Chantale Belanger MD, Presenter: Nothing to Disclose

Marie E Faughnan MD, Abstract Co-Author: Nothing to Disclose

Gilles P. Soulez MD, Abstract Co-Author: Speaker, Bracco Group Speaker, Siemens AG Research grant, Siemens AG Research grant, Bracco Group Research grant, Cook Group Incorporated Research grant, Object Research Systems Inc

Marie-France Giroux MD, Abstract Co-Author: Nothing to Disclose

Vincent Louis Oliva MD, Abstract Co-Author: Nothing to Disclose

Patrick Gilbert MD, Abstract Co-Author: Nothing to Disclose

Eric Therasse MD, Abstract Co-Author: Research support, Johnson & Johnson

To evaluate the sensitivity and specificity of non-enhanced CT to detect PAVMs reperfusion after embolization.

The imaging files of all patients who had pulmonary angiography follow-up after PAVM embolization in our center from 2000 to 2010 were reviewed. PAVM feeding artery, draining vein and aneurismal sac diameters were measured on non-enhanced CT before and after embolization. Follow-up pulmonary angiography was the gold standard to assess reperfusion. ROC curves were used to assess sensitivity and specificity according to the percentage of diameter reduction.

From the 67 patients who had PAVM embolization, 39 (58%) had no follow-up angiography, 9 (13%) no longer had pre or post embolization CT available and 19 (28%) patients had 52 PAVMs assessed by pulmonary angiography, performed during additional PAVMs embolizations (5 patients, 26%) or because of suspicion of PAVM reperfusion on CT (14 patients, 74%). Twenty-seven (51%) PAVMs were occluded and 25 (49%) were reperfused, 18 (72%) of these were re-embolized. Mean diameter reduction in occluded versus reperfused PAVMs were respectively 51,6% +/- 22,6% versus 26,5% +/- 20,7% for the vein and 60,1% +/- 19,5% versus 25,9% +/- 23,8% for the aneurismal sac. A 50% diameter reduction of the PAVM feeding artery, vein and aneurismal sac respectively showed a sensitivity/specificity of 76%/78%, 88%/62%, and 72%/78% for reperfusion.

There is a wide overlap in diameter reduction between occluded and reperfused PAVMs after embolization. A 50% diameter reduction cutoff for the aneurismal sac on CT would still miss 25% of reperfused PAVMs.

Sensitivity and specificity of non-enhanced chest CT to detect PAVM reperfusion after embolization is highly dependant on the selected PAVM diameter reduction threshold, which should be standardized.

Belanger, C, Faughnan, M, Soulez, G, Giroux, M, Oliva, V, Gilbert, P, Therasse, E, PAVMs Reperfusion after Percutaneous Embolization: Sensitivity and Specificity of Nonenhanced CT Assessed with Pulmonary Angiography.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11004514.html