POSTOPERATIVE LOCALIZATION OF PORTA-HEPATIS AND ABDOMINAL VASCULATUREIN PANCREATIC MALIGNANCIES - IMPLICATIONS FOR POSTOPERATIVE RADIOTHERAPY PLANNING

Purpose: To evaluate changes in preoperative and postoperative positio ns of structures used to define target volumes (i,e., pancreatic bed, porta hepatis, local-regional lymph nodes) for postoperative irradiati on of pancreatic malignancies as defined by abdominal computed tomogra phs. Methods and Materials: Eleven consecutive patients who had Whippl e resection and postoperative irradiation for pancreatic cancer were e valuated. Preoperative and postoperative computed tomographs of each p atient were evaluated for the position of the portal vein bifurcation and the origin of the celiac axis and superior mesenteric artery. The length along the x (medial-lateral position) and y (anterior-posterior position) axes was determined with calipers to the closest millimeter . Length along the z axis (cephalad-caudad position) was determined wi th the computed tomographic sectional interval between images. Statist ical significance of the change in the structure's position along the x, y, or z axis between preoperative and postoperative computed tomogr aphs was assessed with the paired t-test. Results: Evaluation of the p reoperative and postoperative positions of the portal vein, celiac aid s, and superior mesenteric artery along the x, y, and z axes revealed a statistically significant change in the location of the portal vein and celiac axis postoperatively. The median change of the celiac axis in the anterior-posterior position was significant (p = 0.0047), but t he mean change was only 2 mm and not considered clinically significant . The median change for the portal vein was 0.97 cm and 1.07 cm along the y and x axes, respectively, and was significant (p = 0.008 and p = 0.0001). The range in position change for the portal vein was 0.0 to 2.0 cm along the y axis and 0.4 to 1.9 along the x axis. The remaining mean changes in position along all axes for all the structures were l ess than 3 mm (not statistically significant). Conclusions: The mean p osition of the portal vein-porta hepatis after Whipple resection is ap proximately 1.0 cm medial and 1.0 cm posterior compared with its preop erative position. These data suggest that postoperative abdominal comp uted tomographs are useful in determining treatment volumes of nodal d rainage basins after Whipple resection of pancreatic malignancies. (C) 1997 Elsevier Science Inc.