| Background: The inferior extremity bile duct obstructive lesions mainly include benign disease such as calculus of bile duct and cholangitis and malignant disease such as carcinoma of common bile duct, cancer of ampullae and cancer of head of pancreas. It is difficult to give qualitative diagnosis according to clinical symptom and exhibition on CT because of similar clinical and imaging appearance. With development of medical imageology, many imaging methods are produced to study such disease of the location. Among these methods, spiral CT is the most important. However, there are some limitations to distinguish malignant and benign obstruction of biliary tract with spiral CT when the tumor is not clear. For the past few years, Binary logistic regression was introduced successfully to analyze the relation of the possibility factor and quantity of many phenomenon. Logistic regression model can calculate these variable's OR value and evaluate the value of these imaging feature to judge disease directly by regression equation.Objective: To explore the spiral CT distinctive imaging appearance of the inferior extremity bile duct obstructive lesions, then to analyze the correlation between clinical or imaging factors and benign or malignant disease and evaluate thevalue of factors in the different carcinoma distinguished.Materials and Methods: A retrospective study was undertaken to investigate the signs on CT and clinical data of 104 patients with obstructive lesions in the infenior extenity bile duct in Nanfang hospital from 2000-1 to 2004-12. All patients included 62 male and 42 female, from 4 to 83 years old, mean age 57, had been suffered from 5 days to 36 months, mean 3.83 months. There were 71 patients who had jaundice, 51 had weight loss, 10 had history of calculosis and 2 had history of operation on biliary tract. They were divided into 2 groups, benign and malignant. 26 with simple calculus of bile duct and 12 with cholangitis ( among these cases, 8 with calculus simultaneously) in benign group. In malignant group, there were 13 with cholangiocarcinoma in the lower bile duct, 20 with ampullary carcinoma and 33 with carcinoma of head of pancreas. All the results of laboratory test before operations were collected, including TBIL, DBIL, IBIL, ALT, AST, ALP, etc. Regular and dynamic enhanced spiral CT scan were underwent and 1.5-3mm reconstruction slice thickness at the site of obstructed area in all the patients. The following imageological signs were measured and collected, (f) Quantity signs: choledochus's diameter, first class bile duct's diameter, choledochus's wall thickness, the length from obstrction site to hepatic portal, gallbladder's size, wall thickness etc. ? Quality signs: tumor and its shape, enhanced degree; intrahepatic second class cholangiectasis degree, scope, duct wall enhanced degree, stone, double duct sign etc. All the diagnosis were confirmed by operation and pathology. Then, all clinical data and imageological signs were put into computers and statistical analysis by SPSS 10.0. Some statistics analysis methods were used such as t-test, analysis of variance, chi-square test and Fisher's exact test. We also used binary logistic regression analysis.Results: The clinical presentation of the patients were similar, which primarily were expressed as obstructive jaundice. Meanwhile, jaundice, weight loss andcalculosis history had significant difference between benign group and malignant group (P<0.05) . But operation history on biliary tract has no significant difference between the two groups (P>0.05) . Direct bilirubin, indirect bilirubin and alkali phosphatase had statistical differences between the two groups. While the average values of total bilirubin, ALT and AST have no statistical differences. Imaging appearances: in benign group, intrahepatic bile duct and extrahepatic bile duct dilate obviously and purulence bile image can be observed in some bile duct. In calculus cases high density image, similar soft tissue density image, sediment-like image and mixed density image was discovered. Mixed density stone had two kinds of appearances: CDcentric low density area was surrounded with high density image; ?punctiform high density image scattered in the centre of low density area. 30 cases with calculus in inferior extremity of common bile duct were diagnosed correctly by CT and correct ratio was 88.2%. In malignant group, intrahepatic bile duct and extrahepatic bile duct dilate was also observed and 11 intrahepatic bile duct medium dilation showed with soft vine appearance. Light bile duct dilation was observed in 40 cases. The average of inner diameter of common bile duct was 12mm. Obvious double duct sign was observed in 55 cases, cancer of head of pancreas, ampulla and distal bile duct were 33, 19, 3 cases respectively. We observed that some indexes, such as length from obstruction to hepatic portal, complicated intrahepatic stone, cholecystolithiasis, double duct sign and enhanced degree of bile duct wall, had significant difference between two groups (P<0.05) . The difference also was observed among common bile duct inferior extremity cancer, carcinoma of head of pancreas and ampullary carcinoma in malignant group (P<0.05) .The other indexes had no significant difference between the two group (P>0.05) . Judgment to mass: in benign group, mass was only found in 3 cases. In malignant group, obvious mass was observed in 33 cases, 32 suspected mass, no mass in 1 case. However, morphouses ofmass were discordantly. Most morphouses were not easily discerned. Round, oval-shap and sublobe lumps were rarely observed. Irregular lumps were relatively more. Enhancement of lump image: the cases in benign group were all not demonstrated with enhancement. In malignant group, 50% cases were not demonstrated with enhancement or expressed as light or medium degree. The cases in carcinoma of head of pancreas group were more demonstrated with no enhancement than those of ampulla (P<0.05) . we selected some clinical presentation and imaging signs to analyze with binary logistic regression model, such as jaundice, body weight decrease, calculosis history, ALP and lump, obstruction location, tubal wall enhanced degree, and double duct sign, etc. We discovered that the occurrence of lump was a strong index to judge the nature of lesion, benign or malignant. And weight loss, double duct sign and tubal wall enhanced degree also can help to judge the nature of lesion (PO.05) .Conclusions: It is very important to distinguish benign and malignant disease that mass, enhanced degree, weight loss, and double duct sign and ALP were discovered by binary logistic model. Meanwhile, mass, enhanced degree, double duct sign and obstruction location were helpful to distinguish different carcinoma of The inferior extremity common bile duct.
|
PDF Full Text Request