Kanako Kunishima Kumamaru MD, Abstract Co-Author: Nothing to Disclose

Nicole Wake MD, MS, Presenter: Nothing to Disclose

Andetta Rotilla Hunsaker MD, Abstract Co-Author: Nothing to Disclose

Michael Tse-Yin Lu MD, Abstract Co-Author: Nothing to Disclose

Jason Signorelli, Abstract Co-Author: Nothing to Disclose

Frank John Rybicki MD, PhD, Abstract Co-Author: Research grant, Toshiba Corporation Research grant, Bracco Group

Arash Bedayat MD, Abstract Co-Author: Nothing to Disclose

To evaluate variability in right-to-left ventricular (RV/LV) diameter ratios introduced by differences in CT 4-chamber reformation post-processing, and the subsequent influence on the accuracy of outcome prediction for patients with acute pulmonary embolism (PE).

RV/LV diameter ratios were measured using three different 4-chamber reformation techniques in 200 consecutive CT pulmonary angiograms positive for acute PE; 4ch-1 is a single oblique method using LV morphology landmarks to generate a 4-chamber view, and both 4ch-2 and 4ch-3 are so-called double oblique methods that create intermediate short-axis images to identify the maximum RV diameter, but with different approaches to reach short-axis images. Inter-observer variability was measured by repeating each reformation for 30 patients by a second examiner. Receiver Operating Characteristics (ROC) analysis compared accuracy of predicting outcome among the three reformation methods for PE-related death and a composite outcome of death or the need for intensive therapies.

The difference in median RV/LV diameter ratios was insignificant (p=0.293) between 4ch-2 (1.01) and 4ch-3 (1.02), whereas that from 4ch-1 dataset (0.93) was significantly (p<0.001) lower. Correlation between observers was excellent for all three datasets (r = 0.902 - 0.925). Compared to 4ch-1, both 4ch-2 and 4ch-3 datasets achieved higher accuracy in predicting PE-related 30-day mortality (Area Under the Curve (AUC): 0.55 vs. 0.69-0.70, p=0.008-0.019) and the composite outcome (AUC: 0.65 vs. 0.77-0.78, p=0.003-0.008).

Double oblique 4-chamber reformation methods are highly reproducible even with different approaches to the intermediate short-axis images. They also have higher accuracy in clinical outcome prediction for patients with acute PE compared to the single oblique reformation method.

When evaluating RV enlargement on a CT 4-chamber reformatted image for patients with acute PE, the double oblique reformation methods are recommended for better clinical outcome prediction.

Kumamaru, K, Wake, N, Hunsaker, A, Lu, M, Signorelli, J, Rybicki, F, Bedayat, A, Variability in RV/LV Diameter Ratios Measured on CT Pulmonary Angiography 4-Chamber Reformatted Images and Its Influence on Prediction of Patient Outcome.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11005766.html