李 平,朱 亮,薛華丹,劉昌義,徐 凱,李 娟,孫 婷,金征宇
中國醫(yī)學(xué)科學(xué)院 北京協(xié)和醫(yī)學(xué)院 北京協(xié)和醫(yī)院放射科,北京 100730
·Force CT專欄 論著·
胰腺局灶性實性病變CT灌注參數(shù)特征及不同算法的比較
李 平,朱 亮,薛華丹,劉昌義,徐 凱,李 娟,孫 婷,金征宇
中國醫(yī)學(xué)科學(xué)院 北京協(xié)和醫(yī)學(xué)院 北京協(xié)和醫(yī)院放射科,北京 100730
目的總結(jié)胰腺局灶性實性病變CT灌注參數(shù)的特征,評估Deconvolution和Maximum slope+Patlak方法所測得的灌注參數(shù)之間的一致性及兩種測量方法之間是否可相互替換。方法2015年12月至2016年11月在北京協(xié)和醫(yī)院行全胰腺CT灌注檢查、經(jīng)術(shù)后及穿刺病理證實為胰腺癌(PAC)患者22例和胰腺神經(jīng)內(nèi)分泌瘤(pNET)患者22例(共37個病灶),全胰腺CT灌注檢查采用管電壓80kV、管電流100mA,進(jìn)行28次連續(xù)動態(tài)體積掃描,靜脈注射45ml碘普羅胺,速率5ml/s,隨后追加40ml鹽水,速率5ml/s。由1名經(jīng)驗豐富的放射科醫(yī)師在西門子后處理工作站上分別用Maximum slope+Patlak及Deconvolution method 方法進(jìn)行數(shù)據(jù)分析,測量并記錄其灌注參數(shù)。結(jié)果Wilcoxon非參配對秩和檢驗結(jié)果顯示,PAC(BFM比BFD,Z=-3.263,P=0.001;BVD 比BVP,Z=-3.978,P=0.000)和pNET(BFM 比BFD,Z=-5.212,P=0.000;BVD比BVP,Z=-2.633,P=0.008)兩種方法所測得的灌注參數(shù)之間差異均有統(tǒng)計學(xué)意義。Spearman’s 相關(guān)系數(shù)分析結(jié)果顯示,PAC(BFM與BFD,r=0.845,P=0.000;BVD 與BVP,r=0.964,P=0.000)和pNET(BFM與BFD,r=0.759,P=0.000;BVD比BVP,r=0.683,P=0.000)兩種方法所測得的灌注參數(shù)間均有顯著相關(guān)性。PAC的BFM/BFD和BVD/BVP幾何均數(shù)及其95%一致性界限(LOA)分別為0.77(0.61~0.99)和1.42(1.13~1.79),pNET的BFM/BFD和BVD/BVP幾何均數(shù)及其95% LOA分別為0.66(0.51~0.86)和1.15(0.88~1.50)。結(jié)論無論在PAC還是pNET中由Deconvolution和Maximum slope+Patlak方法所測得的灌注參數(shù)BF、BV存在顯著性差異,兩種測量方法之間具有顯著相關(guān)性,并且兩者之間的轉(zhuǎn)換范圍較窄,所以兩種測量方法之間可以相互替換。
胰腺實性病變;CT灌注;Deconvolution;Maximum slope+Patlak
ActaAcadMedSin,2017,39(1):80-87
胰腺實性病變可以有多種原因所致,包括腫瘤、炎癥、感染等,其中較常見的乏血供病變?yōu)橐认侔谎┎∽優(yōu)橐认偕窠?jīng)內(nèi)分泌腫瘤(pancreatic neuroendocrine tumor,pNET)。胰腺癌是最常見的胰腺惡性病變,其發(fā)病率近年不斷升高,由于胰腺位置隱蔽,臨床癥狀不典型,發(fā)現(xiàn)時多已是晚期。隨著CT技術(shù)的不斷發(fā)展,胰腺病變的檢出及診斷得到了極大提高,但當(dāng)病變小于2 cm時,其檢出敏感性有所下降[1],此時形態(tài)學(xué)結(jié)合功能學(xué)檢查可能有所幫助。pNET是一種少見的胰腺腫瘤,根據(jù)有無臨床癥狀分為功能性和非功能性,根據(jù)其病理特點(diǎn)分為良性、低度惡性和高度惡性,世界衛(wèi)生組織根據(jù)核分裂數(shù)及增殖細(xì)胞核抗原Ki- 67將其分為G1、G2、G3級[2]。無論是乏血供病變還是富血供病變,腫瘤性病變預(yù)后與腫瘤內(nèi)微血管密度(tumor microvascular density,MVD)密切相關(guān),有研究指出CT灌注成像可以反映MVD[3]。CT灌注為功能性成像,是一種相對先進(jìn)的CT檢查技術(shù)[4],可將功能參數(shù)擴(kuò)展到傳統(tǒng)增強(qiáng)CT上面。CT灌注成像允許逐像素計算一系列的生理參數(shù)并形成彩色函數(shù)圖,這些參數(shù)可反映腫瘤血管生成的生理及分子過程,用于提供生物標(biāo)志物或反映腫瘤的治療反應(yīng)[5]。因此,CT灌注成像是一種理想的無創(chuàng)檢查方法,可反映和量化組織的血流動力學(xué)變化,已有研究表明其在胰腺疾病方面的應(yīng)用價值[6- 7],其在腫瘤病變的檢出、診斷、良惡性鑒別、病理診斷及分級、腫瘤性疾病對放化療反應(yīng)的評估、患者預(yù)后預(yù)測等方面有重要作用[8- 10]。CT灌注成像為定量檢查,其優(yōu)勢在于定量測量灌注參數(shù),以往研究顯示,不同研究所得的灌注參數(shù)數(shù)值之間存在明顯差異,這可能與不同研究的研究對象、灌注掃描方案、所應(yīng)用的灌注后處理軟件及灌注測量方法不同有關(guān)[8,11]。本研究總結(jié)了胰腺實性病變的CT灌注特點(diǎn),評估了Deconvolution和Maximum slope+Patlak方法所測得的灌注參數(shù)之間的一致性及兩種測量方法之間是否可相互替換,以期為今后的臨床應(yīng)用提供參考。
對象2015年12月至2016年11月在北京協(xié)和醫(yī)院行胰腺灌注CT檢查、經(jīng)術(shù)后及穿刺病理證實為胰腺癌或pNET的患者,除外胰腺的彌漫性病變。共入組胰腺癌患者22例,其中,男12例,女10例,平均年齡(57±12歲)(25~79歲);pNET患者22例(共37個病灶),其中,男11例,女11例,平均年齡(44±13)歲(10~60歲),pNET多發(fā)患者5例。本研究經(jīng)北京協(xié)和醫(yī)院倫理委員會批準(zhǔn),所有患者在檢查前均簽署了知情同意書。
容積灌注CT掃描方案采用第3代雙源CT (SOMATOM Definition Force;Siemens Healthcare,F(xiàn)orchheim,Germany)行胰腺灌注檢查,具體掃描方案如下:(1)灌注前腹部CT平掃確定胰腺灌注掃描的范圍:準(zhǔn)直192×0.6 mm,螺距 0.6,重建間隔 5 mm,管電壓、管電流自動選擇,掃描范圍為上腹部,從橫隔至雙腎下極。(2)灌注前告知患者進(jìn)行均勻淺呼吸并應(yīng)用、加壓腹帶以減少患者呼吸運(yùn)動偽影,采用DynMulti 4D模式進(jìn)行灌注掃描:管電壓80 kV,管電流100 mA,進(jìn)行28次連續(xù)動態(tài)數(shù)據(jù)采集,最后1次采集時間間隔為3 s,其余時間間隔均為1.5 s,總灌注掃描時間42 s,掃面范圍為176 mm。采用雙筒高壓注射器經(jīng)肘前靜脈注射對比劑 (碘普羅胺,370 mgI/ml,45 ml,5 ml/s) 和生理鹽水(40 ml,5 ml/s)。在對比劑注射6 s后開始行灌注掃描。
圖像分析及定量灌注評估1名經(jīng)驗豐富的放射科醫(yī)生在西門子后處理工作站(Syngo Volume Perfusion CT Body,VB10,Siemens Healthcare)上分別用Deconvolution 和Maximum slope+Patlak方法進(jìn)行數(shù)據(jù)測量。CT灌注分析軟件采用自動運(yùn)動校正及4D減噪技術(shù),自動選擇腹主動脈作為其輸入動脈,形成動脈輸入函數(shù),從而獲得時間密度曲線,進(jìn)而形成灌注偽彩圖。在增強(qiáng)CT圖像上以病變的最大層面畫取感興趣區(qū)(region of interest,ROI)。劃定ROI區(qū)域時,應(yīng)盡量保證兩種方法測量的ROI大小一致,避開血管、膽管、胰腺邊緣、胰腺導(dǎo)管及病變壞死區(qū)域。然后分別用Deconvolution 和Maximum slope+Patlak方法測量并記錄胰腺病變的灌注參數(shù)血流量(blood flow,BF)、血容量(blood volume,BV),Deconvolution方法所測得的灌注參數(shù)BF、BV用BFD、BVD表示,Maximum slope方法所測得的灌注參數(shù)BF用BFM表示,Patlak方法所測得的灌注參數(shù)BV用BVP表示。
統(tǒng)計學(xué)處理采用SPSS 21.0統(tǒng)計軟件,正態(tài)分布計量資料以均數(shù)±標(biāo)準(zhǔn)差表示,灌注參數(shù)測量值之間的差異性以Wilcoxon 配對非參秩和檢驗評估,兩種測量方法的相關(guān)性以Spearman’s 相關(guān)系數(shù)評估;兩種測量方法所測得的灌注參數(shù)原始數(shù)據(jù)用Bland-Altman散點(diǎn)圖表示,并采用Bland-Altman方法分析兩種測量方法之間一致性,測量數(shù)據(jù)經(jīng)對數(shù)轉(zhuǎn)換;P<0.05為差異有統(tǒng)計學(xué)意義。
兩種方法所測得灌注參數(shù)
Deconvolution法:胰腺癌和pNET的BFD分別為(54.8±22.4)和(282.8±138.5)ml/(100ml·min),BVD分別為(5.2±3.0)和(25.7±9.9)ml/100 ml。
Maximum slope+Patlak法:胰腺癌和pNET的BFM分別(43.4±19.4)和(175.4±56.0)ml/(100 ml·min),BVP分別為(3.9±3.0)和(22.1±8.4)ml/100 ml。
灌注參數(shù)測量數(shù)據(jù)之間的差異Wilcoxon非參配對秩和檢驗結(jié)果顯示,胰腺癌(BFM比BFD,Z=-3.263,P=0.001;BVD 比BVP,Z=-3.978,P=0.000)和pNET(BFM 比BFD,Z=-5.212,P=0.000;BVD比BVP,Z=-2.633,P=0.008)兩種方法所測得的灌注參數(shù)之間差異均有統(tǒng)計學(xué)意義(表1)。
灌注參數(shù)測量數(shù)據(jù)之間的相關(guān)性Spearman’s 相關(guān)系數(shù)分析結(jié)果顯示,胰腺癌(BFM比BFD,r=0.845,P=0.000;BVD 比BVP,r=0.964,P=0.000)和pNET(BFM比BFD,r=0.759,P=0.000;BVD比BVP,r=0.683,P=0.000)兩種方法所測得的灌注參數(shù)間均有顯著相關(guān)性。
灌注參數(shù)測量數(shù)據(jù)之間的一致性Bland-Altman分析結(jié)果顯示,兩種灌注測量方法所測得的灌注參數(shù)BF、BV的均值差及95%一致性界限(limits of agreenment,LOA)存在明顯差異(圖 1~4A)。胰腺癌和pNET由Deconvolution方法所測得灌注參數(shù)BF和BV值均高于由Maximum slope+Patlak方法測量所得(圖1A、B,2A、B,3A、B,4A、B)。由于灌注參數(shù)BF、BV的差值與均值分布不均(圖1~4 B),本研究對原始數(shù)據(jù)進(jìn)行對數(shù)轉(zhuǎn)換后繪圖(圖1~4C),經(jīng)過反對數(shù)計算可得出pNET BFM/BFD的幾何均數(shù)及其95%LOA為0.66(0.51~0.86),BVD/BVP的幾何均數(shù)及其95%LOA為1.15(0.88~1.50);BFM/BFD的均數(shù)及其95%LOA為0.68(0.51~0.85),BVD/BVP的均數(shù)及其95%LOA為1.19(0.84~1.54)。胰腺癌BFM/BFD的幾何均數(shù)及其95%LOA為0.77(0.61~0.99),BVD/BVP的幾何均數(shù)及其95%LOA為1.42(1.13~1.79);BFM/BFD的均數(shù)及其95%LOA為0.80(0.53~1.07),BVD/BVP的均數(shù)及其95%LOA為1.46(1.10~1.81)(圖1~4D)。
表1 PAC與pNET患者CT灌注參數(shù)的差異Table 1 Differences in parameters of perfusion CT between patients with PAC and pNET
PAC:胰腺癌;pNET:胰腺神經(jīng)內(nèi)分泌瘤;BF:血流量;BV:血容量
PAC:pancreatic cancer;pNET:pancreatic neuroendocrine tumor;BF:blood flow;BV:blood volume
BFM:血流量(Maximum slope);BFD:血流量(Deconvolution)
BFM:blood flow(Maximum slope);BFD:blood flow(Deconvolution)
A.BFM與BFD一致性圖,大部分?jǐn)?shù)據(jù)都在y=x線上方;B.Bland-Altman一致性點(diǎn)圖,0為完全一致性,中間連續(xù)的實線為平均差,外邊的兩條線表現(xiàn)95%一致性界限;C.經(jīng)對數(shù)轉(zhuǎn)換的lnBFM及l(fā)nBFD之間的一致性分析,0為完全一致性,中間連續(xù)的實線為平均差,外邊的兩條線表現(xiàn)95%一致性界限;D.BFM/BFD的一致性分析,1為完全一致性,中間連續(xù)的實線為平均差,外邊的兩條線表現(xiàn)95%一致性界限
A.BFM and BFD agreement diagram:the majority of data points lie above the line of equality (y=x);B.Bland-Altman agreement plot of the difference between the BFM and BFD,against their mean values (0:line of perfect agreement;thick continuous line in the middle:the mean difference;outer lines:95% limits of agreement);C.Bland-Altman agreement plot of the difference between the ln-transformed BFM (ln BFM) and BFD (lnBFD),against their mean values (0:line of perfect agreement;thick continuous line in the middle:the mean difference;outer continuous lines:95% limits of agreement);D. Bland-Altman agreement plot of the difference between the BFM/BFD ratios,against their mean values (1:line of perfect agreement;thick continuous line in the middle:the mean difference;outer continuous lines:95% limits of agreement)
圖1 胰腺癌血流參數(shù)的一致性
Fig 1 Agreement diagram of blood flow of pancreatic cancers
BVD:血容量(Deconvolution);BVP:血容量(Patlak)
BVD:blood volume(Deconvolution);BVP:blood volume(Patlak)
A.BVD與BVP的一致性圖,大部分?jǐn)?shù)據(jù)都在y=x線下方;B.Bland-Altman一致性點(diǎn)圖,0為完全一致性,中間連續(xù)的實線為平均差,外邊的兩條線表現(xiàn)95%一致性界限;C.經(jīng)對數(shù)轉(zhuǎn)換的lnBVD及l(fā)nBVP之間的一致性分析,0為完全一致性,中間連續(xù)的實線為平均差,外邊的兩條線表現(xiàn)95%一致性界限;D.BVD/BVP的一致性分析,1為完全一致性,中間連續(xù)的實線為平均差,外邊的兩條線表現(xiàn)95%一致性界限
A.BVD and BVP agreement diagram:the majority of data points lie below the line of equality (y=x);B.Bland-Altman agreement plot of the difference between the BVD and BVP,against their mean values (0:line of perfect agreement;thick continuous line in the middle:the mean difference;outer continuous lines:95% limits of agreement);C.Bland-Altman agreement plot of the difference between the ln-transformed BVD (lnBVD) and BVP (lnBVP),against their mean values (0:line of perfect agreement;thick continuous line in the middle:the mean difference;outer continuous lines:95% limits of agreement);D.Bland-Altma agreement plot of the difference between the BVD/BVP ratios,against their mean values (1:line of perfect agreement;thick continuous line in the middle:the mean difference;outer continuous lines:95% limits of agreement)
圖2 胰腺癌血容量參數(shù)的一致性
Fig 2 Agreement diagram of blood volume of pancreatic cancers
CT灌注成像是一種無創(chuàng)的檢查技術(shù),可以提供組織形態(tài)學(xué)及量化血流動力學(xué)變化,對臨床有重要意義,特別是在腫瘤性疾病的應(yīng)用中顯得尤為重要。而CT灌注參數(shù)的測量受多種因素影響,如灌注成像的軟件及其應(yīng)用的數(shù)學(xué)模型、計算方法、對比劑濃度、注射速度、ROI部位及大小等。分析和比較灌注參數(shù)時,這些因素應(yīng)盡量保持一致。目前常用的計算方法有Deconvolution、Maximum slope及Patlak[12]。不同的計算方法所測得的灌注參數(shù)值明顯不同,本研究結(jié)果顯示Deconvolution和Maximum slope+Patlak方法所得pNET及胰腺癌的BFM與BFD、BVD與BVP間均有明顯差異。Maximum slope用單室模型來評估灌注參數(shù),其假設(shè)對比劑從流入動脈開始至最短通過時間這期間均留存在血管內(nèi),沒有靜脈的流出,所以這種方法測量的BF數(shù)值偏低[13- 14]。本研究結(jié)果亦顯示,由Deconvolution方法所測得的灌注參數(shù)BF值高于由Maximum slope 方法所得,pNET的BFD、BFM 分別為(282.8±138.5)、(175.4±56.0)ml/(100ml·min),BFD值約是BFM值的1.5倍;胰腺癌BFD、BFM分別為(54.8±22.4)、(43.4±19.4)ml/(100ml·min),BFD值約是BFM值的1.3倍,與以往研究結(jié)果類似[3,9,11,15]。Patlak用兩室模型來計算BV,其考慮到對比劑注入血流后,在組織的血管內(nèi)外間存在物質(zhì)交換,把血管內(nèi)外分為2個室區(qū)。而Patlak對噪聲更敏感,低管電壓掃描可能會影響最終結(jié)果,導(dǎo)致計算出現(xiàn)誤差[13]。Deconvolution法是利用推動剩余函數(shù)計算對比劑的靜脈流出,對灌注的流入動脈和流出靜脈進(jìn)行綜合考慮,與實際的血流動力學(xué)相近,計算出的灌注參數(shù)和函數(shù)圖更能反映病變內(nèi)部的實際情況[14]。本研究結(jié)果顯示,pNET的BVD、BVP值分別為(25.7±9.9)、(22.1±8.4)ml/100ml,BVD是BVP值的1.2倍;胰腺癌的BVD、BVP值分別為(5.2±3.0)、(3.9±3.0)ml/100ml,BVD是BVP值的1.4倍。與以往研究結(jié)果相矛盾。Goh等[16]采用GE灌注后處理軟件應(yīng)用Deconvolution方法測量的結(jié)直腸癌BV值低于西門子灌注后處理軟件Patlak方法測量所得。Bisdas等[17]研究發(fā)現(xiàn)在頭頸部腫瘤中由Maximum slope+Patlak方法所測得的BF、BV值均高于由Deconvolution方法測量所得。推測其原因可能與研究對象不同、后處理軟件不同、疾病構(gòu)成不同有關(guān),也有可能與本研究采用80kV管電壓掃描有關(guān),尚需進(jìn)一步研究。
A.BFM與BFD一致性圖,大部分?jǐn)?shù)據(jù)都在y=x線上方;B.Bland-Altman一致性點(diǎn)圖,0為完全一致性,中間連續(xù)的實線為平均差,外邊的兩條線表現(xiàn)95%一致性界限;C.經(jīng)對數(shù)轉(zhuǎn)換的lnBFM及l(fā)nBFD之間的一致性分析,0為完全一致性,中間連續(xù)的實線為平均差,外邊的兩條線表現(xiàn)95%一致性界限;D.BFM/BFD的一致性分析,1為完全一致性,中間連續(xù)的實線為平均差,外邊的兩條線表現(xiàn)95%一致性界限
A.BFM and BFD agreement diagram:the majority of data points lie above the line of equality (y= x);B.Bland-Altman agreement plot of the difference between the BFM and BFD,against their mean values (0:line of perfect agreement;thick continuous line in the middle:the mean difference;outer lines:95% limits of agreement);C.Bland-Altman agreement plot of the difference between the ln-transformed BFM (ln BFM) and BFD (lnBFD),against their mean values (0:line of perfect agreement;thick continuous line in the middle:the mean difference;outer continuous lines:95% limits of agreement);D. Bland-Altman agreement plot of the difference between the BFM/BFD ratios,against their mean values (1:line of perfect agreement;thick continuous line in the middle:the mean difference;outer continuous lines:95% limits of agreement)
圖3 胰腺神經(jīng)內(nèi)分泌腫瘤血流參數(shù)的一致性
Fig 3 Agreement diagram of blood flow of pancreatic neuroendocrine tumor
A.BVD與BVP的一致性圖,大部分?jǐn)?shù)據(jù)都在y=x線下方;B.Bland-Altman一致性點(diǎn)圖,0為完全一致性,中間連續(xù)的實線為平均差,外邊的兩條線表現(xiàn)95%一致性界限;C.經(jīng)對數(shù)轉(zhuǎn)換的lnBVD及l(fā)nBVP之間的一致性分析,0為完全一致性,中間連續(xù)的實線為平均差,外邊的兩條線表現(xiàn)95%一致性界限;D.BVD/BVP的一致性分析,1為完全一致性,中間連續(xù)的實線為平均差,外邊的兩條線表現(xiàn)95%一致性界限
A.BVD and BVP agreement diagram:the majority of data points lie below the line of equality (y=x);B.Bland-Altman agreement plot of the difference between the BVD and BVP,against their mean values (0:line of perfect agreement;thick continuous line in the middle:the mean difference;outer continuous lines:95% limits of agreement);C.Bland-Altman agreement plot of the difference between the ln-transformed BVD (lnBVD) and BVP (lnBVP),against their mean values (0:line of perfect agreement;thick continuous line in the middle:the mean difference;outer continuous lines:95% limits of agreement);D.Bland-Altma agreement plot of the difference between the BVD/BVP ratios,against their mean values (1:line of perfect agreement;thick continuous line in the middle:the mean difference;outer continuous lines:95% limits of agreement)
圖4 胰腺神經(jīng)內(nèi)分泌腫瘤血容量參數(shù)的一致性
Fig 4 Agreement diagram of blood volume of pancreatic neuroendocrine tumor
本研究Spearman’s 相關(guān)系數(shù)分析結(jié)果顯示,胰腺癌和pNET兩種方法所測得的灌注參數(shù)間均有顯著相關(guān)性。以往研究亦顯示兩種灌注測量方法在食管癌及肺內(nèi)結(jié)節(jié)中也具有較高的相關(guān)性[14,18]。本研究采用Bland-Altman法評價了兩種測量方法的一致性,結(jié)果顯示,pNET BFM/BFD和BVD/BVP的幾何均數(shù)及其95%LOA分別為0.66(0.51~0.86)和1.15(0.88~1.50),BFM/BFD和BVD/BVP的均數(shù)及其95%LOA分別為0.68(0.51~0.85)和1.19(0.84~1.54);胰腺癌BFM/BFD和BVD/BVP的幾何均數(shù)及其95%LOA分別為0.77(0.61~0.99)和1.42(1.13~1.79),BFM/BFD和BVD/BVP均數(shù)及其95%LOA分別為0.80(0.53~1.07)和1.46(1.10~1.81)。兩者之間轉(zhuǎn)換范圍較小,所以在兩種方法之間可以相互替換。與Djuric-Stefanovic 等[14]研究不同,后者研究認(rèn)為兩種灌注測量方法在食管癌灌注參數(shù)的測量中不能相互交換,其應(yīng)用灌注后處理軟件采用Deconvolution方法自動測量與手動Maximum slope方法測量比較,而本研究均采用灌注后處理軟件自動進(jìn)行兩種不同計算方法的測量,推測這可能存在一定的影響。
本研究存在以下局限性:(1)只有1名放射科醫(yī)生進(jìn)行數(shù)據(jù)測量,不能保證數(shù)據(jù)及結(jié)果的可重復(fù)性,可能對結(jié)果有一定的影響。(2)采用80kV的掃描管電壓,可能對Patlak方法測量結(jié)果有一定的影響,但是可以相應(yīng)地降低患者的輻射劑量。
綜上,本研究結(jié)果顯示,由Deconvolution及Maximum slope+Patlak方法所測得的pNET及胰腺癌灌注參數(shù)BF、BV之間存在明顯差異,Deconvolution方法測量的BF、BV值均高于Maximum slope+Patlak方法所得,兩者之間具有明顯的相關(guān)性且轉(zhuǎn)換范圍較窄,所以可以相互轉(zhuǎn)換。
[1]Bronstein YL,Loyer EM,Kaur H,et al. Detection of small pancreatic tumors with multiphasic helical CT[J]. AJR Am J Roentgenol,2004,182(3):619- 623.
[2]Ricci C,Casadei R,Taffurelli G,et al. WHO 2010 classification of pancreatic endocrine tumors. Is the new always better than the old[J]. Pancreatology,2014,14(6):539- 541.
[3]d’Assignies G,Couvelard A,Bahrami S,et al. Pancreatic endocrine tumors:tumor blood flow assessed with perfusion CT reflects angiogenesis and correlates with prognostic factors[J]. Radiology,2009,250(2):407- 416.
[4]Delrue L,Blanckaert P,Mertens D,et al. Assessment of tumor vascularization in pancreatic adenocarcinoma using 128-slice perfusion computed tomography imaging[J]. J Comput Assist Tomogr,2011,35(4):434- 438.
[5]Miles KA,Lee TY,Goh V,et al. Current status and guidelines for the assessment of tumour vascular support with dynamic contrast-enhanced computed tomography[J]. Eur Radiol,2012,22(7):1430- 1441.
[6]Park MS,Klotz E,Kim MJ,et al. Perfusion CT:noninvasive surrogate marker for stratification of pancreatic cancer response to concurrent chemo-and radiation therapy[J]. Radiology,2009,250(1):110- 117.
[7]Tsuji Y,Hamaguchi K,Watanabe Y,et al. Perfusion CT is superior to angiography in predicting pancreatic necrosis in patients with severe acute pancreatitis [J]. J Gastroenterol,2010,45(11):1155- 1162.
[8]Kandel S,Kloeters C,Meyer H,et al. Whole-organ perfusion of the pancreas using dynamic volume CT in patients with primary pancreas carcinoma:acquisition technique,post-processing and initial results[J]. Eur Radiol,2009,19(11):2641- 2646.
[9]Delrue L,Blanckaert P,Mertens D,et al. Tissue perfusion in pathologies of the pancreas:assessment using 128-slice computed tomography[J]. Abdom Imaging,2012,37(4):595- 601.
[10]Gr?zinger G,Gr?zinger A,Horger M. The role of volume perfusion CT in the diagnosis of pathologies of the pancreas[J]. Rofo,2014,186(12):1082- 1093.
[11]薛華丹,金征宇,劉煒,等. 正常胰腺及胰島細(xì)胞瘤的多層螺旋CT灌注參數(shù)特征[J]. 中國醫(yī)學(xué)科學(xué)院學(xué)報,2006,28(1):68- 70.
[12]Petralia G,Preda L,D’Andrea G,et al. CT perfusion in solid-body tumours. Part Ⅰ:Technical issues[J]. Radiol Med,2010,115(6):843- 857.
[13]Kaufmann S,Schulze M,Horger T,et al. Reproducibility of VPCT parameters in the normal pancreas:comparison of two different kinetic calculation models [J]. Acad Radiol,2015,22(9):1099- 1105.
[14]Djuric-Stefanovic A,Saranovic DJ,Masulovic D,et al. Comparison between the deconvolution and maximum slope 64-MDCT perfusion analysis of the esophageal cancer:is conversion possible[J]. Eur J Radiol,2013,82(10):1716- 1723.
[15]Xu J,Liang Z,Hao S,et al. Pancreatic adenocarcinoma:dynamic 64-slice helical CT with perfusion imaging[J]. Abdom Imaging,2009,34(6):759- 766.
[16]Goh V,Halligan S,Bartram CI. Quantitative tumor perfusion assessment with multidetector CT:are measurements from two commercial software packages interchangeable[J]. Radiology,2007,242(3):777- 782.
[17]Bisdas S,Konstantinou G,Surlan-Popovic K,et al. Dynamic contrast-enhanced CT of head and neck tumors:comparison of first-pass and permeability perfusion measurements using two different commercially available tracer kinetics models[J]. Acad Radiol,2008,15(12):1580- 1589.
[18]Sun Y,Yang M,Mao D,et al. Low-dose volume perfusion computed tomography (VPCT) for diagnosis of solitary pulmonary nodules[J]. Eur J Radiol,2016,85(6):1208- 1218.
Characteristics of CT Perfusion Parameters of Focal Pancreatic Lesions and Data Comparison of Different Algorithms
LI Ping,ZHU Liang,XUE Huadan,LIU Changyi,XU Kai,LI Juan,SUN Ting,JIN Zhengyu
Department of Radiology,PUMC Hospital,CAMS and PUMC,Beijing 100730,China
XUE Huadan Tel:010- 69155509,E-mail:bjdanna95@hotmail.com
Objective To characterize the CT perfusion parameters of focal pancreatic lesions including pancreatic cancers (PACs) and pancreatic neuroendocrine tumors (pNETs),estimate the confirmity and fungibility of parameters obtained from Deconvolution and Maximum slope+Patlak.Methods From December 2015 to November 2016,22 patients with PACs and 22 patients with pNETs(37 lesions confirmed by surgery and biopsy)underwent preoperative whole-pancreas CT perfusion in our center. The volume perfusion CT of the entire pancreas was performed at 80 kV and 100 mA,using 28 consecutive volume measurements and intravenous injection of 45 ml of iodinated contrast and saline at a flow rate of 5 ml/s. One experienced radiologists measured and recorded the CT perfusion parameters on Siemens post-processing workstation using two mathematical methods:Maximum slope+Patlak analysis versus Deconvolution method.Results Wilcoxon matched-pairs test revealed significant difference between both pairs of the perfusion measurements by the two methods,PACs(BFMvs. BFD,Z=-3.263,P=0.001;BVDvs. BVP,Z=-3.978,P=0.000); pNETs(BFMvs. BFD,Z=-5.212,P=0.000;BVDvs. BVP,Z=-2.633,P=0.008). Spearman’s correlation coefficient showed both pairs of perfusion measurements significantly correlated with each other in PACs (BFMvs. BFD,r=0.845,P=0.000;BVDvs. BVP,r=0.964,P=0.000) and pNETs(BFMvs. BFD,r=0.759,P=0.000),BVDvs. BVP,r=0.683,P=0.000). Geometric mean BFM/BFD ratio in PACs was 0.77 (range:0.61- 0.99),while geometric mean BVD/BVP ratio was 1.42 (range:1.13-1.79),within 95% limits of agreement. Geometric mean BFM/BFD ratio in pNETs was 0.66 (range:0.51- 0.86),while geometric mean BVD/BVP ratio was 1.15 (range:0.88-1.50),within 95% limits of agreement. Conclusions Significantly different CT perfusion values of blood flow and blood volume were obtained by Deconvolution-based and Maximum slope+Patlak-based algorithms in the pNETs and PACs. They correlated significantly with each other. Two perfusion-measuring algorithms are interchangeable because the ranges of the conversion factors are narrow.
pancreatic solid lesions;CT perfusion;Deconvolution;Maximum slope+Patlak
國家自然科學(xué)基金(81371608)和衛(wèi)生公益性行業(yè)科研專項項目(201402001、201402019)Supported by the National Natural Sciences Foundation of China (81371608) and the Health Industry Special Scientific Research Project (201402001,201402019)
薛華丹 電話:010- 69155509,電子郵件:bjdanna95@hotmail.com
R445
A
1000- 503X(2017)01- 0080- 08
10.3881/j.issn.1000- 503X.2017.01.014
2016- 10- 10)