胰腺星状细胞活跃度在胰十二指肠切除术后胰瘘中的预测价值

曹昕彤1,2,朱帅1,2,罗庚求3,林嘉晏1,宁彩虹1,李嘉荣1,申鼎成1,王曦滔1,黄耿文1,2

(中南大学湘雅医院 1.普通外科 2.胰腺外科 3.病理科,湖南 长沙 410008)

摘 要 背景与目的:胰腺质地是胰十二指肠切除(PD)术后胰瘘(POPF)发生的重要影响因素,但胰腺质地的判断缺乏客观评价标准。研究发现,胰腺星状细胞(PSC)的活化与胰腺的纤维化的发生密切相关,因而其活性可能影响胰腺质地。因此,本研究探讨PSC 活跃度预测PD 术后临床相关性胰瘘(CR-POPF)的可行性与有效性。

方法:前瞻性收集2017年12月—2019年9月中南大学湘雅医院连续收治的101例行PD 术患者的切缘处胰腺病理标本。通过免疫组化染色的α- 平滑肌肌动蛋白(α-SMA)检测PSC 的活跃度,并进行分级。分析PSC 活跃度与胰腺质地及CR-POPF 发生的关系,以及其他相关临床因素与CR-POPF 发生的关系,用ROC 曲线评价PSC 活跃度预测CR-POPF 的效能。

结果:101例患者中,41例(40.6%)出现CR-POPF。分析结果显示,随着PSC 活跃度等级的升高,CR-POPF 发生率呈明显递减趋势,不同的胰腺质地之间,PSC 等级的分布具有统计学差异(均P<0.001)。相关性分析结果显示,PSC 活跃度与胰腺质地的硬度之间存在明显正相关性(r=0.456,P<0.001),而与CR-POPF 发生率之间则存在明显负相关性(r=-0.539,P<0.001)。单因素分析结果显示,胰腺质地、胰腺病理、PSC 活跃度分级、术前体质量指数、胰管直径、术前总胆红素、第1 天腹腔引流液淀粉酶与CR-POPF 的发生密切相关(均P<0.05);多元回归分析结果显示,PSC 活跃度分级(OR=0.24,95% CI=0.10~0.56,P<0.001)和术前总胆红素(OR=1.01, 95% CI=1.00~1.01,P=0.008)是CR-POPF 的独立危险因素。ROC 曲线分析显示,PSC 活跃度预测CR-POPF 的AUC 为0.795(95% CI=0.708~0.881),敏感度和特异度分别为63.3% 和87.8%。

结论:PSC 活跃度等级能较客观、准确地反映胰腺的质地情况,是预测PD 术后CR-POPF 的有效指标,具有一定的临床应用价值。

关键词 胰十二指肠切除术;胰腺瘘;胰腺星形细胞

随着外科技术的提高以及围术期管理的日趋完善,胰十二指肠切除术(pancreaticoduodenectomy,PD)的病死率已降至5%以下,但术后胰瘘(postoperative pancreatic fistula,POPF)的发生率仍高达20%~40%[1-4],严重影响患者的术后恢复,甚至影响远期生存率。研究[5]表明,质软胰腺是发生POPF的重要的危险因素之一。然而,胰腺质地的判断往往凭借医生的主观感受,缺乏客观的评价依据。胰腺星状细胞(pancreatic stellate cells,PSC)与胰腺的纤维化密切相关[6-7]。PSC的活跃度能否用来预测POPF,目前尚无研究报道。本研究前瞻性收集中南大学湘雅医院2017年 12月—2019年9月间101例PD术患者资料,通过分析胰腺切缘组织中PSC的活跃度,探讨PSC的活跃度在PD术后POPF中的预测价值。

1 资料与方法

1.1 一般资料

前瞻性收集2017年12月—2019年9月间中南大学湘雅医院连续收治的101例行PD术的患者,其中男55例(54.5%),女46例(45.5%);年 龄27 ~81岁,中位年龄56 岁;开腹手术96例(95.0%),腹腔镜手术5例(5.0%);经典PD 95例(94.0%),保留幽门的PD(pyloruspreserving PD,PPPD)6例(6.0%);病理类型包括十二指肠乳头癌36例(35.6%),胰腺癌 28例(27.6%),胆总管下段癌11例(10.8%),胰腺囊性肿瘤7例(7.0%),壶腹癌6例(6.0%),慢性胰腺炎6例(6.0%),十二指肠或胰腺间叶源性肿瘤5例(5.0%),胰腺神经内分泌肿瘤1例(1.0%),胆总管炎性狭窄1例(1.0%)。

1.2 染色方法及PSC 活跃度分级

取PD 手术中胰腺颈部切缘处胰腺组织行α-平滑肌肌动蛋白(α-smooth muscle actin,α-SMA)免疫组化染色。取该处胰腺组织石蜡块予以切片、脱蜡后,将其置于柠檬酸钠缓冲溶液(0.01 mol/L,pH6.0)中高压环境(125 ℃, 131 kPa)约15 min以修复抗原。3%H2O2溶液消除内源性过氧化物酶的活性。一抗(鼠抗人单克隆平滑肌肌动蛋白抗体1A4,稀释浓度1∶50),湿盒持续4 ℃孵育过夜。二抗(羊抗鼠单克隆抗体)湿盒孵育,在4 ℃下放置约20~30 min,二氨基联苯胺和苏木精分别进行显色和复染,完成染色。

参考Tanaka等[8]的分级方法,由胰腺病理专家盲法阅片,将α-SMA 免疫组化染色强度分为阴性(-)、弱阳性(+)、中阳性(++)和强阳性(+++)4个等级,分别对应0级(S0)、1级(S1)、2级(S2)、3级(S3)对PSC活跃度进行量化,其具体评判细则如表1所示。

表1 PSC 活跃度的分级标准
Table 1 The grading criteria for PSC activity

PSC 活跃等级 评判细则S0 除胰管周围组织外,无胰腺组织染色,胰腺结构正常S1 一般阳性染色,分布不规则,呈小块斑片状,无明显胰腺结构改变S2 中度阳性染色,分布不均匀,主要位于胰腺小叶内及小叶间,伴腺泡细胞轻度萎缩S3 强阳性染色,分布均匀而弥漫,伴明显腺泡细胞萎缩、小叶结构广泛破坏

本研究中所有的临床数据、标本的收集均获得患者的知情同意,并签署书面同意书。本研究获得中南大学湘雅医院医学伦理委员会批准进行(项目编号:201707777)。

1.3 围术期管理

术后常规预防性使用抗生素以及营养支持等治疗。PD患者术后的腹腔引流管均采用延迟性拔管策略(留置时间≥7 d),并常规于术后第1、3、7天动态监测腹腔引流液淀粉酶(drainage fluid amylase,DFA)情况。对于术后7 d或以后的DFA处于正常范围,且引流管引流量<10 m L/d 的患者,考虑拔除腹腔引流管。所有患者均采取快速康复外科模式促进患者早期恢复[9]

1.4 POPF 的定义及诊断标准

临床相关性胰瘘(clinically relevant POPF,C R-POPF)的诊断标准参照2016年版胰腺外科国际研究小组[10](International Study Group of Pancreatic Surgery,ISGPS)的定义:术后>3 d时,DFA大于血清淀粉酶正常值上限的3倍,且与临床治疗及预后相关。在此前提下,仅包括B级和C级CR-POPF,而生化漏(biochemical leakage)不包括在内。生化漏:仅术后第3天或以后腹腔引流液淀粉酶升高达正常值上限3倍,而对临床结局无任何影响。B级CR-POPF:(1) 腹腔引流管留置时间>3周;(2) POPF相关性临床治疗方案变更(加用生长抑素、抗生素升级、肠内外营养支持、输血等);(3) 需经皮或内镜下穿刺引流的POPF继发性胰周积液;(4) 需血管造影介入止血的POPF相关性出血;(5) POPF继发性感染,但无脏器功能衰竭。满足以上5种情况中的任意1种。C级CRPOPF:(1) 需再次开放手术处理的POPF相关并发症(如假性动脉瘤破裂出血、腹腔脓肿形成、吻合口破裂等);(2) POPF继发性器官功能衰竭;(3) POPF相关性死亡。满足以上3种情况中的任意 1种。

1.5 统计学处理

计数资料以例数(百分比)[n(%)]表示,其比较采用χ2检验;计量资料中符合正态分布的数据以均数±标准差(±s)表示,其比较采用 t 检验;符合偏态分布的数据以中位数(M)表示,其比较采用Mann-Whitney U检验。多因素分析采用Logistic回归模型,并应用受试者操作特性曲线(receiver operating characteristic curve,ROC)进行分析。以上均通过SPSS 22.0统计软件分析,P<0.05为差异有统计学意义。

2 结 果

2.1 手术结果

全组101例患者中,41例术后出现CR-POPF,CR-POPF发生率为40.6%,包括B级36例,C 级 5例。无CR-POPF者共60例,包括19例生化漏和41例无胰瘘者。中位手术时长与术中出血量分别为340 min和400 mL。术后总的并发症率为46.5%(47/101),除C R-POPF 外,包括术后胆汁漏 7例(7.0%),延迟性胃排空障碍4例(4.0%),术后出血(包括消化道出血和腹腔内出血)16例(15.8%)和肺部并发症(包括胸腔积液、肺部感染、肺不张及呼吸功能衰竭等)46例(45.5%)。术后中位住院时长为15 d,再入院率为6.0%(6/101),术后90 d内病死率为4.0%(4/101)。

2.2 PSC 活跃度分级

本研究中,切缘处胰腺免疫组化染色的镜下表现及分级如图1 所示。依照评判规则,全组被分级为S0、S1、S2、S3的病例分别为9、49、24、19例,各活跃度等级中的CR-POPF率分别为88.9%(8/9)、57.1%(28/49)、16.7%(4/24)和5.3%(1/19) ,呈逐渐递减趋势,差异有统计学意义(P<0.001)。PSC等级与胰腺质地之间的关系如表2所示。在不同的胰腺质地之间,PSC等级的分布具有明显差异性(P<0.001)。其中,在柔软胰腺中占比最高的PSC等级为S1(60.6%),而在质硬胰腺中为S2(40.0%)。经Spearman相关性检验结果显示,PSC活跃度与胰腺质地的硬度之间存在明显正相关性(r=0.456,P<0.001),而与CR-POPF之间则存在明显负相关性(r=-0.539,P<0.001)。

图1 α-SMA 免疫组化染色所反映的PSC 的活跃度 A:S0;B:S1;C:S2;D:S3
Figure 1 The PSC activity shown by immunohistochemical staining of α-SMA A: S0; B: S1; C: S2; D: S3

表2 PSC 等级与胰腺质地之间的关系[n(%)]
Table 2 Relationship between the PSC activity grade and pancreatic texture [n (%)]

PSC 等级 胰腺质地软硬S0 9(13.6) 0(0.0)S1 40(60.6) 9(25.7)S2 10(15.2) 14(40.0)S3 7(10.6) 12(34.3)

2.3 临床病理因素与CR-POPF 关系的单因素和多元回归分析

单因素分析显示,胰腺质地、肿块病理、PSC活跃度分级、体质量指数(BMI)、胰管直径、术前总胆红素、术后第1 天引流液中淀粉酶含量(DFA1)均与CR-POPF的发生有关(均P<0.05)(表3)。多元回归分析显示,仅PSC活跃度分级(OR=0.24,95% CI=0.10~0.56,P<0.001)和术前总胆红素水平(OR=1.01,95% CI=1.00~1.01,P=0.008)是预测CR-POPF的独立危险因素(表4)。

2.4 PSC 活跃度分级对CR-POPF 的预测价值

PSC活跃度分级预测CR-POPF的ROC曲线如图2 所示。当截断值取1.5 时,其曲线下面积(area under curve,AUC)为0.795(95% CI=0.708~0.881),相应的敏感度和特异度分别为63.3%和87.8%,提示其对CR-POPF具有良好的预测价值。

表3 临床病理因素与CR-POPF 关系的单因素分析
Table 3 Univariate analysis of relationship between clinicopathologic factors and CR-POPF

因素 CR-POPF 组(n=41)无CR-POPF 组(n=60) P 因素 CR-POPF 组(n=41)无CR-POPF 组(n=60) P性别[n(%)] 胰肠吻合方式[n(%)] 男 22(53.7) 33(55.0) 1.000 端-端吻合 36(87.8) 57(95.0) 0.347 女 19(46.3) 27(45.0) 端-侧吻合 5(12.2) 3(5.0)年龄(岁,images/BZ_33_1708_490_1742_536.png±s) 59.4±10.0 56.1±10.1 0.108 门静脉/肠系膜血管切除重建[n(%)]术前BMI(kg/m2,images/BZ_33_1708_490_1742_536.png±s) 22.77±2.24 21.53±2.65 0.016 有 2(4.9) 3(5.0) 1.000胰管直径(mm,images/BZ_33_1708_490_1742_536.png±s) 2.98±2.01 4.76±2.82 <0.001 无 39(95.1) 57(95.0)术中失血量(mL,images/BZ_33_1708_490_1742_536.png±s) 495.6±235.8 472.5±303.0 0.682 预防性使用生长抑[n(%)]手术时长(min,images/BZ_33_1708_490_1742_536.png±s) 364.0±107.8 367.5±122.1 0.884 是 27(65.9) 42(70.0) 0.670术前总胆红素 [μmol/L,M(范围)] 120.9(4.3~305.2) 40.8(5.4~396.3) 0.031 否 14(34.1) 18(30.0)胰腺质地[n(%)]术前血清白蛋白(g/L,images/BZ_33_1708_490_1742_536.png±s) 38.9±4.4 38.8±4.9 0.958 软 36(87.8) 30(50.0) <0.001术后血清白蛋白(g/L,images/BZ_33_1708_490_1742_536.png±s) 32.0±5.6 33.6±4.6 0.118 硬 5(12.2) 30(50.0)DFA1 [U/L,M(范围)] 5660.6(32.6~29139.7) 994.6(4.7~25601.5) 0.002 肿块病理[n(%)]ASA 分级[n(%)] 胰腺癌 7(17.1) 21(35.0)0.018 I~II 20(48.8) 37(61.7) 0.225 壶腹癌 2(4.9) 4(6.7) III~IV 21(51.2) 23(38.3) 胆总管下段癌 10(24.4) 1(1.7)术前糖尿病[n(%)] 十二指肠乳头癌 16(39.0) 20(33.3) 有 7(17.1) 12(20.0) 0.799 慢性胰腺炎 1(2.4) 5(8.3) 无 34(82.9) 48(80.0) 胰腺囊性肿瘤 2(4.9) 5(8.3)术前吸烟史[n(%)] 其他 3(7.3) 4(6.7) 有 12(29.3) 20(33.3) 0.828 PSC 活跃度分级[n(%)] 无 29(70.7) 40(66.7) S0 8(19.5) 1(1.7)腹部手术史[n(%)] S1 28(68.3) 21(35.0) <0.001 有 9(22.0) 14(23.3) 1.000 S2 4(9.8) 20(33.3) 无 32(78.0) 46(76.7) S3 1(2.4) 18(30.0)手术方式[n(%)] PD 36(87.8) 59(98.3) 0.077 PPPD 5(12.2) 1(1.7)

表4 临床病理因素与CR-POPF 关系多因素Logistic 回归分析
Table 4 Multivariate Logistic regression analysis of relationship between clinicopathologic factors and CR-POPF

因素 OR 95% CI PBMI 1.15 0.91~1.45 0.224胰腺质地(软) 2.39 0.55~9.28 0.245肿块病理 1.88 0.44~8.08 0.357胰管直径 0.98 0.75~1.28 0.428术前总胆红素 1.01 1.00~1.01 0.008 DFA1 1.00 1.00~1.00 0.441 PSC 活跃度分级 0.24 0.10~0.56 <0.001

3 讨 论

图2 PSC 等级预测PD 术后CR-POPF 的ROC 曲线
Figure 2 ROC curve of PSC activity for predicting CR-POPF following PD

POPF是PD术后的严重并发症之一,可继发引起术后出血、腹腔感染、延迟性胃排空障碍等一系列并发症,甚至导致患者死亡[11-15]。胰腺质地与POPF的发生密切相关[5,16]。然而,仅靠外科医生的触感判断胰腺质地较为主观。因此,国内外均有研究采用其它较为客观的方法来评判胰腺质地,如MRI、胰腺弹力仪等,取得一定效果,但实用性不高[17-21]。PSC作为胰腺实质中的基质细胞,活化后可分泌大量纤维组织,从而参与胰腺组织的纤维化,理论上可能与胰腺质地变硬密切相关[6-7,21-22]。本研究采用免疫组化检测胰腺切缘组织中的α-SMA蛋白反映PSC的活跃度,前瞻性研究101例PD患者的临床资料,发现PSC活跃度是预测CR-POPF的独立危险因素(OR=0.24,95% CI=0.10~0.56,P<0.001)。此外,ROC曲线分析显示PSC活跃度的预测效能良好(AUC=0.795),从而进一步证实 PSC活跃度是预测PD术后胰瘘的重要的客观指标。

PSC是一种外形类似星形的胰腺基质细胞,最初于1998年被发现并分离培养[23-24]。它存在于胰腺腺泡细胞、微血管及微胰管周围。在大多数情况下,PSC处于静止态,具有免疫调节、吞噬、储存脂质、维持胰腺组织基础内外分泌以及其基本结构等功能[25-26]。然而,在缺氧、酒精、吸烟等恶劣条件刺激下,PSC可转为活化态并获得成纤维细胞样表型[27-30],分泌大量细胞因子(如TNF-α,TGF-β、PDGF等)以及细胞外基质,促进胰腺组织的纤维化进程[7,31]。较为特殊的是,慢性胰腺炎的胰腺组织中有大量单核细胞浸润,后者通过产生TNF-α促进PSC活化[32];胰腺癌细胞与PSC存在纤维化/缺氧循环机制[33],通过正反馈放大作用致使组织持续性的缺氧、缺血、纤维化。以上可能是PSC介导胰腺纤维化的主要机制。然而,其更具体的分子信号通路仍有待于进一步探究。

从组织病理学角度来看,胰腺质地取决于组织中的纤维化程度,它与胰腺的外分泌功能呈负相关性[22,34]。然而究其根本,胰腺纤维组织的蓄积是PSC活跃的结果,故后者更能反映胰腺质地变化的本质。遗憾的是,目前的研究主要集中于探索胰腺纤维化与胰瘘的联系[35-37],而鲜有关于PSC活跃度和POPF关联性的研究。Erkan等[38]认为胰腺癌中的PSC是导致胰腺纤维化的主要原因,可影响外科医生对质地的判断。Tanaka等[8]通过将不同活跃程度的PSC进行分级并对CR-POPF进行预测,率先证实PSC等级在发生POPF与无POPF的患者之间存在统计学差异(P=0.035)。在此基础上,本研究发现随着PSC活跃度提高,CR-POPF率明显降低,这与前述的理论基本契合。而且,在本研究中,PSC活跃度被证实是发生CR-POPF的独立危险因素,而胰腺质地并不是独立的危险因素,从而证实了PSC活跃度在预测CR-POPF方面更具有客观性与优越性。因此,PSC活跃度可代替胰腺质地,作为评估CR-POPF风险的全新预测指标。

本研究存在一定的局限性。首先,本研究所涵盖的样本量不大,且为单中心研究;其次,PSC活跃时分泌或活化的蛋白种类繁多,其特征性蛋白除α-SMA外[39],还包括成纤维细胞活化蛋白-α,潜在转化生长因子结合蛋白等[40-41]。本研究中的标记蛋白较为单一,其客观度稍显不足,有待于在今后的实验中予以完善。

综上所述,PSC的活跃度对CR-POPF的预测具有重要意义,可作为临床指导预测CR-POPF发生的重要参考指标之一。

参考文献

[1] Kawakatsu S, Inoue Y, Mise Y, et al.Comparison of pancreatojejunostomy techniques in patients with a soft pancreas: Kakita anastomosis and Blumgart anastomosis[J].BMC Surg, 2018, 18(1):88.doi: 10.1186/s12893-018-0420-5.

[2] Ecker BL, McMillan MT, Asbun HJ, et al.Characterization and optimal management of high-risk pancreatic anastomoses during pancreatoduodenectomy[J].Ann Surg, 2018, 267(4):608-616.doi: 10.1097/SLA.0000000000002327.

[3] Allen PJ, Gönen M, Brennan MF, et al.Pasireotide for postoperative pancreatic fistula[J].N Engl J Med, 2014, 370(21):2014-2022.doi: 10.1056/NEJMoa1313688.

[4] Ellis RJ, Brock Hewitt D, Liu JB, et al.Preoperative risk evaluation for pancreatic fistula after pancreaticoduodenectomy[J].J Surg Oncol, 2019, 119(8): 1128-1134.doi: 10.1002/jso.25464.

[5] Hamanaka Y, Nishihara K, Hamasaki T, et al.Pancreatic juice output after pancreatoduodenectomy in relation to pancreatic consistency, duct size, and leakage[J].Surgery, 1996, 119(3):281-287.doi: 10.1016/s0039-6060(96)80114-0.

[6] Masamune A, Watanabe T, Kikuta K, et al.Roles of pancreatic stellate cells in pancreatiCInflammation and fibrosis[J].Clin Gastroenterol Hepatol, 2009, 7(11):S48-54.doi: 10.1016/j.cgh.2009.07.038.

[7] Bachem MG, Schünemann M, Ramadani M, et al.Pancreatic carcinoma cells induce fibrosis by stimulating proliferation and matrix synthesis of stellate cells[J].Gastroenterology, 2005, 128(4):907-921.doi: 10.1053/j.gastro.2004.12.036.

[8] Tanaka K, Tomita H, Osada S, et al.Significance of histopathological evaluation of pancreatic fibrosis to predict postoperative course after pancreatic surgery[J].Anticancer Res, 2015, 35(3):1749-1756.

[9] Lassen K, Coolsen MM, Slim K, et al.Guidelines for perioperative care for pancreaticoduodenectomy: enhanced recovery after surgery (ERAS) society recommendations[J].Clin Nutr, 2012, 31(6):817-830.doi: 10.1016/j.clnu.2012.08.011.

[10] Bassi C, Marchegiani G, Dervenis C, et al.The 2016 update of the International Study Group (ISGPS) definition and grading of postoperative pancreatic fistula: 11 Years After[J].Surgery, 2017, 161(3):584-591.doi: 10.1016/j.surg.2016.11.014.

[11] El Nakeeb A, El Sorogy M, Hamed H, et al.Biliary leakage following pancreaticoduodenectomy: prevalence, risk factors and management[J].Hepatobiliary Pancreat Dis Int, 2019, 18(1):67-72.doi: 10.1016/j.hbpd.2018.10.005.

[12] 郑岩松, 石铮, 翁山耕, 等.胰十二指肠切除术后腹腔内迟发性出血的原因与处理[J].中国普通外科杂志, 2011, 20(6):626-628.

Zheng YS, Shi Z, Weng SG, et al.The causes and treatment of delayed intra-abdominal hemorrhage after pancreatoduodenectomy[J].Chinese Journal of General Surgery, 2011, 20(6):626-628.

[13] 刘国华, 谭小宇, 戴东, 等.改良Blumgart胰肠吻合在胰十二指肠切除术中的应用[J].中国普通外科杂志, 2020, 29(3):276-283.doi:10.7659/j.issn.1005-6947.2020.03.004.

Liu GH, Tan XY, Dai D, et al.Application of modified Blumgart pancreaticojejunostomy in pancreaticoduodenectomy[J].Chinese Journal of General Surgery, 2020, 29(3):276-283.doi:10.7659/j.issn.1005-6947.2020.03.004.

[14] Nagakawa Y, Matsudo T, Hijikata Y, et al.Bacterial contamination in ascitic fluid is associated with the development of clinically relevant pancreatic fistula after pancreatoduodenectomy[J].Pancreas, 2013, 42(4):701-706.doi: 10.1097/MPA.0b013e31826d3a41.

[15] Williamsson C, Ansari D, Andersson R, et al.Postoperative pancreatic fistula-impact on outcome, hospital cost and effects of centralization[J].HPB (Oxford), 2017, 19(5):436-442.doi: 10.1016/j.hpb.2017.01.004.

[16] Sugimoto M, Takahashi S, Kojima M, et al.In patients with a soft pancreas, a thick parenchyma, a small duct, and fatty infiltration are significant risks for pancreatic fistula after pancreaticoduodenectomy[J].J Gastrointest Surg, 2017, 21(5):846-854.doi: 10.1007/s11605-017-3356-7.

[17] Maehira H, Iida H, Mori H, et al.Computed tomography enhancement pattern of the pancreatic parenchymAPredicts postoperative pancreatic fistula after pancreaticoduodenectomy[J].Pancreas, 2019, 48(2):209-215.doi: 10.1097/MPA.0000000000001229.

[18] Watanabe H, Kanematsu M, Tanaka K, et al.Fibrosis and postoperative fistula of the pancreas: correlation with MR imaging findings--preliminary results[J].Radiology, 2014, 270(3):791-799.doi: 10.1148/radiol.13131194.

[19] Hong TH, Choi JI, Park MY, et al.Pancreatic hardness: correlation of surgeon's palpation, durometer measurement and preoperative magnetic resonance imaging features[J].World J Gastroenterol, 2017, 23(11):2044-2051.doi: 10.3748/wjg.v23.i11.2044.

[20] Shi Y, Liu Y, Gao F, et al.Pancreatic stiffness quantified with MR elastography: relationship to postoperative pancreatic fistula after pancreaticoenteric anastomosis[J].Radiology, 2018, 288(2):476-484.doi: 10.1148/radiol.2018170450.

[21] Marchegiani G, Ballarin R, Malleo G, et al.Quantitative assessment of pancreatic texture using a durometer: A new tool to predict the risk of developing APostoperative fistula[J].World J Surg, 2017, 41(11):2876-2883.doi: 10.1007/s00268-017-4073-9.

[22] Pereira FL, Vasques FT, Moricz Ad, et al.Correlation analysis between post-pancreatoduodenectomy pancreatic fistula and pancreatic histology[J].Rev Col Bras Cir, 2012, 39(1):41-47.

[23] Apte MV, Haber PS, Applegate TL, et al.Periacinar stellate shaped cells in rat pancreas: identification, isolation, and culture[J].Gut, 1998, 43(1):128-133.doi: 10.1136/gut.43.1.128.

[24] Bachem MG, Schneider E, Gross H, et al.Identification, culture, and characterization of pancreatic stellate cells in rats and humans[J].Gastroenterology, 1998, 115(2):421-432.doi: 10.1016/s0016-5085(98)70209-4.

[25] Nielsen MFB, Mortensen MB, Detlefsen S, et al.Identification of markers for quiescent pancreatic stellate cells in the normal human pancreas[J].Histochem Cell Biol, 2017, 148(4):359-380.doi: 10.1007/s00418-017-1581-5.

[26] Ferdek PE, Jakubowska MA.Biology of pancreatic stellate cells-more than just pancreatic cancer[J].Pflugers Arch, 2017, 469(9):1039-1050.doi: 10.1007/s00424-017-1968-0.

[27] Rebours V, Albuquerque M, Sauvanet A, et al.HypoxiAPathways and cellular stress activate pancreatic stellate cells: development of an organotypic culture model of thick slices of normal human pancreas[J].PLoS One, 2013, 8(9): e76229.doi: 10.1371/journal.pone.0076229.

[28] Haber PS, Keogh GW, Apte MV, et al.Activation of pancreatic stellate cells in human and experimental pancreatic fibrosis[J].Am J Pathol, 1999, 155(4):1087-1095.doi: 10.1016/S0002-9440(10)65211-X.

[29] Hashimoto A, Karim MR, Izawa T, Kuwamura M, Yamate J.Immunophenotypical analysis of pancreatiCInterstitial cells in the developing rat pancreas and myofibroblasts in the fibrotic pancreas in dogs and cats[J].J Vet Med Sci, 2017, 79(12):1920-1926.doi: 10.1292/jvms.17-0423.

[30] Öhlund D, Handly-Santana A, Biffi G, et al.Distinct populations of inflammatory fibroblasts and myofibroblasts in pancreatic cancer[J].J Exp Med, 2017, 214(3):579-596.doi: 10.1084/jem.20162024.

[31] Laklai H, Miroshnikova YA, Pickup MW, et al.Genotype tunes pancreatic ductal adenocarcinoma tissue tension to induce matricellular fibrosis and tumor progression[J].Nat Med, 2016, 22(5):497-505.doi: 10.1038/nm.4082.

[32] Michalski CW, Gorbachevski A, Erkan M, et al.Mononuclear cells modulate the activity of pancreatic stellate cells which in turn promote fibrosis and inflammation in chronic pancreatitis[J].J Transl Med, 2007, 5:63.doi: 10.1186/1479-5876-5-63.

[33] Erkan M, Reiser-Erkan C, Michalski CW, et al.Cancer-stellate cell interactions perpetuate the hypoxia-fibrosis cycle in pancreatic ductal adenocarcinoma[J].Neoplasia, 2009, 11(5):497-508.doi: 10.1593/neo.81618.

[34] Deng Y, Zhao B, Yang M, et al.Association between the incidence of pancreatic fistula after pancreaticoduodenectomy and the degree of pancreatic fibrosis[J].J Gastrointest Surg, 2018, 22(3):438-443.doi: 10.1007/s11605-017-3660-2.

[35] Umezaki N, Hashimoto D, Nakagawa S, et al.Number of acinar cells at the pancreatic stump predicts pancreatic fistula after pancreaticoduodenectomy[J].Surg Today, 2018, 48(8):790-795.doi: 10.1007/s00595-018-1656-5.

[36] Nahm CB, Lui I, Naidoo CS, et al.Density and enhancement of the pancreatic tail on computer tomography predicts acinar score and pancreaticfistula after pancreatoduodenectomy[J].HPB (Oxford), 2019, 21(5):604-611.doi: 10.1016/j.hpb.2018.09.014.

[37] Kang JH, Park JS, Yu JS, et al.Prediction of pancreatic fistula after pancreatoduodenectomy by preoperative dynamic CT and fecal elastase-1 levels[J].PLoS One, 2017, 12(5):e0177052.doi: 10.1371/journal.pone.0177052.

[38] Erkan M, Hausmann S, Michalski CW, et al.How fibrosis influences imaging and surgical decisions in pancreatic cancer[J].Front Physiol, 2012, 3:389.doi: 10.3389/fphys.2012.00389.

[39] Fujita H, Ohuchida K, Mizumoto K, et al.Alpha-smooth muscle actin expressing stromAPromotes an aggressive tumor biology in pancreatic ductal adenocarcinoma[J].Pancreas, 2010, 39(8):1254-1262.doi: 10.1097/MPA.0b013e3181dbf647.

[40] Fu Y, Liu S, Zeng S, et al.The critical roles of activated stellate cells-mediated paracrine signaling, metabolism and oncoimmunology in pancreatic ductal adenocarcinoma[J].Mol Cancer, 2018, 17(1):62.doi: 10.1186/s12943-018-0815-z.

[41] Sarper M, Cortes E, Lieberthal TJ, et al.ATRA modulates mechanical activation of TGF-β by pancreatic stellate cells[J].Sci Rep, 2016, 6:27639.doi: 10.1038/srep27639.

Predictive value of pancreatic stellate cell activity for postoperative pancreatic fistula after pancreaticoduodenectomy

CAO Xintong1,2, ZHU Shuai1,2, LUO Gengqiu3, LIN Jiayan1, NING Caihong1, LI Jiarong1, SHEN Dingcheng1, WANG Xitao1, HUANG Gengwen1,2

(1.Department of General Surgery 2.Department of Pancreatic Surgery 3.Department of Pathology, Xiangya Hospital, Central South University, Changsha 410008, China)

Abstract Background and Aims: The texture of the pancreas is an important factor for the occurrence of postoperative pancreatic fistula (POPF) following pancreaticoduodenectomy (PD).However, there are no objective evaluation criteria for estimating the hardness of the pancreatic texture.Studies have demonstrated that the activation of the pancreatic stellate cells (PSCs) is closely associated with the pancreatic fibrosis, and therefore, their activity may probably influence the texture of the pancreas.This study was designated to investigated the feasibility and effectiveness of using the degree of PSC activity for predicting the clinically relevant postoperative pancreatic fistula (CR-POPF) after PD.

Methods: The surgical margin samples from 101 consecutive patients who underwent PD in the Department of Pancreatic Surgery, Xiangya Hospital, Central South University from December 2017 to September 2019 were prospectively collected.The degree of PSC activity was determined and graded by immunohistochemical staining of α-smooth muscle actin (α-SMA) protein.The relations of PSC activity with CR-POPF and the pancreatic texture, as well as the relations of other relevant clinicopathologic factors with CR-POPF were analyzed.The efficiency of PSC activity in predicting CR-POPF was determined by ROC analysis.

Results: In the 101 patients, CR-POPF occurred in 41 cases (40.6%).Results of analysis showed that the incidence of CR-POPF was decreased progressively with the increase of the grade of PSC activity, and the distributions of PSC activity grades were significantly different among different pancreatic textures (both P<0.001).Results of correlation analysis showed that the degree of PSC activity was positively correlated with the hardness of pancreatic texture (r=0.456, P<0.001), while was negatively correlated with the incidence of CRPOPF (r=-0.539, P<0.001).Results of univariate analysis showed that pancreatic texture, tumor pathology, PSC activity grade, preoperative body mass index, pancreatic duct diameter, preoperative total bilirubin, drainage fluid amylase on postoperative day 1 were significantly associated with the occurrence of CR-POPF (all P<0.05), and the results of multivariate Logistic regression analysis showed that the PSC activity (OR=0.24, 95% CI=0.10-0.56, P<0.001) and preoperative total bilirubin (OR=1.01, 95% CI=1.00-1.01, P=0.008) were the independent risk factors for CR-POPF.Results of ROC analysis showed that the AUC of the degree of PSC activity for predicting CR-POPF was 0.795 (95% CI=0.708-0.881), with a sensitivity of 63.3% and a specificity of 87.8%.

Conclusion: The degree of PSC activity can objectively and accurately reflect the hardness of the pancreatic texture.It is an effective index for predicting the CR-POPF following PD, and has certain clinical application value.

Key words Pancreaticoduodenectomy; Pancreatic Fistula; Pancreatic Stellate Cells

中图分类号:R657.5

基金项目:国家自然科学基金资助项目(81802450);吴阶平医学基金会临床科研专项基金资助项目(320.6750.17518);湖南省技术创新引导计划临床医疗技术创新基金资助项目(2017SK50101);湖南省卫生健康委科研计划课题资助项目(B2019190)。

收稿日期:2020-03-08;

修订日期:2020-08-12。

作者简介:曹昕彤,中南大学湘雅医院住院医师,主要从事胰腺外科和疝外科方面的研究。

通信作者:黄耿文, Email: gengwenhuang@qq.com

doi:10.7659/j.issn.1005-6947.2020.09.002

http://dx.doi.org/10.7659/j.issn.1005-6947.2020.09.002

Chinese Journal of General Surgery, 2020, 29(9):1037-1044.

CLC number: R657.5

(本文编辑 宋涛)

本文引用格式:曹昕彤, 朱帅, 罗庚求, 等.胰腺星状细胞活跃度在胰十二指肠切除术后胰瘘中的预测价值[J].中国普通外科杂志, 2020, 29(9):1037-1044.doi:10.7659/j.issn.1005-6947.2020.09.002

Cite this article as: Cao XT, Zhu S, Luo GQ, et al.Predictive value of pancreatic stellate cell activity for postoperative pancreatic fistula after pancreaticoduodenectomy[J].Chin J Gen Surg, 2020, 29(9):1037-1044.doi:10.7659/j.issn.1005-6947.2020.09.002