ABSTRACT
Objective:To investigate the predictive value of a nomogram based on clinical factors and gadobenate dimeglumine (Gd-BOPTA)-enhanced MRI for predicting the expression of Glypican-3 (GPC-3) in hepatocellular carcinoma (HCC).Methods:The clinical and imaging data of 85 patients with HCC confirmed by pathology in the Provincial Hospital of Shandong First Medical University from July 2018 to June 2021 were retrospectively collected. All the patients underwent Gd-BOPTA-enhanced MRI scan before operation. According to the expression of GPC-3 by immunohistochemistry, the patients were divided into GPC-3 positive group (55 cases) and GPC-3 negative group (30 cases). The clinical data of patients were collected, including gender, age, hepatitis, cirrhosis, alpha-fetoprotein (AFP), alanine aminotransferase, aspartate aminotransferase, and glutamine transferase levels. The MRI qualitative signs including tumor margin, ring enhancement, intratumoral hemorrhage, enhanced capsule, and satellite nodules were reviewed. MRI quantitative parameters including the largest tumor diameter, Gd-BOPTA-enhanced tumor-to-liver parenchyma signal ratio (TLR) and tumor enhancement ratio (TER) in arterial phase (AP), portal venous phase (PP), and hepatobiliary phase (HBP) were calculated. The independent sample t-test or Mann-Whitney U test were used to compare the quantitative data between the two groups, and the χ2 test was used to compare the qualitative data between the two groups. Multivariate logistic regression analysis was used to identify the independent predictors of GPC-3 expression, and a nomogram model was established. The receiver operating characteristic (ROC) curves were used to evaluate the predictive performance of each independent factor and nomogram, and DeLong test was used to compare differences in area under the curve (AUC). Results:There were significant differences in AFP, tumor margin, intratumoral hemorrhage, and TLR-AP, TLR-PP and TLR-HBP between GPC-3 positive and negative groups (all P<0.05). Multivariate logistic regression results showed that AFP≥20 μg/L, intratumoral hemorrhage and TLR-HBP were independent predictors of GPC-3 positive expression in HCC (OR=3.816, 4.788, 0.001, all P<0.05). The preoperative clinical and Gd-BOPTA-enhanced MRI nomogram model for predicting GPC-3 expression in hepatocellular carcinoma was established. The AUC of AFP≥20 μg/L, intratumoral hemorrhage, TLR-HBP and nomogram model in predicting GPC-3 positive expression were 0.688, 0.697, 0.808, and 0.879, respectively. The AUC of nomogram model was significantly better than those of the other three single indicator ( Z=3.82, 4.13, 2.04, P<0.001,<0.001,=0.042). Conclusion:The nomogram model based on indicators of clinical and qualitative and quantitative Gd-BOPTA-enhanced MRI has better performance in predicting the expression of HCC GPC-3 before surgery, which is higher than those of each single indicator.
ABSTRACT
Objective To classify the segmental bronchial patterns of the right upper lobe by combining bronchial tree and CT virtual bronchoscopy(CTVE)post-processing techniques from 64-slice spiral CT.Methods Two hundred and four patients with routine thorax scans were enrolled.The segmental bronchi were demonstrated in terms of bronchial tree and CTVE.First,we checked how many patients can be classified with any one of the above post-processing approaches.Then,integrating the two methods,we classified the fight upper segmental bronchial patterns of the 204 patients.At last,the patterns of the right upper bronchus were analyzed.Results bronchial tree post-processing images were stereoscopic and intuitive.It could be used to identify common stem of bifurcation easily,however,it was hard to be used to differentiate short common stem of bifurcation from trifurcation.CTVE demonstrated segmental bronchi from inside of lumen,and can readily show the short common stem of bifurcation and trifurcation of bronchi.Combining two post-processing techniques,the segmental bronchial ramification of the right upper lobe was mainly classified in three types:trifurcation in 76 patients(37.3%),common stem of apical and posterior segmental bronchi in 50 patients(24.5%)and others in 78 patients(38.2%).Conclusion The combination of multiple post-processing of 64-slice CT posses great superiority to classify the pattern of the right upper lobe bronchi.
ABSTRACT
Objective To classify the segmental bronchial patterns of the left upper lobe by combining three post-processing images from 64 slice spiral CT and to study how to identify different ramifications in transverse thin-section CT.Methods Totally 204 patients with routine thorax scans were enrolled.The segmental bronchi were demonstrated in terms of bronchial tree,virtual bronchoscopy and thin-section CT three post-processing images.Integrated with the three post-processing images,the segmental bronchial patterns of the left upper lobar bronchi were classified into several main types,and displayed in transverse thin-section CT.Results The segmental bronchial ramifications of the left upper lobe were classified into three types mainly:common stem of apical and posterior segmental bronchi(64%,130/200),trifurcation(23%,45/200),common stem of apical and anterior segmental bronchi(10%,21/200),and they could be identified in two typical slices of transverse thin-section CT.There were two dominating types in the left basilar segmental bronchi:bifurcation(75%,163/216),trifurcation(18%,39/216),and they could also be identified in two typical slices of transverse thin-section CT.Conclusion The segmental bronchi of the left lung can be definitely classified by three post-processing images from 64 slice spiral CT.