袁平 馬杏凡
[摘要] 目的 系統(tǒng)評(píng)價(jià)問卷調(diào)查結(jié)合便潛血的篩查方法在中國開展結(jié)直腸癌的篩查效果。 方法 計(jì)算機(jī)檢索中國知網(wǎng)、萬方、PubMed、EBSCO等中文和外文數(shù)據(jù)庫,并輔以參考文獻(xiàn)追溯和手工檢索方法,查找結(jié)直腸癌初篩人群數(shù)量≥5000例的研究文獻(xiàn)。檢索時(shí)限為建庫時(shí)間至2016年6月。采用R3.3.1軟件進(jìn)行meta分析。通過初篩陽性率、結(jié)直腸病變檢出率和早診率、結(jié)直腸癌檢出率,來分析篩查效果,并對(duì)結(jié)直腸病變檢出率進(jìn)行分層分析。 結(jié)果 最終納入28篇文獻(xiàn),共3 398 895人接受篩查。meta分析結(jié)果顯示:初篩陽性率為14%[95%CI(11%~19%),P < 0.0001]。結(jié)直腸病變檢出率為16%[95%CI(12%~19%),P < 0.0001],其中早診率為93%[95%CI(92%~95%),P < 0.0001]。結(jié)直腸癌的檢出率為1%[95%CI(1%~2%),P < 0.0001]。對(duì)結(jié)直腸病變檢出率分層分析發(fā)現(xiàn):城市(17%)高于農(nóng)村(8%);南方地區(qū)(15%)略高于北方(12%);基于體檢人群開展大腸癌伺機(jī)性篩查的結(jié)果(13%)略低于社區(qū)自然人群(16%)。 結(jié)論 問卷調(diào)查結(jié)合便潛血的結(jié)直腸癌序貫篩查方法可增加初篩陽性率,適合腸鏡依從性差的地區(qū)。
[關(guān)鍵詞] 結(jié)直腸癌;篩查;效果;meta分析
[中圖分類號(hào)] R735.5? ? ? ? ? [文獻(xiàn)標(biāo)識(shí)碼] A? ? ? ? ? [文章編號(hào)] 1673-7210(2020)03(b)-0106-07
[Abstract] Objective To systematically evaluate the effectiveness of screening for colorectal cancer in China by questionnaire survey combined with fecal occult blood tests. Methods Computer search was conducted on CNKI, WanFang, PubMed, EBSCO and other Chinese and foreign language databases, supplemented by reference tracing and manual search methods, to find literatures with a population of more than 5000 in the primary screening population of colorectal cancer. The retrieval time was from the database construction to June 2016. R3.3.1 software was used for meta-analysis. The screening effect was analyzed by the initial screening positive rate, the detection rate and early diagnosis rate of colorectal lesions and the detection rate of colorectal cancer, and the detection rate of colorectal lesions was analyzed by stratification. Results In the end, 28 articles were included, and a total of 3 398 895 people were screened. Meta-analysis results showed that the initial screening positive rate was 14% [95%CI (11%-19%), P < 0.0001]. The detection rate of colorectal lesions was 16% [95%CI (12%-19%), P < 0.0001], and the early diagnosis rate was 93% [95%CI (92%-95%), P < 0.0001]. The detection rate of colorectal cancer was 1% [95%CI (1%-2%), P < 0.0001]. Stratified analysis of the detection rate of colorectal lesions found that: urban (17%) was higher than rural (8%), the South (15%) was slightly higher than the North (12%), the results of opportunistic colorectal cancer screening based on physical population (13%) were slightly lower than those of the natural community population (16%). Conclusion The questionnaire survey combined with the sequential screening method of fecal occult blood for colorectal cancer can increase the positive rate of initial screening, suitable for areas with poor compliance with colonoscopy.
[Key words] Colorectal cancer; Screening; Effect; Meta analysis
結(jié)直腸癌是常見的惡性腫瘤,在中國是發(fā)病率上升最快的惡性腫瘤之一[1]。有研究表明,大部分結(jié)直腸癌來源于腺瘤性的息肉惡變,這一過程通常需要5~15年的時(shí)間[2]。經(jīng)篩查,及時(shí)發(fā)現(xiàn)早期病變,并予以治療,即可避免癌變。浙江大學(xué)鄭樹教授課題組,在世界范圍內(nèi)首創(chuàng)序貫結(jié)直腸癌篩查模式:問卷調(diào)查結(jié)合糞便潛血試驗(yàn)為初篩,初篩判斷為高危的人群,繼而接受結(jié)腸鏡檢查[3]。該模式具有簡便易行、成本低等優(yōu)勢(shì),因而被廣泛應(yīng)用。本研究采用系統(tǒng)評(píng)價(jià)加meta分析方法,對(duì)在中國采用上述方法開展的結(jié)直腸癌篩查進(jìn)行綜合分析,同時(shí)分析結(jié)直腸病變檢出率在不同地域、環(huán)境、人群中開展時(shí)的區(qū)別,為今后開展篩查工作提供科學(xué)依據(jù)。
1 資料與方法
1.1 檢索策略
計(jì)算機(jī)檢索中國知網(wǎng)、萬方數(shù)據(jù)庫、PubMed、EBSCO等中文和外文數(shù)據(jù)庫。檢索時(shí)限為建庫時(shí)間至2016年6月。中文檢索詞包括:結(jié)腸癌、直腸癌、結(jié)直腸癌、大腸癌、篩查;英文檢索詞包括:colon cancer,rectal cancer,colorectal cancer,carcinoma of rectum,colorectal carcinoma,screening。
1.2 納入與排除標(biāo)準(zhǔn)
納入標(biāo)準(zhǔn):①論文類型:公開發(fā)表的論文、學(xué)位論文、會(huì)議論文等;②篩查方法:利用問卷調(diào)查與大便潛血檢測(cè)(feces occult blood test,F(xiàn)OBT)進(jìn)行初篩,初篩判斷為高危人群,經(jīng)腸鏡檢查(“精篩”或稱“復(fù)篩”);③初篩人數(shù)≥5000人;④研究樣本來自中國大陸;⑤研究為現(xiàn)況調(diào)查或橫斷面調(diào)查。
排除標(biāo)準(zhǔn):①篩查方法與本文要求不一致;②文獻(xiàn)所含信息不完整;③重復(fù)文獻(xiàn);④綜述。
1.3 文獻(xiàn)篩選
2名研究者同時(shí)獨(dú)立篩選文獻(xiàn)、提取資料。嚴(yán)格遵循納入與排除標(biāo)準(zhǔn)。兩個(gè)判斷結(jié)果不一致時(shí),由第三位高級(jí)職稱的研究者裁定。
1.4 文獻(xiàn)質(zhì)量
遵循橫斷面研究質(zhì)量評(píng)價(jià)標(biāo)準(zhǔn)(美國衛(wèi)生保健治療和研究機(jī)構(gòu)推薦)[4]。該標(biāo)準(zhǔn)共11項(xiàng),每項(xiàng)可選擇“是”“否”“不清楚”。選擇“是”,記1分;選擇“否”或“不清楚”,記0分。依據(jù)總分,判斷文獻(xiàn)質(zhì)量。
1.5資料提取
利用Excel軟件制作數(shù)據(jù)庫,提取文獻(xiàn)中的重要指標(biāo)。提取內(nèi)容包括:作者姓名、題目、發(fā)表時(shí)間、研究單位、研究時(shí)間、研究地點(diǎn)、篩查方式、目標(biāo)人數(shù)、接受問卷調(diào)查人數(shù)、問卷調(diào)查陽性人數(shù)、接受FOBT人數(shù)、FOBT陽性人數(shù)、接受初篩總?cè)巳簲?shù)、初篩陽性人數(shù)、接受腸鏡檢查人數(shù)、各種病理結(jié)果診斷人數(shù)等。
1.6 檢出率及早診率的定義
結(jié)直腸病變檢出率=(腺瘤+伴中、重度異性增生的其他病變+癌)/實(shí)際腸鏡檢查人數(shù)×100%。結(jié)直腸癌檢出率=結(jié)直腸癌/實(shí)際腸鏡檢查人數(shù)×100%。早診率=(腺瘤+伴中、重度異性增生的其他病變+早期癌)/(腺瘤+伴中、重度異性增生的其他病變+癌)×100%[5]。
1.7 統(tǒng)計(jì)學(xué)方法
利用R3.3.1軟件進(jìn)行meta分析。首先,對(duì)原始率、按四種估計(jì)(包括對(duì)數(shù)轉(zhuǎn)換、logit轉(zhuǎn)換、反正弦轉(zhuǎn)換、雙重反正弦轉(zhuǎn)換)方法轉(zhuǎn)換后的率,做正態(tài)性檢驗(yàn)。然后,依據(jù)正態(tài)性檢驗(yàn)結(jié)果,選擇接近正態(tài)分布的方法。以樣本量為權(quán)重,計(jì)算合并值。經(jīng)過上述數(shù)據(jù)變換的逆轉(zhuǎn)換,可得到合并的“發(fā)生率”及其95%CI。利用Homogeneity test(Q檢驗(yàn)),完成異質(zhì)性檢驗(yàn)(檢驗(yàn)水準(zhǔn)為α = 0.1),結(jié)合I2定量判斷異質(zhì)性。若P ≥ 0.10且I2 ≤ 50%,則提示研究結(jié)果間同質(zhì),可以固定效應(yīng)模型meta分析;反之,則提示結(jié)果間存在異質(zhì)性,需隨機(jī)效應(yīng)模型meta分析。meta分析的檢驗(yàn)水準(zhǔn)為α = 0.05。發(fā)表偏倚采用Eggers′檢驗(yàn)。敏感性分析時(shí),利用排除最低質(zhì)量的文獻(xiàn),重新計(jì)算合并率。
2 結(jié)果
2.1 文獻(xiàn)檢索結(jié)果
檢索文獻(xiàn)834篇,經(jīng)多層篩選后,最終納入28篇,共3 398 895人接受初篩。文獻(xiàn)篩選流程見圖1。
2.2 納入文獻(xiàn)的基本特征與質(zhì)量評(píng)價(jià)
納入本研究的文獻(xiàn)共28篇,時(shí)間跨度12年(2004~2015年)。初篩人數(shù)共3 398 895人,不同研究樣本量差異較大,最少5270人,最多2 117 304人。依照研究地區(qū),可分為北方(6篇)、南方(22篇);依照研究地區(qū)類型,可分為城市(20篇)、農(nóng)村(6篇)、城市加農(nóng)村(2篇)。納入文獻(xiàn)質(zhì)量評(píng)價(jià)平均得分為10(9~11)分。
2.3 meta分析結(jié)果
2.3.1初篩陽性率分析? 納入的28篇文獻(xiàn)均報(bào)道了初篩總?cè)藬?shù)和初篩陽性(高危)人數(shù)。這些文獻(xiàn)報(bào)道的初篩陽性率差異較大,最高為52%,最低為6%。將28篇文獻(xiàn)采用隨機(jī)效應(yīng)模型meta分析,結(jié)果顯示初篩總陽性率為14%[95%CI(11%~19%),P < 0.0001]。見圖2。
2.3.2 結(jié)直腸病變檢出率? 28篇文獻(xiàn)均報(bào)道了接受腸鏡檢查人數(shù)、結(jié)直腸病變檢出人數(shù)。不同文獻(xiàn)報(bào)道的檢出率在0%~34%之間。將28篇文獻(xiàn)采用隨機(jī)效應(yīng)模型meta分析,結(jié)果顯示腸鏡檢出率為16%[95%CI(12%~19%),P < 0.0001]。見圖3。
2.3.3 結(jié)直腸癌檢出率? 28篇文獻(xiàn)均報(bào)道了接受腸鏡檢查人數(shù)、結(jié)直腸癌檢出人數(shù)。不同文獻(xiàn)報(bào)道的檢出率在0%~18%之間。將28篇文獻(xiàn)采用隨機(jī)效應(yīng)模型meta分析,結(jié)果顯示腸鏡檢出率為1%[95%CI(1%~2%),P < 0.0001]。見圖4。
2.3.4 早診率? 26篇文獻(xiàn)中報(bào)道檢出了早期病變,不同文獻(xiàn)報(bào)道的早檢率>50%。采用隨機(jī)效應(yīng)模型meta分析,結(jié)果顯示,早診率為93%[95%CI(92%~95%),P < 0.0001]。見圖5。
2.3.5 病變檢出率的分層分析? 由3個(gè)方面,對(duì)結(jié)直腸病變檢出率的分層分析,發(fā)現(xiàn):①按研究地區(qū)類型分類后,城市結(jié)直腸病變檢出率(17%)高于農(nóng)村(8%),并略高于城鄉(xiāng)結(jié)合地區(qū)(12%);②按地域劃分,南方地區(qū)結(jié)直腸病變檢出率(15%)略高于北方(12%);③納入本研究的28篇文獻(xiàn)中,3篇[17,20,28]由醫(yī)療機(jī)構(gòu)的體檢科完成,接受篩查者為正常體檢人群,采用伺機(jī)性篩查方式開展。其余25項(xiàng)研究,均是在社區(qū)開展的基于自然人群的篩查。分析發(fā)現(xiàn),在正常體檢人群中開展的大腸癌篩查,其結(jié)直腸病變檢出率(13%)略低于社區(qū)自然人群篩查結(jié)果(16%)。見表2。
2.3.6 發(fā)表偏倚和敏感性分析? 根據(jù)結(jié)直腸癌篩查的檢出率,繪制漏斗圖(圖6)。Eggers′檢驗(yàn)沒有檢出有統(tǒng)計(jì)學(xué)意義的發(fā)表偏倚(P > 0.05)。剔除文獻(xiàn)質(zhì)量最低的文獻(xiàn),重新計(jì)算效應(yīng)值,其結(jié)果與之前的基本相同。
3 討論
20世紀(jì)90年代,中國首創(chuàng)將數(shù)量化問卷調(diào)查結(jié)合FOBT開展結(jié)直腸癌篩查[3]。2005年,中國國家衛(wèi)生部啟動(dòng)了農(nóng)村癌癥早診早治項(xiàng)目[5],以鄭樹教授團(tuán)隊(duì)確定的篩查模式為基礎(chǔ)開展結(jié)直腸癌篩查,并一直持續(xù)至今。在此期間,北京[19]、上海[9]、天津[10]等大中型城市均開展了類似的結(jié)直腸癌篩查活動(dòng)。本文通過收集整理中國近年來開展的大型結(jié)直腸癌篩查(接受初篩人群≥5000)結(jié)果,評(píng)估上述篩查模式在中國健康人群中的效果。
多項(xiàng)研究發(fā)現(xiàn),單一采用問卷或FOBT篩查的陽性檢出率較低,而同時(shí)將這兩種方法序貫應(yīng)用后可大大提高初篩陽性率[19,21]。中國目前以社區(qū)為單位開展的結(jié)直腸癌篩查均為政府出資、向居民免費(fèi)提供。鑒于腸鏡依從性較低的現(xiàn)狀[34,35],適當(dāng)提高初篩陽性率可鼓勵(lì)更多高危人群參與腸鏡檢查,較適宜中國現(xiàn)階段篩查需要。
本研究發(fā)現(xiàn),初篩陽性并接受腸鏡者中,結(jié)直腸病變檢出率為16%。其中,早期病變的檢出率(早診率)為93%。表明篩查的一大優(yōu)勢(shì),即早期發(fā)現(xiàn)病變,已經(jīng)體現(xiàn)出來。下一步通過早期干預(yù),大大提高這部分患者的生存率,是開展篩查的最大意義所在。
分層分析發(fā)現(xiàn),城市結(jié)直腸病變檢出率(17%)高于農(nóng)村(8%);南方地區(qū)(15%)高于北方地區(qū)(12%)。上述差異與中國結(jié)直腸癌在地域和地區(qū)間的總體發(fā)病特征基本一致[36]。另外,伺機(jī)性篩查的檢出率略低于社區(qū)調(diào)查,可能與參與人群不同有關(guān)。主動(dòng)參加伺機(jī)篩查的人更注意自身健康,因此其結(jié)直腸病變患病率可能低于一般社區(qū)人群。
美國通過大規(guī)模人群動(dòng)態(tài)監(jiān)測(cè)發(fā)現(xiàn)[37],1976~2014年結(jié)直腸癌死亡率下降了51%,這主要?dú)w功于20世紀(jì)70年代開始的結(jié)直腸癌篩查和早診早治[38]。同時(shí),歐美等國家已經(jīng)開展了多項(xiàng)長達(dá)幾十年的結(jié)直腸癌篩查隊(duì)列隨訪研究[39-40],同樣證實(shí)了篩查的效果。但是中國目前開展的篩查多是一次性由政府出資完成,缺乏對(duì)照和長期隊(duì)列隨訪數(shù)據(jù)。急需開展長期隨訪研究獲得結(jié)直腸癌篩查在中國人群中的效果和衛(wèi)生經(jīng)濟(jì)學(xué)評(píng)價(jià)證據(jù),并根據(jù)這些數(shù)據(jù)進(jìn)一步優(yōu)化當(dāng)前的序貫篩查方案[8]。
[參考文獻(xiàn)]
[1]? Chen WQ,Zheng RS,Zhang SW,et al. Report of cancer incidence and mortality in China,2012 [J]. China Cancer,2016,25(1):1-8.
[2]? Guo C,Liu Q,Dai M. Colorectal cancer screening:situation and prospect [J]. Zhonghua Yu Fang Yi Xue Za Zhi,2015,49(5):377-380.
[3]? Zheng GM,Choi BC,Yu XR,et al. Mass screening for rectal neoplasm in Jiashan County,China [J]. J Clin Epidemiol,1991,44(12):1379-1385.
[4]? Rostom A,Dube C,Cranney A,et al. Celiac Disease- Appendix D. Quality Assessment Forms [M]. US:Agency for Healthcare Research and Quality,2004. sep. http://www.ncbi.nlm.nih.gov/books/NBK35156/
[5]? Dong ZW. Technical Proposal of Chinese Cancer Screening and Early Diagnosis [M]. Beijing:People′s Medical Publishing House,2009:123.
[6]? Huang YQ,Cai SR,Zhang SZ,et al. Primary evaluation of a mass screening program for colorectal tumor in China [J]. Zhonghua Yu Fang Yi Xue Za Zhi,2011,45(7):601-604.
[7]? Zhang F,Wang WG,Xie Y,et al. Analysis on the influencing factors of compliance in colorectal cancer screening among residents in Songjiang District [J]. Zhejiang Prev Med,2015,27(1):17-20.
[8]? Li J,Li QL,Xue F,et al. Application of Feces Occult Blood Testing and Questionnaire Risk Assessment in Population Screening for Colorectal Cancer [J]. China Cancer,2015, 24(5):385-389.
[9]? Lou WP,Zhang L,Zhang W,et al. Analysis on the result of screening colorectal cancer in residents in communities of Fengxian District in Shanghai [J]. Chin J Prev Contr Chron Dis,2015,23(9):657-660.
[10]? Zhao LZ,Zhang WH,Ma DW,et al. Analysis of colorectal cancer screening result in Tianjing [J]. China Cancer Clinic,2015,42(15):760-764.
[11]? Pan X,Wang GZ,F(xiàn)ang DN,et al. colorectal screening in Xuhui district of shanghai from 2012 to 2013 [J]. Chin J Clinicians(Electronic Edition),2015,9(18):3365-3368.
[12]? Shen YZ,Huang YQ. Evaluation of colorectal cancer early detection screening program in Haining [J]. Chin J Prev Contr Chron Dis,2015,23(11):836-838.
[13]? Liu Y,Jiang YL,Gui P. Analysis on colorectal cancer screening result in Yinhang community of Shanghai in 2013 [J]. Chinese Primary Health Care,2014,28(6): 71-73.
[14]? Sun C,Liu SS,Cai BX,et al. Analysis on the result of colorectal cancer screening on natural population in community of Songhua district in Shanghai [J]. Chinese Primary Health Care,2014,28(7):89-92.
[15]? Zhu J,Liu CH,Zhao Y,et al. Colorectal cancer screening in Sujiatun district [J]. China Cancer,2014,23(9):748-751.
[16]? Mao XC,Xiang F,Xu XL,et al. Analysis of colorectal cancer screening results in west Qingpu district of Shanghai [J]. Health Education and Health Promotion,2015,10(5):365-368.
[17]? Xu F,Ying LY,Chen MT,et al. Analysis of opportunistic screening of colorectal cancer in Yongkang,Zhejiang Province [J]. China Cancer,2016,25(2):96-98.
[18]? Cai BX,Ji FM,Liu SS,et al. Analysis on the result of screening colorectal cancer in residents in communities of Songjiang district in Shanghai [J]. Health Education and Health Promotion,2016,11(1):22-24.
[19]? Zhang K,Zhai AJ,He BF,et al. Practice of colorectal screening in Yungang community of Beijing,2010-2011 [J]. Chin J Clinicians:Electronic Edition,2012,6(16):4872-4873.
[20]? Li F. Screening and early prevention for asymptomatic physical examination people colorectal cancer above 50 years old [D]. Shijiazhuang:Hebei Medical College,2012, 16-23.
[21]? Han J,Wang JP,Zhang JL,et al. Analysis of colorectal screening results of Yingze district,Taiyuan [J]. Chinese Remedies and Clinics,2015,15(10):1504-1506.
[22]? Xu AG,Yu ZJ,Zhong XH,et al. Screening of high-risk group with colorectal cancer [J]. National Med J China,2010,90(2):116-118.
[23]? Zhang YY,Yuan PH,Yuan HF,et al. Jiading district community residents colorectal cancer screening results [J]. Medical Information,2015,28(5):30.
[24]? Wen M,Shan RC,Lun Z,et al. Performance value of high risk factors in colorectal cancer screening in China [J]. World Journal of Gastroenterology,2009,15(48):6111-6116.
[25]? Yu ZJ. Analysis on colorectal screening results and cost- effectiveness [D]. Guangdong:Guangdong Medical University,2006,45-50.
[26]? Cai SR,Zheng S,Zhou L,et al. Practice of colorectal cancer screening on nature population in Hangzhou [J]. Journal of practical oncology,2006,21(2):177-178.
[27]? Zhao Y. Practice of colorectal cancer screening on nature population in Shanghai [D]. Shanghai:Fudan University,2009:34-36.
[28]? Zhai AJ,Chen H,Wang GQ,et al. Cost - effectiveness of opportunistic screening for colorectal cancer in community population [J]. Chinese General Practice,2015,18(34):4184-4186.
[29]? Chen YM,Shi YL,Gu T. Analysis of colorectal cancer screening results of Chongming County [J]. Shandong Medical,2014,54(24): 64-65.
[30]? Yang F,Zhuang JM,He J,et al. Screening results analysis of colorectal cancer in Haicang district of Xiamen city [J]. J Mod Med Health,2013,29(20):3076-3077.
[31]? Gao MM,Gao XY. Analysis of colorectal cancer screening results of Jinfeng district,Yinchuan [J]. Ningxia Medical,2016,38(4):378-379.
[32]? Lu ZM,Chen JG,Zhang YX,et al. Preliminary report about the screening program for colorectal cancer by sequential fecal occult blood in Wuhan area for 4 years [J]. Chinese Journal of Gastrointestinal Surgery,2009,12(5):474-476.
[33]? Wang Q,Zhang YX,Chen JG. Preliminary report about the colorectal cancer screening in Wuhan City [J]. Colorectal Surgery,2012,18(3):198-200.
[34]? Lieborman D. Screening,surveillance,and prevention of colorectal cancer [J]. Gnstrointest Endosc Clin N Am,2008,18:595-605.
[35]? Deng SX,Cai QC,An W,et al. Factors influencing patient compliance in colorectal cancer screening:qualitative research synthesis [J]. Zhonghua Yi Xue Za Zhi,2010,90(38):2679-2673.
[36]? National cancer center. Chinese cancer registry annual report [N].2012:56-59.
[37]? Siegel RL,Miller KD,Jemal A. Cancer statistics,2017 [J]. CA:A Cancer Journal for Clinicians,2017,67(1):7-30..
[38]? Lin JS,Piper MA,Perdue LA,et al. Screening for Colorectal Cancer Updated Evidence Report and Systematic Review for the US Preventive Services Task Force [J]. JAMA,2016,315(23):2576-2594.
[39]? Shaukat A,Mongin SJ,Geisser MS,et al. Long-Term Mortality after Screening for Colorectal Cancer [J]. N Engl J Med,2013,369(12):1106-1114.
[40]? Scholefield JH,Moss S,Sufi F,et al. Effect of faecal occult blood screening on mortality from colorectal cancer: results from a randomised controlled trial [J]. Gut,2002,50(6):840-844.
(收稿日期:2019-08-08? 本文編輯:顧家毓)