Predicting Outcome in Pulmonary Embolism Using CT Angiography

Abstract & Commentary

Synopsis: No CT variables predicted severe in-hospital morbidity and mortality (death from pulmonary embolism, death from any cause, or cardiac arrest) in patients with PE. However, ventricular septal bowing and increased RV/LV diameter ratio were both strongly predictive of less severe morbidity, namely, subsequent ICU admission, and oligemia was associated with subsequent intubation and vasopressor use.

Source: Araoz PA, et al. J Thorac Imaging. 2003;18(4): 207-216.

The radiology group at the university of California, San Francisco, attempted to determine whether variables measured on the basis of CT angiography would predict in-hospital morbidity and mortality in patients with pulmonary embolism. Between January 1998 and January 2001, CT scans from 173 patients that were positive for pulmonary emboli were reviewed from 2 hospitals using (initially) single detector and (later) multidetector scanners. The scans were reviewed for the following potential indicators of adverse outcome: leftward ventricular septal bowing; increased right ventricle to left ventricle diameter ratio; clot burden, increased pulmonary artery to aorta diameter ratio; and oligemia. Outcome measurements included death from pulmonary embolism, death from any cause, cardiac arrest, and milder outcomes such as intubation, the use of vasopressors, or admission to the intensive care unit.

Ventricular septal bowing and increased RV/LV diameter ratio both predicted ICU admission, and oligemia was associated with subsequent intubation and vasopressor use. However, no CT variable predicted in-hospital morbidity and mortality in patients with pulmonary emboli. There were several limitations to the study, including the use of single detector scanners early on, a potential bias in patient selection due to the 2 hospitals’ lack of use of CT in 1997 for PE, a relatively small sample size, and technical limitations of CT angiography for evaluating right ventricular dysfunction.

Comment by James E. McFeely, MD

In the last several years, the use of CT angiography has greatly increased as a diagnostic test for proximal pulmonary emboli. The reasons for the increase are various and include the increased availability of multidetector helical scanners with improved sensitivity for clot, the ease of obtaining CT studies, and the associated difficulty in obtaining nuclear medicine studies, particularly during off-hours. CT technology for finding pulmonary emboli is rapidly improving. It is possible that, in the future, electrocardiographically gated helical multiplaner CT pulmonary angiography may be more useful for finding wall motion abnormalities.

As stated Araoz and colleagues, the surface echocardiogram remains the test of choice to evaluate right ventricular dysfunction in patients with acute pulmonary emboli. It is also probably a better choice if you are attempting to predict the in-hospital morbidity and mortality that may result from this phenomenon.

James E. McFeely, MD, Pulmonary and Critical Care Medicine, Alta Bates Summit Medical Center, Berkeley, CA, is Associate Editor for Critical Care Alert.