CHANG Zhi－h(huán)ui，LIN Kun，DU Xiao－li，YIN Xiao－li，LU Zhao，LIU Zhao－yu

Computed tomography versus transthoracic echocardiography in the detection of complex congenial heart diseases in china：a meta－analysis

CHANG Zhi－h(huán)ui，LIN Kun，DU Xiao－li，YIN Xiao－li，LU Zhao，LIU Zhao－yu

Objective：To perform a meta－analysis to evaluate the diagnostic performance of computed tomography（CT）and transthoracic echocardiography（TTE）in complex congenital heart diseases（CHD）in China.Methods：MEDLINE，Cochrane library and China National Knowledge Infrastructure（CNKI）database from January 1966 to October 2010，were searched for initial studies in China.All the studies，published in English or Chinese，used TTE，CT，or both as diagnostic tests for CHD and reported the rate of true－positive，true－negative，false－positive and false－negative diagnoses of CHD from TTE and CT findings with the surgical results as the“gold－standard”（15 studies，XX patients）were collected.The statistic software package，“Meta－Disc 1.4”，was used to conduct data analysis.A covariate analysis was used to evaluate the influence of patient or study－related factors on sensitivity.Results：Pooled sensitivity for diagnosis of CHD were 95%[95%confidence interval（CI）：94%～96%]for CT studies and 87%（95%CI：85%～88%）for TTE studies.The difference between the pooled sensitivity of CT and that of TTE was statistically significant（P＜0.001）.TTE had higher sensitivity[0.96（95%CI：0.94～0.97）]for cardiac malformation but lower sensitivity[0.78（95%CI：0.76～0.81）]for extracardiac malformation than CT.Conclusion：CT can provide added diagnostic information compared with TTE in patients with CHD in China，especially for patients suspected of extracardiac malformation.

Echocardiography；Tomography，X－ray computed；Heart Defects，Congenital；Meta－analysis

Since the last century，the management of CHD has been vastly improved，contributing to increased survival into adulthood[1-2].This progress largely depends on the advancements of surgical repair，which would not have been possible without a clear delineation of the anatomy and physiology of the defects[3].With the characteristics of noninvasiveness，speed，safety and easy availability，transthoracic echocardiography（TTE）is always the firstline study of choice for patients with CHD.One of its limitations is the relatively small acoustic window for which TTE may be inadequate if cardiac，thoracic and visceral anomalies are needed to be clearly delineated[4].In recent years，CT technology has advanced rapidly，Cardiac CT which contributes to valuable information on congenital abnormalities is capable of complimenting echocardiography and replacing further diagnostic cardiac catheterization for anatomical delineation if performed with good contrast medium injection technique[5].

Although extensive research has been performed in regard to the diagnostic performance of CT and TTE for the detection of congenial heart diseases，the optimal imaging staging strategy has not yet been defined.The aim of our study was to perform a Meta－analysis to compare current CT and TTE in detecting abnormalities of heart diseases in China，which as far as we know，had not previously been studied.

Methods

1.Literature search

A comprehensive computer literature search[6]of abstracts about studies in human subjects was performed to identify articles about the diagnostic performance of CT and TTE for the detection of CHD in China.The MEDLINE databases，from January 1966 to October 2010，were searched with the following keywords：computed tomography（CT）or transthoracic echocardiography（TTE）and congenial heart diseases（or CHD）and sensitivity（or specificity，false－negative，false－positive，diagnosis，detection or accuracy）.The China bio－medicine databases was used for Chinese articles with the following keywords：（CT or TTE）and（congenial heart diseasesor CHD）.Other database such as Cochrane Library，and China National Knowledge Infrastructure（CNKI）database were also searched for relevant articles.The list of articles was supplemented with extensive cross－checking of the reference lists of all retrieved articles.

2.Study selection

Three observers independently checked all retrieved articles for inclusion criteria.One observer（Z.H.C）checked all articles，and two observers checked a subset of articles：one observer（Z.L）checked studies that predominantly focused on evaluation of CT，another one（Y.J.G）checked studies that predominantly focused on evaluation of TTE.Disagreements were resolved in consensus.The inclusion criteria were as follows：①articles were reported in the English or Chinese；②CT or TTE（alone or in combination，but not in sequence）was used to detect CHD；③only surgical findings were used as the reference standard；④for per－lesion statistics，sufficient data were presented to calculate the accuracy for imaging techniques；⑤when data or subsets of data were presented in more than one article，the article with the most details or the most recent article was chosen；⑥no less than 20 patients in the study was in cluded；⑦studies using sequential test combinations（e.g，CT in patients selected on the basis of CDH which are not unambiguous on TTE）were excluded，because the selection of patients on the basis of diagnostic test results could have unpredictably modified the estimate of the select characteristics of the tests themselves.

3.Data extraction

The same observers independently extracted relevant data about study characteristics and examination results by using a standardized form.Obse－rvers were not blinded with regard to such unnecessary information as the journal name，authors′name and affiliation，or year of publication，since this has been shown to be unnecessary[7].To resolve disagreement between reviewers，a third reviewer assessed all discrepant items，and the majority opinion was used for analysis.

Relevant studies were further examined with Quality Assessment of Diagnostic Accuracy Studies（QUADAS）criteria[8].To perform accuracy analyses，we extracted the following items：description of study population（age），study design（prospective，retrospective or unknown），patient selection（consecutive or not），interpretation of the test results（blinded or not）.The following features were also included：as to CT，the slices of scanner were included，and as to TTE imaging，two－dimensional or three－dimensional were included.The numbers of true positive（TP），false－negative（FN），false－positive（FP）and true－negative（TN）results in the detection of CHD were extracted on per－lesion.

4.Statistic analysis

The statistical software named“Meta－Disc”，version 1.40，was used to analyze data for CT and TTE.We only calculated pooled sensitivity for each modality.The Galbraith plots（visual inspection of forest plots of accuracy estimates）and a chi－squared statistical test were used to assess heterogeneity among the studies included in the meta－analysis.A fixed－effects model（FEM）was utilized if homogeneity existed among different studies，while a random－effects model（REM）was used if heterogeneity existed.

Results

1.Literature search and selection of studies

Fig 1 Article selection process.

The detailed procedure of study selection in the meta－analysis was showed in Fig 1.Total of 936 initial studies were searched from all the databases.After reading the title or abstracts，we reviewed 128 studies in detail.Of all these articles，113 were excluded because①the aim of the articles was not to reveal the diagnostic value of CT and TTE for detection of CHD（n＝64）；②insufficient data were reported to construct（n＝12）；③researchers in the articles did not only use surgery finding as the reference standard（n＝10）；④the studies were not based on per－lesion statistics（n＝6）；⑤the number of“yes”response to the 14 questions of QUADAS was fewer than 7（n＝2）；⑥fewer than 20 patients reported in the study（n＝9）.Finally，15 articles[9-23]fulfilled all the inclusion criteria and were selected for data extraction and data analysis.

2.Study design characteristics

The principal characteristics of the 15 studies included in the meta－analysis is in Table 1.635 patients were included in this meta－analysis.In the selected 15 studies，all studies were for both of CT and TTE.

3.Study quality

We used the“QUADAS”quality assessment tool to evaluate each selected study.All the eligible studies′score was more than 6 in the 11 questions（Table 2）.All studies did not describe the interval time between index tests diagnostics and confirmation.Patients in most studies did not receive the same reference standard.In most of the studies，the interpretation of the reference standard results with or without knowledge of the index test results was not clear.To the other questions，the“yes”responses were more than 80.0%.

4.Summary estimates of sensitivity and heterogeneity assessing

The pooled sensitivity for CT and TTE was 0.95（95%CI：0.94～0.96）and 0.87（95%CI：0.85～0.88），respectively.CT had significantly higher sensitivity estimates than TTE（P＜0.001）.However，to the studies both for CT and TTE，the sensitivity were highly heterogeneous，which affected the diagnostic value.There was no heterogeneity among studies about TTE for cardiac malformation（sensitivity：heterogeneitychi－squaredwas18.34，P＝0.1917，I2＝23.7%）.The forest plots for the sensitivity of CT and TTE were showed in Fig 2～7.

Table 1 Participants′characteristics and types of studies included

Table 2 Evaluation of quality of included studies using the QUADAS tool

Fig 2 Sensitivities and 95%confidence intervals（CI）for studies assessing the diagnostic accuracy of CT in all malformation of CDH. Fig 3 Sensitivities and 95%CI for studies assessing the diagnostic accuracy of CT in cardiac malformation.of CDH.Fig 4 Sensitivities and 95%CI for studies assessing the diagnostic accuracy of CT in extracardiac malformation.of CDH.Fig 5 Sensitivities and 95%CI for studies assessing the diagnostic accuracy of TTE in all malformation.of CDH. Fig 6 Sensitivities and 95%CI for studies assessing the diagnostic accuracy of TTE in cardiac malformation.of CDH. Fig 7 Sensitivities and 95%CI for studies assessing the diagnostic accuracy of TTE in extracardiac malformation of CDH.

5.Subgroup analysis

The results of subgroup analysis performed for deformity position（cardiac and extracardiac）were also presented.Cardiac malformation included atrial septal defect，ventricular septal defect，valve deformity，patent foramen ovale and endocardial cushion defects.Extracardiac malformation included patent ductus arteriosus，variation of coronary anatomy，coarctation of aorta，anomalous pulmonary venous drainage and anomalous origin of pulmonary artery.

For cardiac malformation，the sensitivity of CT was significantly lower than TTE（P＜0.05），but the sensitivity for extracardiac malformation，CT was higher than TTE（P＜0.05）.We also did a subgroup analysis for the dual－source CT，and found the sensitivity about dual－source CT was similar to the overall pooled sensitivity mentioned before.

Discussion

For the diagnosis of CHD，various imaging tools have been used，including cardiac catheterization，CT，echocardiography and MRI.For the past 20 years，TTE remains a first－line imaging examination in patients with suspected CHD.It is inexpensive and reproducible and demonstrates cardiac function in real time[24].This examination method has been considered the clinical gold standard，and often can eliminate the need for cardiac catheterization.With the increased use of CT angiography，cardiac CT has shown its value of providing valuable information in the detecton of congenital abnormalities.The purpose of this study is to perform a meta－analysis to compare the diagnostic performance of CT and TTE in the diagnosis of CHD.

To avoid selection bias，both the MEDLINE database and the EMBASE and the Cochrane Database of Systematic Review were searched for relevant articles.In addition，all reference lists were checked manually.To minimize bias in the selection of studies and in data extraction，reviewers independently selected articles on the basis of inclusion criteria，and scores were assigned to study design characteristics and examination results by using a standardized form that was based on the QUADAS tool.The QUADAS tool is an evidence－based quality assessment tool，which was developed for systematic reviews of studies of diagnostic accuracy[25].Data were analyzed by means of a random effects approach，which accounts for the heterogeneity between studies.

To our knowledge，this meta－analysis was the first report that assessed and compared summary estimates of overall diagnostic ability for CT and TTE.The results of our meta－analysis demonstrated CT had higher pool sensitivity than TTE in the detection of CHD.TTE was always a routine examination of CHD，however，TTE provided not enough information concerning heart－large vascular connecting and extracardiac malformation，which may influence the further treatment strategy.Therefore，to a highly suspected patient with heart－large vascular connecting or extracardiac malformation，additional imaging information was necessary.Regarding that CT can provide more anatomic information than TTE，CT might be helpful in the selection of patients who may derive a significant survival benefit.CT may be a more useful supplement to current surveillance techniques，particularly for those patients with heartlarge vascular connecting or extracardiac malformation.

TTE has been long known to have a better performance in young children due to increased permeability of the ribs with a broader acoustic window.In our analysis，Echocardiography appears more suitable for the evaluation of lesions of the cardiac valves as well as the cardiac septa.We attribute these findings to the technical implications of each imaging modality.CT can′t display function and ventricular blood flow but TTE can，which has been recognized as very useful for the evaluation of atrioventricular valve defects[26].Moreover，the color flow Doppler feature of TTE is particularly useful for evaluating flow across atrial and ventricular septal defects，such as malformations，sclerosis and fusion[27].

However，despite its advantages in showing muscular wall movement and ventricular blood flow，TTE does have limitations compared to CT，mainly because of its limited acoustic window，Therefore，CT might be more useful for congenital coronary artery anomalies.For example，the assessment of the coronary vessels largely benefits from simultaneous delineation of the entire coronary vascular system，including its origins.Therefore，CT might be more useful for congenital anomalies of the coronary vessels.

Further analysis revealed that pulmonary stenosis and aortic coarctation are the most common diseases leading to the false positive results.Our data also indicate that CT is useful for the diagnosis of congenital anomalies of the great arteries.Finally，intraand inter－observer variability，depending sonographer experience and relatively poor temporal and spatial resolution compared with radiologic tools are disadvantages of echocardiography.The development of new echocardiography－based technologies，including three－dimensional echocardiography and transesophageal echocardiography appears to hold significant promise to improve CHD assessment[29].

During the years，substantial improvements in CT（eg，introduction of spiral CT，multisection CT，dual－source CT）have been introduced.To account for these improvements，data for techniques were analyzed separately，and subgroup analyses were performed.The study shows that，there was no significant difference between dual－source CT and singlesource CT both in cardiac and extracardiac malformation.But dual－source CT can shorten the time and reduce the radiation dose and reduce the influence of heart rate.With 256－，320－or 640－section CT applied gradually，we believe CT can provide accurate information in the diagnosis of CHD in future.

Conclusion

This meta－analysis was the first report that assessed and compared summary estimates of overall diagnostic ability of CT and TTE that were currently used for detecting CHD in China.The results of our meta－analysis demonstrated that CT might be a useful technique for CHD，particularly for those patients with heart－large vascular connecting and extracardiac malformation，however，TTE may have additional value in detecting cardiac malformation.

[1]Freedom R，Lock J，Bricker JT.Pediatric cardiology and cardiovascular surgery：1950－2000[J].Circulation，2000，102（20）：58－68.

[2]Choussat A.The best of pediatric cardiology in 2002[J].Arch Mal Coeur Vaiss，2003，96（1）：51－55.

[3]Frommelt M，F(xiàn)rommelt PC.Advances in echocardiographic diagnostic modalities for the pediatrician[J].Pediatr Clin North Am，1999，46（4）：427－439.

[4]Soongswang J，Nana A，Laohaprasitiporn D，et al.Limitation of transthoracic echocardiography in the diagnosis of congenital heart diseases[J].J Med Assoc Thai，2000，83（Suppl 2）：S111－S117.

[5]Khatri S，Varma SK，Khatri P，et al.Sixty－four－slice multidetector－row computed tomographic angiography for evaluating congenital heart disease[J].Pediatr Cardiol，2008，29（2）：755－762.

[6]Deville WL，Bezemer PD，Bouter LM.Publications on diagnostic test evaluation in family medicine journals：an optimal search strategy[J].J Clin Epidemiol，2000，53（1）：65－69.

[7]Berlin JA.Does blinding of readers affect the results of meta－analyses？University of Pennsylvania meta－analysis blinding study group[J].Lancet，1997，350（9072）：185－186.

[8]Whiting P，Rutjes AW，Reitsma JB，et al.The development of QUADAS：a tool for the quality assessment of studies of diagnostic accuracy included in systematic reviews[J].BMC Med Res Methodol，2003，10（3）：25.

[9]HUANG MP，LIANG CHZ，ZENG H，et al.Multi－slice computed tomography in infants and children with complex congenital heart disease[J].Chin J Radiol，2004，38（7）：726－731.

[10]WANG RP，XIAN ZHY，YANG MF，et al.The diagnostic value of multi－slice spiral CT in complex congenital heart disease[J].J Clin Radiol，2007，26（4）：341－346.

[11]YANG YY，WANG SY，ZHOU XH，et al.The clinical application of 64－slice computed tomography in the diagnosis of complex congenital Heart disease[J].J Clin Radiol，2007，26（10）：1029－1032.

[12]LUO DD，LIANG HX，ZHONG J.Multi－slice computed tomography in infants with complex congenital heart disease[J].GuangDong Med J，2008，29（12）：2033－2035.

[13]ZHONG JS，SHEN HL，DING Y.Clinical application investigation of 64 multi－slice spiral computed tomography angiography in infants and children with congenital heart diseases[J].J Chin Clin Med Imaging，2009，20（7）：560－563.

[14]GUO JG，GUO SL，ZHOU HQ，et al.64－slice spiral CT in the diagnosis of congenital heart disease[J].Radiol Pract，2009，24（7）：732－735.

[15]HUANG XM，DUAN Q，XUE YJ，et al.Clinical application of dual－source computed tomography in diagnosis of complex congenital heart disease[J].Chin J Cardiovasc Rehabil Med，2009，18（2）：167－170.

[16]RUAN Weiyong，YANG Ming，TENG Gaojun.Clinical value of 64－slice computed tomography in the diagnosis of complex congenital heart disease in infants and children[J].J Southeast Univ（Med Sci Edi），2009，28（3）：212－216.

[17]QIN WH，CHEN WY，QIN LQ，et al.16－slice spiral CT in diagnosis of complicated congenital heart disease[J].J Pract Radiol，2009，25（7）：1068－1070.

[18]WANG J，LIU XD，LI ZHY.The application value of dual source spiral computed tomography and echocardiogram on the diagnosis of child congeIlital heart disease[J].Acta Acad Med Weifang，2009，31（3）：168－170.

[19]LI GW，LIU B，WANG WQ，et al.Clinical value of 64－slice computed tomography in the diagnosis of complex congenital heart disease[J].J Pract Radiol，2010，26（2）：183－188.

[20]LV JL，YI YCH，HAN B，et al.Clinical value of dual－source spiral CT in the diagnosis of congenital heart disease in children[J].Journal of Shandong University（Health Sciences），2010，48（7）：77－82.

[21]WANG HZH，WANG XH，HAN X，et al.Comparative study of 64－slice spiral CT and echocardiography in the diagnosis of congenital heart diseases[J].Journal of Harbin Medical University，2010，44（3）：279－282.

[22]XIAO Y，TIAN JM，GONG J，et al.Application of 64－slice spiral CT angiography in congenital heart disease[J].Chin Comput Med Imaging，2007，13（3）：167－170.

[23]LI JL，LI YF，Zhang ZW，et al.Multislice spiral computed tomography in detecting congenital heart diseases in children[J].Chin J Pract Pediatr，2005，20（5）：300－301.

[24]Hagendorff A，Pfeiffer D.Echocardiographic functional analysis of patients with rheumatoid arthritis and collagen diseases[J].Z Rheumatol，2005，64（10）：239－248.

[25]Whiting PF，Rutjes AW，Westwood ME，et al.QUADAS－2：a revised tool for the quality assessment of diagnostic accuracy studies[J].Ann Intern Med，2011，155（8）：529－536.

[26]Geibel A.Echocardiographic evaluation in unoperated congenital heart disease in adults[J].Herz，1999，24（8）：276－292.

[27]Jain P，Shrivastava S，Saxena A，et al.Significance of left ventricular inflow gradients in patients with ventricular septal defect[J].Indian Pediatr，1995，32（6）：743－747.

[28]Mao S，Shinbane JS，Girsky MJ，et al.Coronary venous imaging with electron beam computed tomographic angiography：three－dimensional mapping and relationship with coronary arteries[J].Am Heart J，2005，150（4）：315－322.

[29]Salehian O，Chan KL.Impact of three－dimensional echocardiography in valvular heart disease[J].Curr Opin Cardiol，2005，20（8）：122－126.

R814.42；R445.1；R541.1

A

1000－0313（2012）11－1168－06

110004 沈陽，中國醫(yī)科大學附屬盛京醫(yī)院放射科

暢智慧（1984－），男，山西陽城人，碩士，住院醫(yī)師，主要從事影像診斷及外周介入治療工作。

劉兆玉，E－mail：liuzy＠sj－h(huán)ospital.org

2012－04－05

2012－08－22）