ThepredictiveanalysisofspectralCTportalveinimagingforesophagealGgastricvaricealbleeding / 实用放射学杂志

Objective ToinvestigatethevalueoftheoptimalCNR monoenergeticimagesofportalveininpredictingesophagealGgastric varicealbleeding(EVB).Methods FortypatientswithlivercirrhosisandportalhypertensionunderwentcontrastGenhancedspectral CTscan.TheoptimalCNR monoenergeticimagesand70keV monoenergeticimagesoftheportalveinphasewereselectedtocompare theimagingquality.TheconsistencyoftheoptimalCNR monoenergeticimageswithendoscopyforesophagealGgastricvaricealwas analyzed by Kappa test.A ccording to the E V B history ,the patients w ere divided into bleeding group and nonGbleeding group.T he ROCcurveswerecalculatedtoassessthevaluesofthemainportalvein(MPV)anditstributariesdiameterstopredictingEVB.Results TheCNRandobjectivescoreoftheoptimalCNR monoenergeticimagesweresignificantlyhigherthanthoseofthe70keVimages(P＜0.05).TheoptimalCNR monoenergeticimageshadgoodconsistency withgastroscopeintheclassificationofesophagealGgastric variceal(k＝0.769).ThereweresignificantdifferencesofthediametersofMPV,intrahepaticleftbranchofportalvein(LPV),splenic vein(SV)andleftgastricvein (LGV)betweenthebleedinggroupandnonGbleedinggroup (P＜0.05).ROCcurveanalysisshowed thattheincidenceofEVBwashighestwithasensitivityof76.92%andaspecificityof85.71%,whentheLGVwasgreaterthan6.1mm. Conclusion TheoptimalCNR monoenergeticimagesofportalveincouldimprovetheimagingquality,andLGVdiametershouldbe ariskindicatorforpredictingEVBinlivercirrhosis.

The clinical significance of CT angiography in the diagnosis of arterioportal shunts associated with hepatocellular carcinoma / 中华放射学杂志

Objective To evaluate the clinical significance of CT angiography(CTA)in the diagnosis of arterioportal shunts(APS)associated with hepatocellular carcinoma(HCC).Methods One hundred and twenty-seven consecutive HCC patients accepted both dynamic enhancement CT and DSA examinations.The interval between CT and DSA exam was from 3 to 15 days.Based on transverse CT images in hepatic artery phase,CTA was performed for all the patients.By contrast with DSA results,the capabilities of transverse CT and transverse images combined with CTA in APS diagnosis were analyzed. Results In all 127 HCC cases,52 cases with APS were confirmed by DSA(40.94%),33 with central type of APS and 19 with peripheral type.Diagnostic sensitivity of APS based on transverse CT and combined CTA with transverse CT images were both 94.23%(49/52).However,specificity was 84.00%(63/75) and 97.33%(73/75),respectively,accuracy was 88.19%(112/127)and 96.06%(122/127),the predictive value of positive cases was 80.33%(49/61)and 96.08%(49/51),and the predictive value of negative cases was 95.45%(63/66)and 96.05%(73/76).Combined with CTA,false positive cases of 4 central type of APS and 6 peripheral type of APS were excluded which were demonstrated by transverse CT images.By contrast with DSA,the coincidence rate of the type of APS diagnosed by transverse images combined with CTA was 88.46%(46/52),including 90.91%(30/33)of central type of APS and 84.21%(16/19)of peripheral type.The supplying arteries of central type of APS were intuitively displayed by CTA in 23 cases,19 from proper hepatic artery and 4 from gastro-duodenal artery.Conclusion CTA techniques based on the dynamic enhancement CT exams could effectively promote the specificity and the accuracy of APS diagnosis.

Correlation Between Contrast Enhancement of Portal Vein and Spleen Size in Dual-phase Spiral CT

URPOSE: To evaluate using spiral CT the effect of spleen size on blood flow in the portal venous system andto know the usefulness of this evaluation. MATERIALS AND METHODS: Fifty-one patients without evidence on spiralCT scan of abnormality thought to affect portal venous flow presented between December 1994 and June 1995. We measured spleen size and Hounsfield units of portal vein in dual-phase, and calculated the ratio of the unit inthe portal phase to that in the arterial phase. Spleen size was measured, using the length of X-axis by that of Z-axis on spiral CT scan. We then measured the correlation between the two values. CT was performed with a Somatom Plus-S scanner(Siemens, Erlangen, Germany). A total dose of 120ml of non-ionic contrast material(Ultravist) was administered at a rate of 3ml/sec. Arterial and portal phase were obtained after 30 seconds and 60 seconds fromthe beginning of the contrast agent injection. RESULTS: The correlation between spleen size and contrast enhancement of the portal vein was relatively significant(Pearson's correlation coefficient(r)=0.41801). CONCLUSIONS: Spleen size significantly affects portal venous flow on spiral CT scan. The evaluation of spleensize and contrast enhancement of the portal vein could be useful in the differential diagnosis of diseases which affect portal venous flow.