Stephane Dumas, Presenter: Nothing to Disclose

Donatella Distefano, Abstract Co-Author: Nothing to Disclose

Vincent Moutardier, Abstract Co-Author: Nothing to Disclose

Genevičve Monges, Abstract Co-Author: Nothing to Disclose

Serge Brunelle, Abstract Co-Author: Nothing to Disclose

Jean Robert Delpero, Abstract Co-Author: Nothing to Disclose

Dominique Margain, Abstract Co-Author: Nothing to Disclose

Blandine Le Brigand, Abstract Co-Author: Nothing to Disclose

Frederic Ternier, Abstract Co-Author: Nothing to Disclose

Inna Baklanova, Abstract Co-Author: Nothing to Disclose

Roland Rosello, Abstract Co-Author: Nothing to Disclose

Bernard Lelong, Abstract Co-Author: Nothing to Disclose

et al, Abstract Co-Author: Nothing to Disclose

Value of virtual hepatectomy (VH) to simulate the surgical margins in the preoperative evaluation of hepatic lesions.

Between November 1999 and march 2003, preoperative helical CT (2.5 mm collimation) and corresponding VH were performed on 155 patients expecting a surgical decision for metastatic or primary hepatic lesions. A iodinated contrast material was injected intravenously ( 140 mL ; rate, 2.0 – 3.0 mL/sec) for arterial and portal examination. From CT data, 3D models were generated with portal veins, hepatic veins and lesions visualized by transparency through the liver. The line of VH was drawn along the hepatic segmental boundaries defined by Couinaud. The simulated margins were measured between the lesions and the edge of the virtual piece of hepatectomy. Fifty three cases were confronted to the histopathologic measures of margins. The available follow-up of 31 patients were consulted to determine the exact location of recurrences. The following parameter - healthy or ill margins - were examined 6 months after the surgery. Then a Chi 2 test compared the group “initial negative margins” (≥10 mm) to the set “initial positive margins” (<10 mm)in order to detect an impact on the relapse.

Fourteen virtual line were negative among the 16 specimens with healthy margins (ss = 0.88) and 33 VH showed positive margins out of 37 surgical pieces with ill margins (sp = 0.89). The probability to predict a healthy margin by VH was 0.78 (14/18) and the one to perform a positive margin was 0.94 (33/35). We observed 47 out of 53 (88.7%) simulated lines (including positive and negative margins), which predicted resection’s margins agree with the histopathologic data. Moreover, the comparison of follow-up at 6 months between the group with initial healthy histopathologic margins and the set of actual pieces with positive margins showed no significant difference. The statistical Chi 2 test yielded a p-value equal to 0.73.

VH appears to be a sensitive tool to assess pre-operatively the margins. However, we didn’t observe any impact on the risk of hepatic relapse, 6 months after the surgery.

Dumas, S, Distefano, D, Moutardier, V, Monges, G, Brunelle, S, Delpero, J, Margain, D, Le Brigand, B, Ternier, F, Baklanova, I, Rosello, R, Lelong, B, et al, , Accuracy of Hepatectomy’s Margins Assessed Preoperatively by Virtual Hepatectomy Performed onto Three Dimensional Computed Tomography Hepatic Model (3DCT) with Histopathologic Comparison.  Radiological Society of North America 2004 Scientific Assembly and Annual Meeting, November 28 - December 3, 2004 ,Chicago IL. http://archive.rsna.org/2004/4407206.html