中国医科大学学报  2022, Vol. 51 Issue (6): 529-533

文章信息

孙逸飞, 张伟鹏, 黄丽萍
SUN Yifei, ZHANG Weipeng, HUANG Liping
脾脏剪切波弹性成像对需要治疗的乙型肝炎肝硬化食管胃底静脉曲张的无创评估
Noninvasive evaluation of esophageal and gastric varices in patients with hepatitis B cirrhosis using splenic shear wave elastography
中国医科大学学报, 2022, 51(6): 529-533
Journal of China Medical University, 2022, 51(6): 529-533

文章历史

收稿日期:2021-06-22
网络出版时间:2022-06-01 17:00
脾脏剪切波弹性成像对需要治疗的乙型肝炎肝硬化食管胃底静脉曲张的无创评估
孙逸飞1 , 张伟鹏2 , 黄丽萍1     
1. 中国医科大学附属盛京医院超声科,沈阳 110013;
2. 中国医科大学附属盛京医院普通外科,沈阳 110013
摘要目的 研究剪切波弹性成像(SWE) 测量的脾脏硬度(SS) 对乙型肝炎肝硬化患者是否发生需要治疗的食管胃底静脉曲张(VNT) 的无创评估能力,并与其他无创评估指标,包括血小板(PLT) 数、白蛋白(ALB)、总胆红素(TBIL)、丙氨酸氨基转移酶(ALT)、门冬氨酸氨基转移酶(AST)、脾厚度、AST和PLT比率指数(APRI)、肝纤维化4因子指数(FIB-4) 的评估能力对比。方法 对2017年4月至2020年12月于中国医科大学附属盛京医院行胃镜检查的70例乙型肝炎肝硬化患者及42例正常对照者行常规腹部超声及脾脏SWE检查,收集一般资料、血清学及胃镜检查结果。所有SS测量成功的受试者依据有无肝硬化、有无VNT进行分组。采用Mann-Whitney U检验对肝硬化VNT组、非VNT组与正常对照组两两比较,采用logistic回归分析有统计学意义的指标对VNT的预测能力。采用受试者操作特征(ROC) 曲线评估有预测能力指标的诊断效能。结果 乙型肝炎肝硬化患者SS测量成功64例,合并VNT 23例。肝硬化VNT组、非VNT组与正常对照组两两比较各指标差异均有统计学意义。仅PLT、ALB、FIB-4、脾厚度及SS的差异对区分肝硬化患者有无VNT有统计学意义。logistic回归分析上述指标对VNT的预测能力,仅SS和脾厚度有意义。该模型灵敏度、特异度、阳性预测值、阴性预测值分别为78.3%、92.7%、85.7%、88.4%。脾厚度和SS及二者联合的ROC曲线下面积分别为0.879、0.901、0.924。结论 脾厚度和SS对VNT诊断有预测能力,二者联合可提高VNT诊断能力。
关键词剪切波弹性成像    肝硬化    食管胃底静脉曲张    脾脏硬度    
Noninvasive evaluation of esophageal and gastric varices in patients with hepatitis B cirrhosis using splenic shear wave elastography
1. Department of ultrasound, Shengjing Hospital of China Medical University, Shenyang 11001, China;
2. Department of General Surgery, Shengjing Hospital of China Medical University, Shenyang 11001, China
Abstract: Objective To study the noninvasive evaluation ability of spleen stiffness (SS) measured by shear wave elastography (SWE) on the occurrence of varices needing treatment (VNT) in patients with hepatitis B liver cirrhosis and compare it with other noninvasive indexes, such as platelet (PLT) number, albumin (ALB), total bilirubin (TBIL), alanine aminotransferase (ALT), aspartate aminotransferase (AST), spleen thickness, AST and PLT ratio index (APRI), and hepatic fibrosis 4 index (FIB-4). Methods A total of 70 patients with hepatitis B liver cirrhosis who underwent gastroscopy in Shengjing Hospital of China Medical University from April 2017 to December 2020 and 42 healthy controls were subjected to routine abdominal ultrasound and splenic SWE examination. The general data, serological test results, and gastroscopy results were collected. All valid subject data were grouped based on cirrhosis and presence of VNT. The Mann-Whitney U test was used to assess whether there were statistical differences among healthy controls, VNT, and non-VNT groups. Logistic regression analysis was used to analyze and compare the predictive ability of the statistically different indicators for the diagnosis of VNT. The receiver operating characteristic (ROC) curve was used to assess the diagnostic efficacy of the predictors. Results SS measurement was successfully performed in 64 patients with hepatitis B liver cirrhosis, including 23 patients with VNT. In 42 cases of the healthy controls group, 37 cases were successfully measured. The differences of each index between the healthy controls, VNT, and non-VNT groups were statistically significant. In patients with liver cirrhosis, the Mann-Whitney U test results showed that only PLT, ALB, FIB-4, spleen thickness, and SS were statistically significant when distinguishing the presence of VNT (P < 0.05). Only SS and spleen thickness were significant for the predictive ability of VNT by logistic regression analysis. The sensitivity, specificity, positive predictive value, and negative predictive value of the predictive model were 78.3%, 92.7%, 85.7%, and 88.4%, respectively. The area under curve of spleen thickness and SS and their combination were 0.879, 0.901, and 0.924, respectively. Conclusion Spleen thickness and SS have the predictive ability in the diagnosis of VNT and their combination can improve the diagnostic ability of VNT.

肝硬化门静脉高压主要表现为脾功能亢进、肝性脑病、腹水及食管胃底静脉曲张(gastroesophageal varices,GOV) 出血等,多数患者的主要死因为GOV破裂出血[1]。约50%的肝硬化患者发生GOV,但发展至需要治疗的食管胃底静脉曲张(varices needing treatment,VNT) 者仅占15%~25%[2],故大部分胃镜检查的患者无GOV,或虽有GOV但暂无需预防性治疗。现行的指南和共识[1, 3]提出应对VNT进行一级预防,所有大型GOV (扩张静脉直径≥5 mm) 及Child-Pugh C型患者的小型GOV (扩张静脉直径 < 5 mm) 或小型GOV伴有红斑征象均为VNT。

准确评估GOV风险的方法有2种,即肝静脉压力梯度检查(诊断门静脉高压金标准) 和胃镜检查(诊断GOV金标准)。但二者均有侵入性,长期随访困难。因此亟需非侵入性手段准确评估及预测VNT。门静脉高压肝硬化患者常伴随脾增大,脾内高动力血液循环较肝内更明显,与门静脉充血及脾淋巴组织过度活化、血管生成增加和肝纤维化发展有关。脾脏硬度(spleen stiffness,SS) 测量不受肝炎、脂肪肝的影响,可减少相应误差,理论上SS比肝脏硬度(liver stiffness,LS) 更敏感。因此,本研究通过应用剪切波弹性成像(shear wave elastography,SWE) 测量SS对乙型肝炎(简称乙肝) 肝硬化患者有无VNT进行评价。

1 材料与方法 1.1 研究对象

选择2017年4月至2020年12月于中国医科大学附属盛京医院行胃镜检查的70例乙肝肝硬化患者和42例体检肝功能及肝胆脾超声检查正常的健康对照者。所有研究对象均签署知情同意书。乙肝肝硬化的诊断标准根据《慢性乙型肝炎防治指南(2019年版) 》 [4]。排除门静脉内有血栓及脾梗死,合并肝癌,脾脏原发病,接受GOV治疗(如肝脾介入治疗、经颈静脉肝内门静脉分流术等),脾切除者。

1.2 方法

1.2.1 超声检查

选用法国Supersonic Aixplorer型SWE超声诊断仪,SC6-1型凸阵探头,频率1~6 MHz。常规超声检查时,受检者禁食8 h以上,仰卧位,双上肢抬至头部,检查肝、脾回声、大小及有无占位等。弹性检查时,将探头置于左侧腋中线或腋后线紧贴9~11肋间隙,使二维图像清晰显示脾脏,放大图像至120%,切换SWE模式,取样框范围4 cm×3 cm,感兴趣区10.0~20.0 mm,SCALE 70 kPa;上缘选定脾包膜下约1~2 cm的脾实质,避开大血管,嘱受检者屏气3~5 s,图像均匀稳定后测量SS值;同一患者测量3次,取平均值。

1.2.2 收集资料

收集患者一般资料、胃镜检查和血清学检查结果,包括性别、年龄、血常规、肝功能。计算血清模型肝纤维化4因子指数(fibrosis 4 index,FIB-4) =年龄(岁) ×门冬氨酸氨基转移酶(aspartate aminotransferase,AST,U/L) / [血小板(platelet,PLT) 数(109 /L) ×丙氨酸氨基转移酶(alanine aminotransferase,ALT,U/L) 1/2],AST和PLT比率指数(aspartate aminotransferase to platelet ratio index,APRI) =AST (U/L) /AST正常值上限(U/L)×100/PLT (109 /L)。

1.3 统计学分析

采用SPSS 23.0软件进行统计学分析。首先采用Mann-Whitney U检验两两比较肝硬化VNT组、非VNT组与正常对照组,分析每2组各指标间差异有无统计学意义。采用logistic回归分析上述指标对VNT诊断的预测能力并进行比较。采用受试者操作特征(receiver operating characteristic,ROC) 曲线评估有预测能力指标的诊断效能,评价各因素组合的模型,比较ROC曲线下面积(area under curve,AUC)。用约登指数确定最佳截断值,并计算灵敏度及特异度。P < 0.05为差异有统计学意义。

2 结果 2.1 肝硬化VNT组、非VNT组与正常对照组各指标比较

70例乙肝肝硬化患者中,SS测量成功64例,其中,VNT 23例(VNT组),非VNT 41例(非VNT组)。正常对照组42例,SS测量成功37例。分析显示数据不符合正态分布,所有连续性变量采用M (P25~P75) 表示,见表 1。Mann-Whitney U检验结果显示,肝硬化VNT组和非VNT组与正常对照组的各指标两两比较均有统计学差异(均P < 0.05)。肝硬化VNT组与非VNT组PLT、ALB、FIB-4、脾厚度以及SS有统计学差异(均P < 0.05)。见表 2

表 1 正常对照组、肝硬化非VNT组及VNT组各评价指标比较[M (P25~P75)] Tab.1 Index comparision of normal group, non-VNT group and VNT group [M (P25-P75)]
Index Healthy control group (n = 37) Non-VNT group (n = 41) VNT group (n = 23)
PLT (×109/L) 243.0(197.0-292.0) 136.0(90.5-187.5) 80.0(46.0-131.0)
ALB (g/L) 44.9(40.5-46.6) 42.9(40.6-45.2) 39.4(31.7-42.0)
TBIL (µmol/L) 10.9(8.8-14.4) 15.4(10.9-21.4) 19.3(13.7-24.1)
ALT (U/L) 12.0(7.0-17.0) 27.0(19.0-37.0) 21.0(16.0-39.0)
AST (U/L) 16.0(13.0-18.0) 24.0(20.5-42.0) 25.0(18.0-52.0)
APRI 0.16(0.13-0.20) 0.53(0.30-1.05) 1.07(0.46-1.56)
FIB-4 0.63(0.50-0.78) 1.48(0.99-3.83) 4.48(2.51-7.02)
Spleen thickness (cm) 3.1(2.8-3.4) 3.8(3.2-4.5) 5.4(4.7-6.1)
SS (kPa) 16.3(14.3-19.0) 24.5(21.3-32.9) 43.6(38.1-47.3)
SS,spleen stiffness;PLT,platelet count;ALB,albumin;TBIL,total bilirubin;ALT,alanine aminotransferase;AST,aspartate aminotransferase;APRI,AST and PLT ratio index;FIB-4,hepatic fibrosis 4 index.

表 2 肝硬化VNT组、非VNT组与正常对照组各评价指标两两比较(P) Tab.2 Comparison of each index between liver cirrhosis VNT group, non-VNT group and healthy control group (P)
Index VNT group vs non-VNT group VNT group vs healthy control group Non-VNT group vs healthy control group
PLT 0.011 < 0.001 < 0.001
ALB 0.005 0.001 0.009
TBIL 0.152 < 0.001 0.002
ALT 0.275 < 0.001 < 0.001
AST 0.634 < 0.001 < 0.001
APRI 0.056 < 0.001 < 0.001
FIB-4 0.004 < 0.001 < 0.001
Spleen thickness < 0.001 < 0.001 < 0.001
SS < 0.001 < 0.001 < 0.001

2.2 各评价指标对VNT预测能力比较

logistic回归分析上述有统计学意义的的预测因素,结果显示,脾厚度与SS有统计学意义(P < 0.05),二者可用于预测VNT,见表 3。所获模型拟合度良好,有统计学意义(P < 0.05)。模型能正确分类87.5%的研究对象,灵敏度78.3%,特异度92.7%,阳性预测值85.7%,阴性预测值88.4%。

表 3 各评价指标预测VNT的logistic回归分析结果 Tab.3 Logistic regression analysis results of each index predicting VNT
Index B Exp (B) P 95% CI for Exp (B)
Lower limit Upper limit
Spleen thickness 0.740 2.096 0.040 1.035 4.241
SS 0.148 1.159 0.003 1.053 1.276
Since the P values of the other indicators were all > 0.05,they were meaningless and were not included in the final equation.

2.3 单独和联合指标对VNT的诊断效能

将脾厚度和SS以及二者联合后的预测概率绘制ROC曲线,AUC分别为0.879、0.901及0.924,见图 1。诊断效能由高至低依次为二者联合> SS > 脾厚度。根据约登指数计算,联合诊断、单独使用SS诊断、单独使用脾厚度诊断的灵敏度分别为100%、100%、78.3%,特异度分别为75.6%、73.2%、87.8%,单独使用SS或脾厚度诊断的最佳截断值分别为31.4 kPa、4.7 cm。肝硬化VNT患者以及非VNT患者的SS测量图像见图 23

图 1 脾厚度、SS以及二者联合后的预测概率的ROC曲线 Fig.1 ROC curve of spleen thickness, SS, and their combined predicted probability

图 2 肝硬化合并VNT的脾弹性图像(平均杨氏模量47.0 kPa) Fig.2 Spleen elasticity image of liver cirrhosis with VNT (mean Young's modulus 47.0 kPa)

图 3 肝硬化不合并VNT的脾弹性图像(平均杨氏模量27.8 kPa) Fig.3 Spleen elasticity image of liver cirrhosis without VNT (mean Young's modulus 27.8 kPa)

3 讨论

肝脾弹性是预测GOV/VNT的良好工具,且SS的预测能力优于LS [5-7]。研究[5]显示,应用瞬时弹性成像技术(transient elastography,TE) 诊断VNT时,TE-SS和TE-LS的AUC分别为0.989以及0.955,在诊断GOV破裂出血时,TE-SS和TE-LS的AUC分别为0.923和0.860,表明SS对VNT的诊断较LS更准确。本研究发现,单独SWE-SS对VNT诊断时AUC为0.901,诊断效能良好。

SS用于VNT诊断具有许多优势。研究[6]发现,门静脉高压肝脏急性炎症及ALT、AST上升时,弹性成像测量的LS值增加,SS值未增加,因此,当患者处于炎症急性期或ALT异常时测量的LS不能真实地反映此时的肝硬化程度,但SS却不受肝脏炎症的影响;该研究还发现SS与任何程度的门静脉高压均有相关性,证明可以通过SS对VNT进行预测。肝硬化常伴随肝脂肪变性,但SS测量完全不受影响。肝硬化进展时,肝右叶不同程度缩小,而脾脏常因充血等原因增大,更便于测量。因此,本研究通过SWE测量SS评估VNT风险,同时收集患者常规检查血清学指标,联合诊断VNT。logistic分析结果发现,脾厚度与SS有显著的鉴别能力,二者可用于预测VNT。得到的logistic模型有统计学意义,能正确分类87.5%的研究对象,灵敏度为78.3%,特异度为92.7%,阳性预测值为85.7%,阴性预测值为88.4%。本研究中,APRI、FIB-4血清学模型及ALT、AST等血清学指标对VNT预测能力不佳,可能与炎症的不同阶段、药物治疗的不同及无法避免的伴发疾病导致的血清学模型准确性降低有关。

本研究发现SS也具有某些局限性,如SS测量失败率高,本研究中SS测量的总体失败率为9.8%,文献[8]报道SS测量失败率甚至近30%,原因包括部分受检者脾脏解剖位置过高,受肺气干扰图像欠佳;脾切除患者无法获得SS值;门静脉高压达一定阈值时,部分患者除形成GOV,还可有胃肾分流、脾肾分流、脐静脉开放等自发性分流降低门静脉压力[9],影响弹性测量以及评估等。

PATERNOSTRO等[10]对弹性成像技术测量SS预测GOV/VNT的大量实验进行了对比和分析,结果显示,单独TE-SS对GOV诊断的界值分布于21.4~55 kPa之间,单独TE-SS对VNT诊断的界值分布于41.3~54 kPa之间;不同研究单独使用点剪切波弹性成像(point shear wave elastography,pSWE) 技术测量的SS (pSWE-SS) 对GOV诊断的界值分布于2.55~3.17 m/s之间,单独pSWE-SS对VNT诊断的界值分布于2.55~4.13 m/s之间;不同实验单独SWE-SS对GOV和VNT诊断的界值分别为30.3 kPa和25.6 kPa。本研究中,单独SWE-SS对VNT的诊断界值为31.4 kPa,与其他研究结果有一定差异。因测量技术出现的时间先后问题,文献数量也分别是TE多于pSWE、SWE。由此可见,不同地区、不同人群中测量获得的诊断界值差异明显,测量技术(如TE、SWE等) 不同,收集的病例病因(如病毒性肝炎、酒精性肝硬化等) 不同,实验目的(如计算阳性或阴性预测值等) 不同,均导致众多研究的最终结果差异过大,无法直接对比。本研究中,单独SWE-SS或脾厚度诊断VNT的灵敏度分别为100%、78.3%,特异度分别为73.2%、87.8%,最佳截断值分别为31.4 kPa、4.7 cm。二者联合后灵敏度(100%) 及特异度(75.6%) 均有所改善。因此,在临床工作中仅凭SS一个指标诊断GOV及VNT并不严谨,需进一步联合其他指标以提高诊断的准确性。

综上所述,SWE-SS是一种较好的无创诊断VNT的方法。本研究的不足之处在于样本量不够大,在现有的基于SWE的研究中,对测量的标准和规范未达到共识,因此未来需进一步深入研究。对于具体的GOV/VNT的诊断,也需要后续扩大样本量进行研究,以完善截断值的准确性。

参考文献
[1]
DE FRANCHIS R, FACULTY BV. Expanding consensus in portal hypertension: report of the BavenoⅥ Consensus Workshop: stratifying risk and individualizing care for portal hypertension[J]. J Hepatol, 2015, 63(3): 743-752. DOI:10.1016/j.jhep.2015.05.022
[2]
KIM BK, HAN KH, PARK JY, et al. A liver stiffness measurement-based, noninvasive prediction model for high-risk esophageal varices in B-viral liver cirrhosis[J]. Am J Gastroenterol, 2010, 105(6): 1382-1390. DOI:10.1038/ajg.2009.750
[3]
REIBERGER T, PÜSPÖK A, SCHODER M, et al. Austrian consensus guidelines on the management and treatment of portal hypertension (BillrothⅢ)[J]. Wien Klin Wochenschr, 2017, 129(Suppl 3): 135-158. DOI:10.1007/s00508-017-1262-3
[4]
王贵强, 王福生, 庄辉, 等. 慢性乙型肝炎防治指南(2019年版)[J]. 中国医学前沿杂志(电子版), 2019, 11(12): 51-77.
[5]
WANG XK, WANG P, ZHANG Y, et al. A study on spleen transient elastography in predicting the degree of esophageal varices and bleeding[J]. Medicine, 2019, 98(9): e14615. DOI:10.1097/MD.0000000000014615
[6]
MEISTER P, DECHÊNE A, BÜCHTER M, et al. Spleen stiffness differentiates between acute and chronic liver damage and predicts hepatic decompensation[J]. J Clin Gastroenterol, 2019, 53(6): 457-463. DOI:10.1097/MCG.0000000000001044
[7]
MA XW, WANG L, WU H, et al. Spleen stiffness is superior to liver stiffness for predicting esophageal varices in chronic liver disease: a meta-analysis[J]. PLoS One, 2016, 11(11): e0165786. DOI:10.1371/journal.pone.0165786
[8]
SINGH S, EATON JE, MURAD MH, et al. Accuracy of spleen stiffness measurement in detection of esophageal varices in patients with chronic liver disease: systematic review and meta-analysis[J]. Clin Gastroenterol Hepatol, 2014, 12(6): 935-945. e4. DOI:10.1016/j.cgh.2013.09.013
[9]
LIPINSKI M, SABOROWSKI M, HEIDRICH B, et al. Clinical characteristics of patients with liver cirrhosis and spontaneous portosystemic shunts detected by ultrasound in a tertiary care and transplantation centre[J]. Scand J Gastroenterol, 2018, 53(9): 1107-1113. DOI:10.1080/00365521.2018.1498913
[10]
PATERNOSTRO R, REIBERGER T, BUCSICS T. Elastography- based screening for esophageal varices in patients with advanced chronic liver disease[J]. World J Gastroenterol, 2019, 25(3): 308-329. DOI:10.3748/wjg.v25.i3.308