| BackgroundEsophageal varices(EV) in patients with hepatic cirrhosis is one of main performances of portal hypertension, its incidence rate is about60%-80%. About30%～40%of patients with hepatic cirrhosis occurre varicosis rupturting and hemorrhaging. At present, endoscopic examination for the dia-gnosis of esophageal varices is the gold standard, and it can find the presence of esophageal varices and predict the risk of varicosis rupturting in the patients. But the main drawbacks of the endoscopy examination are not easy to be tolerance by patients, and poor compliance. And the processes of examination probably induce hemorrhaging. This study mainly researches whether the noninvasive indexes, such as sublingual collaterals, portal vein diameter (PVD), spleen thickness (ST) and the platelet count(PLT) can predict the degree of esophageal varices. But due to that it is generally belived the abnormalities of sublingual collaterals are related to blood stasis syndrome, it needs to analyse the correlation of sublingual collaterals and blood stasis syndrome in hepatic cirrhosis patients firstly so as to study the correlation between sublingual collaterals and EV better.These noninvasive indexes, comparing to gastroscopy, have better compliance by patients, can find the esophageal varices ealrier, and guide clincal better.ObjectiveFirstly to study the correlation between sublingual collaterals and blood stasis syndrome in patients with hepatic cirrhosis, and then studying the correlations between the degree of esophageal varices and the noninvasive indexes, such as sublingual collaterals, portal vein diameter (PVD), spleen thickness (ST) and the platelet count (PLT).And to find the best combination of noninvasive indexes which can predict the degree of esophageal varices better.MethodsClinical epidemiology survey methods were used. In strict accordance with hepatic cirrhosis diagnosis standard, a total of173patients were enrolled. Firstly it was performed on patients with a cross-sectional observation, collection of information by four diagnostic methods of traditional Chinese medicine and syndrome differentiation, then divided the patients into blood stasis syndrome group and non blood stasis syndrome group, in orde to analyse the relationship between sublingual collaterals and blood stasis syndrome. Patients were examined by gastroscopy, then the degree of EV were defined, which can be divided into no esophageal varices, mild, moderate, severe, four grades. Sublingual collaterals in patients were observed, and divided into normal, I, II, III, four grades. And patients should undergo B ultrasonography and have blood routine tests in order to record the diameter of the portal vein, thickness of spleen and platelet count. All of the examinations were completed in1week, gastroscopy and B ultrasonography were in charged by specially-assigned doctors. The SPSS18.0statistical software was used to set up a database and analysis data. Firstly to compare the differences of sublingual collaterals grading in patients between blood stasis syndrome group and non blood stasis syndrome group and study the grades of sublingual collaterals whether would be changed by the two differnt groups. And then the correlation between sublingual collaterals, PVD, ST, PLT and the degree of EV would be studied. Finally logistic regression was used to analyse the data, in order to find best combination of noninvasive indexes which can predict the grades of EV in patients with hepatic cirrhosis.Results1. There was no obviously statistical difference of sublingual collaterals grading between blood stasis syndrome group and non blood stasis syndrome group (P>0.05). And the differences of PVD, ST, PLT between the two groups were not statistically significant. So it was no need to divide patients into blood stasis syndrome group and non blood stasis syndrome group, when studying the correlations between every noninvasive indexes and the degree of EV. But it was found that the degree of EV in the blood stasis group were severer than that in the non blood stasis syndrome group (P<0.05), and the incidence of severe esophageal varices in blood stasis group was higher than non blood stasis syndrome group(P<0.05), up to48.7%. It meant that the factor of blood stasis syndrome or not probably can be used to predict esophageal varices.2. The relationship between sublingual collaterals grading and esophageal varices grading was linear trend (The values of linear-by-linear association were18.794, P<0.05). It meant that in general, when the grades of esophageal varices were higher, the grades of sublingual collaterals grade were higher, too. And sublingual collaterals grading in the patients of severe esophageal varices is level III mostly.3. Both PVD and ST had positive correlations with the degree of EV, the values of rank correlation coefficient rs were0.242and0.386(P<0.05). Both means of PVD and ST in the severe esophageal varices group were greater than that in the non severe esophageal varices. But the PLT had negative correlation with the degree of EV, and the values of rank correlation coefficient rs were-0.468(P<0.05). The mean of PLT in severe esophageal varices group was greater than non severe esophageal varices(P<0.05).4. Uesing logistic regression to analyse, it was found that PLT, sublingual collaterals and blood stasis syndrome or not, these three noninvasive indexes for the prediction of esophageal varices grading had statistical significance (P<0.05). And those three noninvasive indexes could be uesd to establish a logistic regression model which can predict the probabilities of every grades of EV in patients. To predict the probability of severe esophageal varices, but only PLT was statistically significant(P<0.05), and when its critical values of67.00X10e9/L, the prediction of severe esophageal varices had sen-sitivity of76.4%and specificity of65.7%.ConclusionSublingual collaterals grading in hepatic cirrhosis patients with blood stasis syndrome did not differ obviously from the patients with non blood stasis syndrome. And sublingual collaterals, PVD, ST, PLT and blood stasis syndrome or not, these noninvasive indexes, correlated with the degree of EV. And the combination of sublingual collaterals, PLT and blood stasis syndrome or not, these three indexes, can predict the degree of EV more significantly. The logistic regression model can be used to predict the probabilitise of every grades of EV.But for the prediction of severe esophageal varices, only PLT was needed.All of these provided a noninvasive, simple and quick way for the prediction of EV and prevention of varicosis rupturting and hemorrhaging, patients can also accepted it more easily. |
PDF Full Text Request