劉　慧　林　江△　陸秀良　顧君英　姚家美

(1復旦大學附屬中山醫(yī)院放射科,2病理科　上?！?00032)

320排容積CT雙血供灌注評估肺占位性病變的良惡性及與微血管密度的相關性

劉慧1林江1△陸秀良1顧君英1姚家美2

(1復旦大學附屬中山醫(yī)院放射科,2病理科上海200032)

【摘要】目的評價肺占位性病變雙重血供CT灌注(dual-input CT perfusion,DI-CTP)的可重復性、鑒別良惡性病變的能力及與病變微血管密度(microvessel density,MVD)的相關性。方法116例經病理證實的肺占位性病變患者接受320排容積CT的DI-CTP檢查,由兩名觀察者單獨進行DI-CTP參數測量,獲得病變的肺動脈血流量(pulmonary flow,PF)、支氣管動脈血流量(bronchial flow,BF)及血流灌注指數(perfusion index,PI),并計算灌注總量(total perfusion,TPF)。評價觀察者內及觀察者間的可重復性;分析良惡性病變DI-CTP參數的差異;并對其中94例外科手術切除病灶進行CD34免疫組化染色分析DI-CTP參數與MVD間的相關性。結果觀察者間和觀察者內的可重復性達到良好以上(ICC>0.90)。良惡性肺占位的BF、PF、PI差異有統(tǒng)計學意義(P<0.05),其中PI的ROC曲線下面積為0.936。良惡性肺占位間的MVD差異具有統(tǒng)計學意義(P<0.05);BF、PF及TPF與MVD呈正相關。結論320排容積DI-CIP可重復性良好,其參數可反映肺占位性病變的血管生成情況,并為鑒別肺占位的良惡性提供依據。

【關鍵詞】肺部病變;CT灌注成像;雙重血供;可重復性;微血管密度

CT 灌注成像(CT perfusion,CTP)作為一種無創(chuàng)性評價組織血流灌注的功能成像方式,已經成為評價腫瘤血管生成及抗血管生成治療藥物效果的主要檢查方法[1]。以往肺癌CTP由于受掃描技術限制,一般只能采用主動脈作為輸入動脈,即采用單血供灌注(single-input CT perfusion,SI-CTP)模型,該模型只能反映肺腫瘤的支氣管動脈供血情況[2-3]。然而大量研究均證實肺癌的供血來自支氣管動脈和肺動脈[4-6]。

320排容積CT成像由于探測器寬度達到16 cm,在無需移床的情況下,一次曝光掃描范圍即可覆蓋成人肺臟的大部,單次容積掃描一般可同時包含肺內病灶、肺動脈和主動脈,該掃描技術能夠同時捕捉到肺動脈、主動脈以及病灶的動態(tài)強化特征,在此基礎上建立的雙血供灌注(dual-input CT perfusion,DI-CTP)模型可以獲得反映肺部病變兩套血供的參數,對于肺內病灶的診治有重要意義。

目前使用320排容積CT進行肺占位DI-CTP的研究才剛剛起步,國內外文獻報道不多[7-9],且主要用于良惡性病變的鑒別診斷,尚未見系統(tǒng)性地評估肺占位性病變DI-CTP觀察者可重復性的研究,尚未見將DI-CTP參數與病變的微血管密度(micro-vessel density,MVD)這一評價病變血管生成“金標準”進行對照的報道。本研究目的是評價肺占位性病變DI-CTP觀察者的可重復性和對肺占位性病變的鑒別診斷價值,并研究其與MVD之間的相關性。

資 料 和 方 法

一般資料本研究為前瞻性,經復旦大學附屬中山醫(yī)院醫(yī)學倫理委員會批準。共116例肺占位性病變患者接受灌注成像檢查,其中男93例,女23例,平均年齡60歲(42～80歲)。納入標準:(1)病灶最大徑≥1cm,為實性病灶;(2)檢查前未接受任何治療;(3) 肝、腎功能良好,無對比劑過敏史;(4)有能力配合檢查。所有病例在CT檢查后均經病理證實(外科手術切除94例,經支氣管鏡活檢16例,CT引導下穿刺活檢6例),其中惡性病變95例,包括腺癌39例、鱗癌34例、腺鱗癌3例、小細胞肺癌14例、類癌3例、肉瘤樣癌l例、不能分型肺癌l例;良性病變21例,包括錯構瘤4例、炎性病變14例、肺結核2例和肺硬化性血管瘤l例。

檢查方法采用Toshiba Aquilion One 320排螺旋CT掃描儀,先進行常規(guī)平掃定位,確定灌注掃描范圍,采用高壓注射器從單側上肢靜脈注射60 mL碘普羅胺(370 mg/mL),注射流率8 mL/s,開始注藥時即屏氣,2 s后啟動17個回合動態(tài)容積掃描(圖1)。掃描參數:管電壓80 kV,管電流80 mA,探測器寬度16 cm,轉速0.5 s/r,層厚0.5 mm。一次灌注掃描的平均輻射量是5.07 mSv。

圖像后處理及分析采用Toshiba雙血供灌注軟件,在肺門水平的肺動脈主干、降主動脈、左心房及肺內病灶內繪制感興趣區(qū)(regions of interest,ROI)生成4條時間-密度曲線(time-density curve,TDC)(圖2)。設置灌注窗寬范圍為0～150 HU,以排除骨質及肺組織的干擾,確保病灶灌注測量圖得到良好顯示。運行灌注軟件,自動生成512×512矩陣編碼的偽彩圖像,分別顯示肺動脈血流量(pulmonary flow,PF)、支氣管動脈血流量(bronchial flow,BF)及灌注指數[perfusion index,PI=PF/(PF+BF)](圖3),并計算灌注總量(total perfusion flow,TPF)。為了解觀察者間一致性(inter-observer agreement),患者的圖像分析由兩名高年資放射科醫(yī)師采用盲法獨立完成上述所有后處理程序,在病灶的3個代表層面各測量1次,最終采用3次測量值的平均值;為了解觀察者內一致性(intra-observer agreement),并最大程度降低回憶偏倚,其中1位醫(yī)師在1個月后再次對所有圖像進行測量和分析。

Seventeen intermittent low-dose volume acquisitions were made 2 s after the bolus injection (1.5 s per acquisition for the first 8 acquisitions and 2s per acquisition for the subsequent 9 acquisitions).

圖1動態(tài)容積掃描流程式圖

Fig 1Dynamic volumetric scanning process diagram

The peak enhancement time point of the left atrium (indicated by the max line) was located between the two peaks of PA and aorta, distinguishing between pulmonary and bronchial circulation. The lesion had two ascending slopes representing pulmonary and bronchial circulation respectively. The latter was much steeper than the former, which suggests that bronchial circulation was dominant in this case.

圖2輸入動脈(肺動脈,主動脈)、左心房及病灶的時間密度曲線

Fig 2Time-density curves of input arteries

(PA and aorta),left atrium and lung lesion

BF was 40.9 mL·min-1·100 mL-1, PF was 27.4 mL·min-1·100 mL-1and PI was 42.2%. Microvessel density with CD34 immunostaining was 73 (Original magnification, ×200).

圖363歲男性鱗癌病例

Fig 3Squamous cell carcinoma of a 63-year-old man

病理標本免疫組化處理及觀察94例外科手術切除病灶進行CD34染色,其中惡性病變80例(腺癌34例、鱗癌31例、小細胞肺癌7例、腺鱗癌3例、類癌3例、肉瘤樣癌l例、不能分型肺癌l例),良性病變14例(肺結核1例、炎性病變9例、錯構瘤4例)。以PBS液代替一抗作為陰性空白對照,MVD計數采用Weidner方法,結果判定在盲法下進行。

統(tǒng)計學分析使用SPSS 21.0及MedCale 14.8.1軟件完成。采用Bland-Altman法及組內相關系數(intraclass correlation coefficient,ICC)分析DI-CTP檢查的觀察者一致性;采用非參數秩和檢驗(Wilcoxon法)比較惡性與良性肺占位性病變DI-CTP參數,并繪制有鑒別診斷價值參數的受試者工作特性曲線(receiver operating characteristic curve,ROC曲線);運用Youden指數決定ROC曲線的最佳工作點(optimal operating point,OOP),并計算OOP、相應敏感性、特異性、似然比(1ikelihood ratio,LR)、陽性預測值(positive predictive value,+PR)和陰性預測值(negative predictive value,-PR);采用Spearman相關系數分析各DI-CTP參數與MVD的相關性。

結果

DI-CTP結果觀察者間和觀察者內的可重復性均達到良好以上(ICC>0.90),觀察者間測定病灶的PF、BF、PI值的ICC分別為0.96、0.98和0.95;觀察者內ICC則分別為0.97、0.99和0.97。病變最大徑(40.96±17.74) mm,范圍14～120 mm。良、惡性肺占位的支氣管動脈血流量BF、肺動脈血流量PF、灌注指數PI均數間差異具有統(tǒng)計學意義(P<0.05);灌注總量TPF均數間的差異無統(tǒng)計學意義,具體統(tǒng)計結果見圖4、表1。BF、PF、PI值鑒別良惡性肺占位的ROC曲線下面積(area under curve,AUC)分別為0.644、0.711、0.936,其中PI數值最大,具有最高的診斷準確性(圖5)。PI=51.15%時,鑒別良、惡性肺占位的敏感度是0.95,特異度是0.84,+LR是17.54,-LR是0.17,+PR是17.54,-PR是0.17。

PF:Pulmonary flow;BF: Bronchial flow;PI:Perfusion index;TPF:Total pulmonary flow.

圖4　良惡性病變CTP參數箱線圖(n=116)

BF,PF,TPF:mL·min-1·100 mL-1;PI:100 %.

Parameters for identifying lung lesions.

圖 5CT灌注成像四個參數診斷肺占位性病變的ROC曲線

Fig 5ROC curves of the four perfusion parameters

for identifying lung lesions

MVD計數結果良惡性肺占位的MVD分別為(139.49±107.85)條/視野和(62.04±35.43)條/視野,差異具有統(tǒng)計學意義(P=0.011)。

DI-CTP參數與MVD相關性分析 94例免疫組化染色的肺占位性病變BF、PF、TPF與MVD呈正相關(P值均<0.05),PI與MVD無相關性(P>0.05,表2)。

表2　CT灌注參數與MVD相關性分析

討論

應用CTP對肺占位性病變進行鑒別診斷的病理生理基礎是良惡性病變的血管生成和血管反應的性質不同造成其血液動力學差異。既往研究已證實肺部CTP檢查可以用來鑒別診斷良惡性病變[10-12],并認為CTP參數與MVD具有較好的相關性[2,13]。

肺占位性病變320排容積CT雙血供灌注成像研究的技術優(yōu)勢肺占位血流灌注不均勻[2],如CTP掃描只覆蓋其中心區(qū)域,則可重復性有限,故需要擴大掃描覆蓋范圍,實現病灶容積CTP[13-14]。320排CT的覆蓋范圍達16 cm,可實現真正意義的容積灌注,另外,320排容積CT的球管轉速為0.5 s/r,掃描時間間隔大大縮短,時間分辨率進一步提高,故能更真實準確地反映病變的血供情況,其可重復性明顯提高。灌注研究一直受輻射劑量的限制,本研究采用低劑量掃描,平均輻射量是5.07 mSv,為歐共體標準的55.7%,極大地降低了受檢者的輻射劑量。

目前,國內外灌注研究[2,11,15]多應用SI-CTP模型計算灌注參數,然而SI-CTP模型通常只能反映肺癌的優(yōu)勢供血來源,而供血相對少的那一部分血流灌注則被忽略。Yuan等[7]認為DI-CTP模型可以很好地應用于肺癌的灌注研究,Ohno等[16]對肺孤立結節(jié)的灌注研究發(fā)現DI-CTP模型比SI-CTP模型具有更好的定性診斷價值。本研究也發(fā)現,使用DI-CTP模型可以將肺占位性病變內肺動脈和支氣管動脈供血分開觀察,測量的灌注值能夠更全面、準確地反映肺占位的血供特點,且使用DI-CTP模型進行灌注研究的可重復性很好,可能更有利于病灶血供的觀察與比較。

CTP參數在肺占位病變鑒別診斷中的價值肺部病變具有肺動脈及支氣管動脈雙重血供,并且惡性病變多以支氣管動脈供血為主,良性病變多以肺動脈供血為主[6,8]。本研究中BF反映了支氣管動脈供血豐富程度,PF反映了肺氣管動脈供血豐富程度,PI反映了肺動脈血流量在總灌注量中所占比例。本研究顯示惡性病變組的BF值高于良性病變組,而PF及PI值低于良性病變組,該結果與不同性質肺占位的病理生理基礎相一致,其中PI的ROC AUC最大,具有最高的診斷準確性,并且可用PI<51.15%作為良惡性病變的診斷閾值。本研究中惡性病變的PF及PI值均高于Yuan 等[8]的研究,LR和PR值也有所差異,這可能與對比劑注射方法和掃描技術不同有關,也可能與病變大小、位置和個體生理差異等因素有關。

肺占位性病變CTP參數與MVD相關性分析腫瘤內MVD可定量反映腫瘤血管生成狀態(tài),是目前評價腫瘤血管生成的“金標準”[14,17-18]。MVD可以通過一些特異性抗體進行標記,且多采用Weidner的計數標準[19]。CTP成像則是可以定量反映病變微循環(huán)特征的功能成像方法。既往CTP成像運用的是單血供模型,對肺占位病變CTP參數與MVD相關性分析顯示,血流量和血容量與MVD有較好的相關性[12,20-21]。本文首次應用DI-CTP模型對其參數與MVD做相關性研究,結果發(fā)現肺占位病變BF、PF、TPF與MVD呈正相關;且良惡性肺占位的MVD差異具有統(tǒng)計學意義;另外TPF與MVD相關系數最高,故BF、PF及TPF可以無創(chuàng)性間接反映肺占位性病變的MVD這一病理特征,并具備幫助鑒別病變的良惡性、指導腫瘤治療和評判療效的潛能。本研究中良性病例多為炎性病變,可能導致良性病變的血流總量及MVD高于惡性病例。

綜上所述,320排容積DI-CTP可重復性好,其參數可反映肺占位性病變的血管生成情況,并對肺占位的良惡性鑒別提供幫助。

參考文獻

[1]MILES KA,LEE TY,GOH V,etal.Current status and guidelines for the assessment of tumour vascular support with dynamic contrast-enhanced computed tomography[J].EurRadiol,2012,22(7):1430-1441.

[2]MILES KA,GRIFFITHS MR.Perfusion CT:a worthwhile enhancement?[J].BrJRadiol,2003,76(904):220-231.

[3]MA SH,LE HB,JIA BH,etal.Peripheral pulmonary nodules:relationship between multi-slice spiral CT perfusion imaging and tumor angiogenesis and VEGF expression[J].BMCCancer,2008,8:186.

[4]MILNE EN,ZERHOUNI AE.Blood supply of pulmonary metastases[J].ThoracicImaging,1987,2(4):15-23.

[5]TACELLI N,REMY-JARDIN M,COPIN M C,etal.Assessment of non-small cell lung cancer perfusion:pathologic-CT correlation in 15 patients[J].Radiology,2010,257(3):863-871.

[6]KIESSLING F,BOESE J,CORVINUS C,etal.Perfusion CT in patients with advanced bronchial carcinomas:a novel chance for characterization and treatment monitoring?[J].EurRadiol,2004,14(7):1226-1233.

[7]YUAN X,ZHANG J,AO G,etal.Lung cancer perfusion:can we measure pulmonary and bronchial circulation simultaneously?[J].EurRadiol,2012,22(8):1665-1671.

[8]YUAN X,ZHANG J,QUAN C,etal.Differentiation of malignant and benign pulmonary nodules with first-pass dual-input perfusion CT[J].EurRadiol,2013,23(9):2469-2474.

[9]柳維義,譚理連,李志銘,等.320排螺旋 CT 低劑量容積灌注在孤立性肺結節(jié)鑒別診斷中的價值[J].實用放射學雜志,2014,30(5):755-758,803.

[10]SITARTCHOUK I,ROBERTS HC,PEREIRA AM,etal.Computed tomography perfusion using first pass methods for lung nodule characterization[J].InvestRadiol,2008,43(6):349-358.

[11]SHAN F,ZHANG Z,XING W,etal.Differentiation between malignant and benign solitary pulmonary nodules:use of volume first-pass perfusion and combined with routine computed tomography[J].EurJRadiol,2012,81(11):3598-3605.

[12]LI Y,YANG ZG,CHEN TW,etal.First-pass perfusion imaging of solitary pulmonary nodules with 64-detector row CT:comparison of perfusion parameters of malignant and benign lesions[J].BrJRadiol,2010,83(993):785-790.

[13]YI CA,LEE KS,KIM EA,etal.Solitary pulmonary nodules:dynamic enhanced multi-detector row CT study and comparison with vascular endothelial growth factor and microvessel density[J].Radiology,2004,233(1):191-199.

[14]NG QS,GOH V,KLOTZ E,etal.Quantitative assessment of lung cancer perfusion using MDCT:does measurement reproducibility improve with greater tumor volume coverage?[J].AJRAmJRoentgenol,2006,187(4):1079-1084.

[15]KAN Z,PHONGKITKARUN S,KOBAYASHI S,etal.Functional CT for quantifying tumor perfusion in antiangiogenic therapy in a rat model[J].Radiology,2005,237(1):151-158.

[16]OHNO Y,NISHIO M,KOYAMA H,etal.Comparison of quantitatively analyzed dynamic area-detector CT using various mathematic methods with FDG PET/CT in management of solitary pulmonary nodules[J].AJRAmJRoentgenol,2013,200(6):W593-W602.

[17]JAIN RK.Normalization of tumor vasculature:an emerging concept in antiangiogenic therapy[J].Science,2005,307(5706):58-62.

[18]WEIDNER N.Intratumor microvessel density as a prognostic factor in cancer[J].AmJPathol,1995,147(1):9-19.

[19]WEIDNER N,SEMPLE JP,WELCH WR,etal.Tumor angiogenesis and metastasis-correlation in invasive breast carcinoma[J].NEnglJMed,1991,324(1):1-8.

[20]SPIRA D,NEUMEISTER H,SPIRA SM,etal.Assessment of tumor vascularity in lung cancer using volume perfusion CT (VPCT) with histopathologic comparison:a further step toward an individualized tumor characterization[J].JComputAssistTomogr,2013,37(1):15-21.

[21]BAI RJ,CHENG XG,QU H,etal.Solitary pulmonary nodules:comparison of multi-slice computed tomography perfusion study with vascular endothelial growth factor and microvessel density[J].ChinMedJ(Engl),2009,122(5):541-547.

Dual-input perfusion of lung lesions with 320-detector-row CT:its value in differentiating malignant from benign lesions and correlation with lesion micro-vessel density

LIU Hui1, LIN Jiang2△, LU Xiu-liang3, GU Jun-ying4, YAO Jia-mei5

(1DepartmentofRadiology,2DepartmentofPathology,ZhongshanHospital,FudanUniversity,Shanghai200032,China)

【Abstract】ObjectiveTo investigate the reproducibility of dual-input CT perfusion (DI-CTP) of lung lesions with 320-detector-row CT,its value in differentiation of malignant and benign lesions and the correlation between CTP parameters and microvessel density (MVD).MethodsOne hundred and sixteen patients with various lung lesions confirmed later by pathology underwent DI-CTP with 320-detector-row CT.The pulmonary trunk and the descending aorta were selected as input arteries for measuring contributions from pulmonary and bronchial circulation to the lesions.Pulmonary flow (PF),bronchial flow (BF),and perfusion index (PI) were recorded by two independent radiologists,and then total perfusion (TPF) was calculated.Intraclass correlation coefficient (ICC) and Bland-Altman statistics were used to evaluate intra-and inter-observer agreement.DI-CTP parameters were compared between malignant and benign lesions.The correlation between DI-CTP and MVD was analysed in 94 cases of lesions with immunohistochemical staining of CD34.ResultsBoth intra- and inter-observer agreements were good to excellent (ICC>0.90).PF and PI of benign lesions were higher than those of malignant lesions.BF of malignant lesions was higher than that of benign lesions.Statistically significant differences of BF,PF and PI were found between malignant and benign lesions (P<0.05) with the area under the ROC curve of PI being 0.936,the largest of the three perfusion parameters.There was statistically significant difference in MVD between benign and malignant lesions (P<0.05).BF,PF and TPF values were positively correlated with MVD .ConclusionsDI-CTP is reproducible and reflects the angiogenesis of lung lesions.It can provide additional information for differential diagnosis of malignant from benign lung lesions.

【Key words】lung lesions;CT perfusion;dual-input;reproducibility;micro-vessel density

【中圖分類號】R812

【文獻標識碼】A

doi:10.3969/j.issn.1672-8467.2016.03.012

(收稿日期:2015-11-03;編輯:王蔚)

△Corresponding authorE-mail:lin.jiang@zs-hospital.sh.cn