulmonary hypertension (PH) in patients undergoing computed tomography pulmonary angiography (CTPA).

Onno Spruijt (VU University Medical Center, Amsterdam, the Netherlands) and study co-authors stress that CTPA is not suitable as a primary screening tool for PH.

Nonetheless, “CTPA is often performed early in the diagnostic process of patients with unexplained dyspnea”, they write in the International Journal of Cardiovascular Imaging.

“Combining ventricular and PA measurements decreases the chance that the diagnosis of precapillary PH is missed.”

The researchers studied 51 precapillary PH patients who underwent CTPA and right heart catheterisation at baseline assessment. They also included 25 control patients who underwent both tests but had no evidence of PH.

A standard CTPA model, using the ratio of the PA diameter to the ascending aorta diameter (PA/AA), had good accuracy for distinguishing between patients with and without PH, at an area under the receiver operating characteristic curve of 0.901 (90.1% accuracy).

However, adding in the ratio of the maximum transverse diameter of the right ventricle to that of the left ventricle (RV/LV) improved the accuracy further, to more than 98%.

The researchers assessed RV/LV as measured on an axial view and as measured on a manually reconstructed four-chamber view. They found similar benefits with both methods and no statistically significant differences between the two.

“Therefore, determination of the RV/LV ratio in the axial view seems preferable as it does not require a manual reconstruction of the image”, they say.

In decision curve analyses, the addition of the RV/LV ratio (axial view) significantly decreased the number of false–positive cases without increasing the number of false–negative cases, by between nine and 24 per 100 patients, depending on the baseline probability of PH.

The researchers used low baseline probabilities, ranging from 1% to 20%, and gave more weight to false negatives than false positives, arguing that, for PH, “missing the diagnosis is worse than performing unnecessary diagnostic tests”. This strengthens the clinical relevance of adding RV/LV to arterial measurements, they say.

“When there is suspicion of precapillary PH, and a CTPA is made, we recommend radiologists to assess not only the diameters of the great vessels, but also of both ventricles”, concludes the team.

Licensed from medwireNews with permission from Springer Healthcare Ltd. ©Springer Healthcare Ltd. All rights reserved. Neither of these parties endorse or recommend any commercial products, services, or equipment.