Logistic回归模型在肝硬化再出血患者心得安干预治疗监测中的作用——前瞻性4年随访应用研究  

Effect of Logistic regression model on monitoring efficacy of interfering treatment of inderal for patients with recurrent bleeding from hepatic cirrhosis： a 4-year prospective follow-up study

作　　者：郑永平[1] 许岳俊[1] 郭俊雄[1] 林奕芬[1] 

机构地区：[1]汕头市中心医院,广东汕头515031

出　　处：《中国中西医结合急救杂志》2011年第3期135-137,共3页Chinese Journal of Integrated Traditional and Western Medicine in Intensive and Critical Care

基　　金：广东省汕头市重点科研项目[汕府科（2004）65]

摘　　要：目的探讨采用基于脾静脉及肠系膜上静脉血流量建立的Logistic数字回归模型在选用、监测药物防治门脉高压再出血疗效中的临床应用价值。方法选择有门脉高压出血史的肝硬化食管静脉曲张破裂出血患者，服用心得安后采用彩色多普勒测定门静脉、脾静脉及肠系膜上静脉血流量，同时采集血液学指标，依据Logistic回归模型计算P值，如P值〈0．5为达标组；否则为不达标组，则加服五苓散3个月后再次行多普勒检查，如P值〈0．5作为五苓散应答组，否则为无应答组。观察患者继续原方案用药1年内及长期服药随访3年的再出血率。结果40例患者完成研究，用药1年内达标组再出血率明显低于不达标组[17．6％（3／17）比52．2％（12／23），P〈0．05]。3年随访期内，31例完成终点随访，加服五苓散后达标组再出血率低于不达标组[57．1％（8／14）比88．2％（15／17）]，但差异无统计学意义（P〉0．05）；在持续服用五苓散的患者中，服药后有效应答组再出血率较无应答组明显降低[40．0％（4／10）比84．6％（11／13），P〈0．05]。结论Logistic数字回归模型能方便了解用药后门脉高压患者腹腔血管动力循环改变，P值0．5可作为药物选用与剂量调整的目标截断值，能达标者早期再出血危险性下降。目前常规应用的心得安达标率低，部分患者对加用五苓散后有治疗反应，值得扩大病例数深入研究。Objective To investigate the clinical significance of Logistic regression model created on the basis of color Doppler blood flow quantity in portal, splenic and superior mesentery veins （Qspv and Qsmv） parameters for predicting efficacy of interfering treatment for re-bleeding from hepatic cirrhosis. Methods Fifty-nine eases with hepatic cirrhosis complicated with portal vein hypertension and esophageal varices bleeding history were enrolled ; after taking inderal, the parameters of Qspv and Qsmv were checked by color Doppler instrument and analyzed with the Logistic regression model established in previous study. In the mean time, hematological indexes were collected. The response group was defined as the result of calculation reaching P〈0.5 and non-response group, P〉0.5. Wulingsan （五苓散） was applied for patients in the latter group and 3 months later they were rechecked by Doppler examination. The re-bleeding rates in patients treated for 1 year and those long-term treated for 3 years and follow-up were observed. Results Forty cases finished the study in 1 year, re-bleeding rate in response group was lower than that in non-response group [17.6% （3/17） vs. 52.2% （12/23）, P〈0. 05]. In 3 years of follow-up, 31 eases reached the end-point, the re-bleeding rate in response group with the addition of Wulingsan was lower than that in non-response such group ]57.1% （8/14） vs. 88.2% （15/17）], but no statistical significant difference was seen （P〉0. 05）, however the re-bleeding rate was significantly lower in response group of patients persistently taking Wulingsan than that in such Wulingsan non-response group （40. 0% （4/10） vs. 84. 6% （11/13）, P〈0. 053. Conclusion Logistic regression model based on Qspv and Qsmv parameters can facilitate to realize the abdominal vascular dynamic circulatory changes of eases with portal vein hypertension after drug treatment. P of 0. 5 can be a cut-off point to select drug and regulate dosage. After treatment, the patients whose thera

关 键 词：LOGISTIC回归模型 肝硬化 再出血 干预治疗 监测 

分 类 号：R285.6[医药卫生—中药学] R575.2[医药卫生—中医学]