Exenteration and Assessement of Loco-Regional Resectability

Objectives: To evaluate the accuracy of preoperative MRI scans in the assessement of loco-regional tumorresectability of pts. with exenterations.

Methods: A retrospective chart review of all pts. undergoing pelvic exenteration by one single gynaecologic oncology service (Klinikum Leverkusen) between 1999 and 02/2005 was conducted. Data on multiple variables of the MRI scans were analysed including pelvic side wall infiltration, ureter obstruction, vessel involvement, peritoneal disease. To avoid biased interpretation of radiographic findings all MRI scans were re-evaluated in a blinded fashion. With each variable the sensitivity and specificity for surgical resectability (R0/R1 vs. R2) and for the final histological margin status were calculated. Calculations were performed with 2×2 tables according to the various investigated risk factors.

Results: 63 pts. were identified: all of them underwent exenteration. The histopathological investigation of the exenteration specimen demonstrated R0 resection status in 23 cases, i.e. the prevalence of microscopically tumor-free margins was 36 %. In 40 cases, microscopically tumor-free margins could not be achieved. The sensitivity varied between 43 % and 71 % for the macroscopic evaluation during surgery and 39 % and 85 % for the final microscopic result based upon the preoperative assessement of the radiographic results. Accordingly the specificity varied between 59 % and 62 % and 49 % and 56 % (as above). Diagnostic accuracy was particularly low in pts. with a history of previous radiochemotherapy treatment.

Conclusions: The diagnostic accuracy of preoperative radiographic diagnostic procedures is low and therefore alone no appropriate tool for assessing the risk of surgical resectability and the final margin status in pts. undergoing eviscerations.