A tumour score with multidetector spiral CT for venous infiltration in pancreatic cancer: influence on borderline resectable.

PURPOSE: A tumour score for venous invasion in patients with pancreatic adenocarcinoma was evaluated by means of computed tomography (CT), in order to improve the assessment of medical treatment and clinical outcome with special attention to borderline resectable disease. MATERIALS AND METHODS: Fifty-six consecutive patients who underwent curative surgical resection for pancreatic cancer were analysed. On the basis of CT criteria, tumour involvement of the portal vein (PV) and superior mesenteric vein (SMV) was graded according to an adapted 4-point scale: score 1, definite absence of invasion; score 2, probable absence of invasion; score 3, probable presence of invasion; score 4, definite presence of invasion. Correlations between the venous infiltration scores and the patients' clinical features were also evaluated. RESULTS: After radiological evaluation of PV and SMV grades of infiltration, 21/56 (37 %) and 37/56 (66 %) patients, respectively, were found to have borderline resectable disease. The 4-point scale achieved a sensitivity of 80 %, a specificity of 96 % and an accuracy of 93 % in the evaluation of the PV, and a sensitivity of 100 %, a specificity of 94 % and an accuracy of 95 % in the evaluation of the SMV. Analysis of the distribution of clinical characteristics by PV and SMV infiltration showed that both scores correlated with the presence of distal metastasis (p = 0.016 and p = 0.028, respectively), and resection margins status (p = 0.015 and p = 0.006, respectively). CONCLUSIONS: This adapted tumour score is reliable for assessing venous invasion and might improve preoperative staging in patients with borderline resectable pancreatic cancer.

Clinical impact of MR cholangiopancreatography in patients with biliary disease.

PURPOSE: To evaluate the role of MR Cholangiopan-creatography (MRCP) as a first imaging modality in patients with suspected biliary tree pathology and indications to endoscopic retrograde cholangiopancreatography (ERCP). MATERIAL AND METHODS: Eighty-eight patients, with clinical signs of biliary tree pathology underwent MRCP, performed with a 1.5 T unit and a phased-array coil. Surgery, intraoperative cholangiography, percutaneous transhepatic cholangiography (PTC) or ERCP were regarded as the gold standard in patients with obstruction; the remaining patients underwent follow-up MRCP examinations at 6-9 months. The MR examination was performed with baseline T1w 2D FLASH and T2w TSE sequences, followed by the MRCP study (single-slab breath-hold RARE and multislice breath-hold HASTE sequences). The MR images were independently evaluated by two radiologists. RESULTS: MRCP showed normal findings in 20 patients; 68 patients had biliary duct dilatation. In 11 out of 68 patients MRCP did not identify any obstruction (9/11 were true negative cases). A diagnosis of benign obstruction was expressed in 36/59 patients (4 chronic pancreatitis, 29 choledocolithiasis, 4 inflammatory obstruction, 2 primary sclerosing cholangitis), with 1 false positive and 5 false negatives (sensitivity, specificity and diagnostic accuracy of 86%, 95% and 90%, respectively). MRCP identified 23 neoplastic stenoses (20/23 were true positives): the sensitivity, specificity and diagnostic accuracy values were 100%, 87% and 95%, respectively. MRCP correctly identified the level of obstruction in 100% of cases. CONCLUSIONS: MRCP may be considered as a first-step imaging method in patients with clinical signs of biliary disease. The workload of ERCP in the diagnostic stage could therefore be reduced and its use be reserved for therapeutic indications.

Preoperative T and N staging of colorectal cancer: accuracy of contrast-enhanced multi-detector row CT colonography--initial experience.

PURPOSE: To evaluate the accuracy of contrast material-enhanced multi-detector row computed tomographic (CT) colonography for preoperative staging of colorectal cancer. MATERIALS AND METHODS: Forty-one patients with colorectal carcinoma underwent preoperative contrast-enhanced multi-detector row CT colonography. Images were obtained in the arterial (start delay of 35 seconds) and portal venous (start delay of 70 seconds) phases. The arterial phase was focused on the suspected region of neoplasm, whereas the venous phase included the whole abdomen and pelvis. Two radiologists independently evaluated the depth of tumor invasion into the colorectal wall (T) and regional lymph node involvement (N) on transverse CT images alone and in combination with multiplanar reformations (MPRs). Disagreements were resolved by means of consensus. CT findings were compared with pathologic results, which served as the reference standard. Sensitivity, specificity, accuracy, positive predictive value, and negative predictive value were assessed. Differences in accuracy for T and N staging were assessed by using the McNemar test. RESULTS: In T staging, overall accuracy was 73% when transverse images were evaluated alone and 83% when they were evaluated in combination with MPRs. This difference was not significant. N staging was associated with an overall accuracy of 59% with transverse images alone and 80% with combined transverse and MPR images (P <.01). CONCLUSION: Contrast-enhanced multi-detector row CT colonography is an accurate technique for preoperative local staging of colorectal tumors.