Journal of the Neurological Sciences 346 (2014) 197–203
Contents lists available at ScienceDirect
Journal of the Neurological Sciences journal homepage: www.elsevier.com/locate/jns
Meta-analysis of diagnostic significance of sixty-four-row multi-section computed tomography angiography and three-dimensional digital subtraction angiography in patients with cerebral artery aneurysm Wei Guo a,b, Xu-Ying He a, Xi-Feng Li a, Dong-Xiang Qian b, Jian-Quan Yan b, De-Lin Bu b, Chuan-Zhi Duan a,⁎ a b
Department of Neurosurgery, Zhujiang Hospital, Southern Medical University, Guangzhou 510282, PR China Department of Neurosurgery, The Third Affiliated Hospital of Guangzhou Medical University, Guangzhou 510150, PR China
a r t i c l e
i n f o
Article history: Received 2 August 2014 Received in revised form 15 August 2014 Accepted 17 August 2014 Available online 27 August 2014 Keywords: Sixty-four-row multi-section computed tomography angiography Three-dimensional digital subtraction angiography Cerebral artery aneurysm Diagnostic Odds ratio Meta-analysis
a b s t r a c t Objective: Imaging methods are essential in evaluating cerebral artery aneurysms and they have evolved with recent technical advances. Sixty-four-row multi-section computed tomography (64-MSCT) angiography and three-dimensional digital subtraction angiography (3D-DSA) are two of the most popular methods. We sought to systematically explore and find out which one would be better in imaging cerebral artery aneurysm, and try to investigate the potential use and value of 64-MSCT angiography and 3D-DSA in cerebral artery aneurysm. Method: Followed by a predefined comprehensive literature search, we carefully searched both English and Chinese electronic databases for potentially relevant studies following our meta-analysis. Two reviewers independently assessed the methodological quality of the included eligible trials based on quality assessment of studies of diagnostic accuracy studies (QUADAS). Pooled summary statistics for sensitivity, specificity, positive and negative likelihood ratios (positive LR and negative LR), and diagnostic odds ratio (ORs) with their 95% confidence intervals (CIs) were utilized. Results: Final meta-analysis of 923 cerebral artery aneurysm cases were incorporated from eight cohort studies and selected for statistical analysis. Pooled sensitivity and specificity of 64-MSCT angiography in the diagnosis of cerebral artery aneurysm were 0.97 (95% CI, 0.96–0.98) and 0.91 (0.88–0.94), respectively. The pooled positive LR was 7.68 (95% CI, 3.34–17.67); and the pooled negative LR was 0.04 (95% CI, 0.03–0.05). The pooled diagnostic OR was 263.69 (95% CI, 121.19–573.77). The area under the SROC curve was 0.9934 (standard error [SE] = 0.0031). No significant evidence of publication bias was detected (P N 0.05). Conclusion: The main finding of our meta-analysis revealed that 64-MSCT angiography relative to the 3D-DSA may have a high diagnostic accuracy for the cerebral artery aneurysm. Thus, 64-MSCT angiography may be an effective tool for the early detection of cerebral artery aneurysm. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Cerebral artery aneurysm, also named intracranial cerebral aneurysm, is widely known as one of the most common and serious cerebrovascular events [1]. A previous report in this field has demonstrated that chronic inflammation might have a negative influence on degeneration in the cerebral artery aneurysm wall, which will increase the potential risk of rupture rate of an aneurysm [2]. Furthermore, ruptured cerebral aneurysms, which vary with its morphological features and patient's medical history, is one of the leading causes of morbidity and fatality in patients with subarachnoid hemorrhage [3]. In the United States, a recent epidemiologic study has shown that among a population of 100, ⁎ Corresponding author at: Department of Neurosurgery, Zhujiang Hospital, Southern Medical University, No. 253, Industrial Avenue, Haizhu District, Guangzhou 510282, PR China. E-mail address: [email protected] (C.-Z. Duan).
http://dx.doi.org/10.1016/j.jns.2014.08.023 0022-510X/© 2014 Elsevier B.V. All rights reserved.
approximately 2 people might be affected by cerebral artery aneurysms, and approximately 15,000 patients will die of complications from ruptured cerebral artery aneurysms each year [4]. In addition, it has been reported that the incidence of cerebral artery aneurysm varies greatly across gender, with a higher prevalence in women and conversely a lower prevalence in men [5]. Meanwhile, investigations have found that the recurrence rate of cerebral artery aneurysms is higher than expected, which becomes a considerable challenge [6]. As for the etiology of cerebral artery aneurysms, it has been widely accepted to be induced by interaction between ambient and genetic factors [7]. A number of studies have shown that various factors such as chronic inflammation, hypertension, hemodynamic stress, arteriovenous malformations, and excessive alcohol consumption may be significantly related with an increased risk of cerebral artery aneurysms [1,5,8]. As we all know, a rapid radiological imaging evaluation of ruptured cerebral artery aneurysms can play an important role to the successful treatment of cerebral artery aneurysms [9]. To date, researchers have paid more attention to
198
W. Guo et al. / Journal of the Neurological Sciences 346 (2014) 197–203
spiral computed tomography and digital subtraction angiography, two popular and useful imaging methods, which are extensively used in neurovascular imaging [10,11]. Sixty-four-row multi-section computed tomography (64-MSCT) angiography, also named x-ray computed tomography angiography (CTA), can reconstruct tomographic images in a variety of directions in the human body via three-dimensional horizontal scan with the help of computer-processed x-rays [12]. The whole practical process is achieved by digital geometry processing, which forms and produces a three-dimensional image of the scanned object from a series of twodimensional radiographic images [13]. The usage of CTA has dramatically increased over the last two decades, especially in the United States, Europe, and Asia, and the main target crowd is children and adults [14,15]. This excellent technique is prevalent and widely applied in medical imaging for diagnostic and therapeutic goals [16]. For example, with regard to the application, 64-MSCT angiography can provide more accurate and detailed data for analyzing the internal and external structures of organs and tissues of the patients examined, such as patients with gastric cancer and perihilar cholangiocarcinoma [17,18]. Up to now, a great number of researchers have shown that CTA with the remarkable performance of rotational time of 0.6 s and slice thickness of 0.625 mm may be an alternative method for diagnosing and treating intracranial aneurysms [19,20]. To be more specific, imaging acquisition of 64-MSCT angiography requires only 1 minute or even less, and 64MSCT is well tolerated by the majority of patients with high-risk acute subarachnoid hemorrhage [21]. The main characteristics of 64-MSCT angiography are that it can depict intracranial aneurysms at different locations and with high sensitivity and specificity; then, its unique advantages include non-invasivity, simplicity, rapidity, and better visualization, which are the prime determinants for it to become the preferred and first-intention diagnostic technique [22]. On the other hand, three-dimensional digital subtraction angiography (3D-DSA), used in a bony or dense soft tissue environment, is regarded as a fluoroscopy technique assisted by radiology to obtain vascular images [23]. The operating process is to form a pre-contrast image from radiological equipment; then, images are acquired after injecting contrast medium into the blood vessels and removing distracting structures from the first image [24]. Therefore, it was widely accepted that DSA primarily aims to image blood vessels with high image quality (b 3 mm), results that are not possible using CTA imaging [25]. Although 3D-DSA is still considered as the gold standard for arterial imaging, especially for intracranial aneurysms, it is used less and less commonly and may be substituted by 64-MSCT angiography [26]. The reasons are that it is invasive, time-consuming, expensive, and stressful for patients with intracranial aneurysms [27]. Hence, previous large studies demonstrated and concluded that 64-MSCT angiography is a more appropriate and feasible alternative to diagnose intracranial aneurysms with significant advantages in the reduction of contrast medium dose and x-ray exposure time [21]. Yet inconsistent results were also found in other relevant documents [22,28]. In the current meta-analysis, the purpose of this study was to explore the utility of 3D-DSA and 64-MSCT angiography in describing the imaging characteristics of patients with cerebral artery aneurysms and have a potential judgment on which imaging detection method is more available and reliable to the present diagnosis of cerebral artery aneurysm. 2. Materials and methods
(“Multi-slice computed tomography angiography” or “Multi-slice CT angiography” or “MSCTA” or “MS-CTA” or “64-section multidetector CT angiography” or “MSCT” or “multislice computed tomography coronary angiography” or “64-section CT angiography” or “64-MSCT angiography” or “64-section CT scanner”) and (“intracranial aneurysm” or “intracranial aneurysm” or “intracranial aneurysms” or “Anterior Communicating Artery Aneurysm” or “Anterior Communicating Artery Aneurysms” or “Basilar Artery Aneurysms” or “Basilar Artery Aneurysm” or “Middle Cerebral Artery Aneurysms” or “Posterior Cerebral Artery Aneurysm” or “Posterior Cerebral Artery Aneurysms” or “Berry Aneurysm” or “Berry Aneurysms” or “Brain Aneurysm” or “Brain Aneurysms” or “Cerebral Aneurysm” or “Cerebral Aneurysms” or “Giant Intracranial Aneurysm” or “Giant Intracranial Aneurysms” or “Intracranial Mycotic Aneurysm” or “Intracranial Mycotic Aneurysms” or “Anterior Cerebral Artery Aneurysm” or “Anterior Cerebral Artery Aneurysms” or “Posterior Communicating Artery Aneurysm” or “Posterior Communicating Artery Aneurysms”) were entered in the database searches as medical subject heading terms and text words with a highly sensitive search strategy. Manual searches were used to screen other eligible studies. 2.2. Study selection After reading the abstracts, full papers were retrieved and assessed for their suitability with the following inclusion criteria: (1) only those cohort studies conducted within a human population to explore the relationship between 64-MSCT angiography and the diagnostic value of cerebral artery aneurysm were incorporated; (2) the diagnosis and management of patients with cerebral artery aneurysm should be confirmed by its clinical etiology and manifestations, and the combined diagnosis from the laboratory and imaging methods observing the occurrence and extent of subarachnoid hemorrhage (SAH) via CT or cerebral angiography, and magnetic resonance imaging [29,30]; (3) articles must be published in a peer-reviewed journal especially providing with original data. Corresponding major exclusion criteria were those experimental articles or trails that: (1) did not satisfy the above inclusion criteria designed in the current study; (2) abstracts, reviews, case report, letters, meta-analysis, or proceedings; (3) duplication publications or studies with overlapping data; and (4) subgroup analysis of the included trials. 2.3. Data extraction and quality assessment We used a standard reporting form to extract data from each included study, and the following descriptive information were collected: surname and initials of the first author, year of submission, country and racial descent, total numbers of included cases, gender information, mean age and floating range, demographic variables, 2 × 2 table information, and confirmation of diagnosis. Diagnostic results data were also counted in the research including the true positive (TP), false positive (FP), false negative (FN), and true negative (TN) information results in four-fold (2 × 2) tables for cerebral artery aneurysm diagnosis. The quality of involved studies was assessed independently by two authors based on a tool for the quality assessment of studies of diagnostic accuracy studies (QUADAS) [31]. Fourteen assessment items were implicated in the QUADAS criteria. Each of these items was scored as “yes” (2), “no” (0), or “unclear” (1). QUADAS scores ranged from 0 to 28; and a score of ≥22 indicates good quality.
2.1. Literature search 2.4. Statistical analysis The following computerized bibliographic databases were applied to identify relevant articles related to the association of 64-MSCT angiography with cerebral artery aneurysm diagnosis, with no restrictions on language or data collection: PubMed, Embase, CINAHL, and Science Citation Index, the Cochrane Library, Current Contents Index, Chinese Biomedical, the Chinese Journal Full-Text, and the Weipu Journal.
The association of the diagnostic significance of 64-MSCT angiography with cerebral artery aneurysm was judged by the positive likelihood ratio (positive LR) and negative likelihood ratio (negative LR) combined with its 95% confidence interval (95% CI). Data were first extracted to quantify heterogeneity among studies with Cochran's Q-
W. Guo et al. / Journal of the Neurological Sciences 346 (2014) 197–203
Identification
statistic (P b 0.05 was considered significant) and I2 tests in Meta-DiSc1 software [32]. Threshold effects were the main reason contributing to the existence of heterogeneity among all the included diagnostic trails, and P N 0.05 indicated no threshold effect and thus no significant evidence of heterogeneity was shown. Threshold effect was assessed using Spearman correlation coefficients. Meanwhile, with regard to the heterogeneity among different studies, χ2 test was applied (P b 0.05 or I2 test exhibiting N 50% indicated the occurrence of heterogeneity). Additionally, when the included studies were relatively small that they could not continue the exploration of between-studies heterogeneity within meta-analysis, we always chose the random-effects model; otherwise, when the P N 0.05 or I2 test exhibited b 50% (no apparent evidence of heterogeneity), a fixed-effects model may be picked. Furthermore, independent random-effects summary sensitivity and specificity estimations were weighted; diagnostic odds ratios (DOC) were also calculated, and based on the results, summary receiver operating characteristic (sROC) curves were concluded to represent the performance of the present diagnostic test [33]. The area under the receiver operating characteristic (ROC) curve (AUC) was calculated as a performance measure for machine detection significance [34]. P b 0.05 was considered as the difference and was statistically significant. Publication bias was conducted, using a scatter plot, mainly by a regression of diagnostic odds ratio (ORs) against 1/sqrt (effective sample size), weighting by effective sample size, with the usage of slope coefficient indicating asymmetry [35]. All statistical analyses were undertaken using the SAS statistical software version 8.2 (SAS Institute Inc., Cary, NC, United States) and Meta. Disc (version 1.4) statistical software for metaanalysis of testing accuracy data.
Articles identified through electronic database searching (N = 296)
3. Results 3.1. Selection of eligible studies Fig. 1 shows the flow chart of identified publications and the main reason for exclusion. Firstly, 298 potential articles were resulted from the electronic databases. Of the 298 articles, 1 study was a duplicate and was therefore removed. After screening of the titles and abstracts, 134 irrelevant studies were excluded subsequently. Then, 153 studies were excluded after more detailed full text assessment, and 10 studies were selected for the qualitative analysis. An additional 2 studies were removed after a further assessment. Finally, 8 cohort studies published from 2008 to 2013 were analyzed in the current meta-analysis [20–22,24,25,28,36,37].
3.2. Demographic variables The 8 included studies consisted of 923 cerebral artery aneurysm with a sample size ranging from 30 to 512. Of the 8 eligible studies, only two were from Caucasians [Serbia (Milosevic Medenica S.), and Belgium (Lubicz B.)], the remaining six studies were performed in Asians [China (Tian W.B., Wang H.S., Lu L., Xing W., and Ye H.W.) and Turkey (Chang J.W.)]. In addition, all of the included studies provided the baseline characteristics information. The basis of the included trials was divided into two biases, namely, per-patient and per-aneurysm basis. The baseline characteristics in the individual studies were summarized in Table 1.
Additional articles identified through a manual search (N = 2)
Screening
Articles reviewed for duplicates (N = 298)
Articles after duplicates removed (N = 297) Studies were excluded, due to: (N = 41) Letters, reviews, meta-analysis (N = 28) Not human studies (N = 62) Not related to research topics
Eligibility
Full-text articles assessed for eligibility (N = 163)
Studies included in qualitative synthesis (N = 10)
Included
199
Studies were excluded, due to: (N = 32) Not cohort study (N = 63) Not relevant to 64-MSCT and 3D-DSA (N = 58) Not relevant to cerebral artery aneurysm
Studies included in quantitative synthesis (meta-analysis) (N = 8)
Fig. 1. Flow chart of literature search and study selection. Eight cohort studies were included in this meta-analysis.
200
W. Guo et al. / Journal of the Neurological Sciences 346 (2014) 197–203
Table 1 The characteristics and methodological quality of the included studies in this meta-analysis. First author
Year
Country
Ethnicity
Number
Gender (M/F)
Age (years)
Basis
QUADAS score
Tian WB [37] Wang HS [28] Lu L [22]
2013 2012 2012
China China China
Asians Asians Asians
30 39 512
18/12 17/22 220/293
50.0 (38–76) 47.5 (41–71) 52.0 ± 12.0
22 2 27
Xing W [21]
2011
China
Asians
133
63/70
52.0 (9–87)
Milosevic Medenica S [24] Ye HW [36] McKinney AM [25] Lubicz B [20]
2010 2009 2008 2008
Serbia China Turkey Belgium
Caucasians Asians Asians Caucasians
47 42 66 54
40/7 18/24 35/31 37/17
54.3 (13–76) 27–67 54.5 (14–93) 55.0 (18–85)
Per-patient basis Per-aneurysm basis Per-aneurysm basis Per-patient basis Per-aneurysm basis Per-patient basis Per-aneurysm basis Per-patient basis Per-aneurysm basis Per-aneurysm basis
26 23 23 24 24
M, male; F, female; QUADAS, quality assessment of studies of diagnostic accuracy studies.
3.3. Test for heterogeneity In the meta-analysis, the relationship between diagnostic significance of 64-MSCT angiography and cerebral artery aneurysms were assessed for the observed heterogeneity. The random effects model was used due to the obvious existence of heterogeneity among studies. There was no significant association (I2 = 0, P = 0.8272) of sensitivity with specificity, which indicates the absence of the threshold effect. 3.4. Sensitivity and specificity analyses of 64-MSCT angiography Concluded from the present meta-analysis, the sensitivity and specificity analyses results of the diagnostic value of 64-MSCT angiography in cerebral artery aneurysm were 0.97 (95% CI, 0.96–0.98) and 0.91 (0.88–0.94), respectively. There were a total of 8 clinical trials conducted for the analysis, and the results of the meta-analysis based on the combined data indicated that up to five trials in the pooled specificity data were 100%. Random effects model results indicated that the pooled positive LR and negative LR were 7.68 (95% CI, 3.34–17.67) and 0.04 (95% CI, 0.03–0.05), respectively (Fig. 2). 3.5. SROC and AUC analysis The ROC curve was calculated for all observers and is shown in Fig. 3. In the current clinical analyses, the pooled diagnostic ORs of 64-MSCT angiography in cerebral artery aneurysms were 263.69 (95% CI, 121.19– 573.77). The results were plotted as a symmetrical SROC curve, and the corresponding AUC based on the 64-MSCT angiography diagnostic image of observers were 0.9934 (standard error [SE] = 0.0031). The above results indicated that there were significant values in 64-MSCT angiography in the diagnosis of trigeminal neuralgia, 3.6. Publication bias result Publication bias was assessed visually by using a scatter plot. It should hold a symmetrical funnel shape; otherwise, publication bias may be detected. Furthermore, we found no evidence of obvious asymmetry based on the symmetrical funnel shape in the present systemic analyses (P N 0.05; Fig. 4). 4. Discussion In the present meta-analysis, we systematically evaluated the technical performance and accuracy of 64-MSCT angiography and 3D-DSA in diagnosing cerebral artery aneurysms. Our results showed that the pooled sensitivity and specificity of 64-MSCT angiography in diagnosing cerebral artery aneurysms were 97% and 91%, respectively. These were consistent with the previous results, in which 64-MSCT angiography had a high diagnostic accuracy accompanied with a lower rate of missed diagnosis and misdiagnosis for cerebral artery aneurysms, suggesting that 64-MSCT angiography may be an effective tool for cerebral artery
aneurysm diagnosis. As a matter of fact, it has been suggested that ruptured cerebral artery aneurysms were one of the major reasons for the occurrence of spontaneous subarachnoid hemorrhage, and without timely treatment, approximately half of the patients could experience and possibly die from ruptured cerebral artery aneurysms [38]. Previously, DSA was confirmed as the “gold standard” in the diagnosis of cerebral artery aneurysms [39]. Nevertheless, due to the normal 2D imaging presentation, it may not be easily utilized for the observation of cerebral aneurysm, traditional imaging detection actually may result in a potential partial of false negative rate [40]. In recent decades, with the development of imaging detection technology, the updated 3DDSA based on the three-dimensional reconstruction of cerebral blood vessels can, to some extent, lead to a comprehensive observation of cerebral vascular from any respect, and hence would decrease the rate of misdiagnosis [25,41]. On the other side, widely acknowledged as an invasive examination method, and characterized by a long operation and examination times, there is no doubt that 3D-DSA would not be the optimal choice for cerebral artery aneurysm patients [42]. Methods and apparatus for a non-invasive examination of cerebral vascular for cerebral artery aneurysm patients therefore became popular. Our results showed that 64-MSCT angiography had a higher sensitivity and specificity for the detection of cerebral artery aneurysm, and the reasons for this phenomenon could be attributed to the fact that 64-MSCT angiography can itself reconstruct tomographic images in a variety of directions in the human body via 3D horizontal scan with the help of computer-processed x-rays [43]. 64-MSCT angiography and its unique advantages (non-invasivity, simplicity, rapidity, and better visualization as mentioned in the Introduction) have become the dominant determinants making 64-MSCT angiography the more popular and more suitable diagnostic technique for patients and hospital staff [44,45]. Additionally, the ROC curve is a broad overview of diagnostic tests, and is also considered a relative operating characteristic curve; to a large extent, it can be attributed that it is a comparison of two operating characteristics (true positive rate and false positive rate) when the criterion changes over time [46].The best possible prediction method would yield a point in the upper left corner or coordinate (0, 1) of the ROC space, representing 100% sensitivity (no false negatives) and 100% specificity (no false positives) [47]. The present results indicate that the corresponding AUC, based on the 64-MSCT angiography diagnostic image of observers, was 0.9934, revealing a significantly high efficacy of 64MSCT angiography in brain tumors. In addition, the diagnostic ORs measure the effectiveness of a diagnostic test, and the ratio ranges from zero to infinity; higher diagnostic odds ratios are indicative of better test performance; otherwise, it means a poor detected outcome [48]. Our findings demonstrated that the pooled diagnostic ORs of 64-MSCT angiography in cerebral artery aneurysm were 263.69, suggesting that 64-MSCT angiography holds a higher value in the diagnosis of brain tumors. Furthermore, due to the restriction of small sample size, heterogeneity tests failed to continue the following stratified analyses, which, in some way, have an adverse effect on the whole reliability of the outcomes.
W. Guo et al. / Journal of the Neurological Sciences 346 (2014) 197–203
201
Fig. 2. Forest plots for the diagnostic accuracy of 64-MSCT in cerebral artery aneurysm. (A) Sensitivity, (B) specificity, (C) positive likelihood ratio, and (D) negative likelihood ratio.
202
W. Guo et al. / Journal of the Neurological Sciences 346 (2014) 197–203
Log Odds Ratio versus 1/sqrt(Effective Sample Size)(Deeks) (Egger’s test: P > 0.05)
10
100
Study Regression Line
1
Diagnostic Odds Ratio
1000
Fig. 3. Forest plot of SROC curve and DOR on the diagnostic accuracy of 64-MSCT in cerebral artery aneurysm. SROC, summary receiver operator characteristic; DOR, diagnostic odds ratio; AUC, area under the curve; SE, standard error.
0.10
0.15
0.20
0.25
0.30
1/root(ESS) Fig. 4. Funnel plot of publication bias on the pooled DOR of 64-MSCT in cerebral artery aneurysm. No publication bias was detected in this meta-analysis. DOR, diagnostic odds ratio.
Although this is the first meta-analysis focused on the comparison of 64-MSCT angiography and 3D-DSA accuracy in the diagnosis and differential diagnosis of cerebral artery aneurysm, our study still has some limitations that merit a deeper consideration. Firstly, our results lacked sufficient statistical power to evaluate the accuracy of 64-MSCT angiography due to a relatively small sample size. Secondly, meta-analysis is a retrospective study that may lead to subject selection bias. Thirdly, this meta-analysis failed to obtain original data from the included studies, which limited further clinical assessment of 64-MSCT angiography for the diagnosis of cerebral artery aneurysm. Fourthly, we failed to find potential articles with an integrity dataset that focused on the comparison of diagnostic performance between 64-MSCT angiography and 3D-DSA accuracy for cerebral artery aneurysm. Importantly, the inclusion criteria of cases and controls were not well defined in all included studies and thus might have influenced our results. In conclusion, our findings provide empirical evidence that 64-MSCT angiography may have a high diagnostic accuracy for cerebral artery aneurysm. Thus, 64-MSCT angiography may be an effective tool for the early detection of cerebral artery aneurysms. However, due to the limitations mentioned above, further detailed studies are still needed to provide a more representative statistical analysis.
W. Guo et al. / Journal of the Neurological Sciences 346 (2014) 197–203
Conflict of interest We declare that we have no conflicts of interest.
Acknowledgments We would like to acknowledge the helpful comments on this paper received from our reviewers.
References [1] Aoki T, Nishimura M, Matsuoka T, Yamamoto K, Furuyashiki T, Kataoka H, et al. PGE(2)-EP(2) signalling in endothelium is activated by haemodynamic stress and induces cerebral aneurysm through an amplifying loop via NF-kappaB. Br J Pharmacol 2011;163(6):1237–49. [2] Hasan DM, Mahaney KB, Brown Jr RD, Meissner I, Piepgras DG, Huston J, et al. Aspirin as a promising agent for decreasing incidence of cerebral aneurysm rupture. Stroke 2011;42(11):3156–62. [3] Lin N, Cahill KS, Frerichs KU, Friedlander RM, Claus EB. Treatment of ruptured and unruptured cerebral aneurysms in the USA: a paradigm shift. J Neurointerv Surg 2012;4(3):182–9. [4] Villablanca JP, Duckwiler GR, Jahan R, Tateshima S, Martin NA, Frazee J, et al. Natural history of asymptomatic unruptured cerebral aneurysms evaluated at CT angiography: growth and rupture incidence and correlation with epidemiologic risk factors. Radiology 2013;269(1):258–65. [5] Caranci F, Briganti F, Cirillo L, Leonardi M, Muto M. Epidemiology and genetics of intracranial aneurysms. Eur J Radiol 2013;82(10):1598–605. [6] Hasan DM, Nadareyshvili AI, Hoppe AL, Mahaney KB, Kung DK, Raghavan ML. Cerebral aneurysm sac growth as the etiology of recurrence after successful coil embolization. Stroke 2012;43(3):866–8. [7] Bor AS, Rinkel GJ, van Norden J, Wermer MJ. Long-term, serial screening for intracranial aneurysms in individuals with a family history of aneurysmal subarachnoid haemorrhage: a cohort study. Lancet Neurol 2014;13(4):385–92. [8] Kelliny M, Maeder P, Binaghi S, Levivier M, Regli L, Meuli R. Cerebral aneurysm exclusion by CT angiography based on subarachnoid hemorrhage pattern: a retrospective study. BMC Neurol 2011;11:8. [9] Luo Z, Wang D, Sun X, Zhang T, Liu F, Dong D, et al. Comparison of the accuracy of subtraction CT angiography performed on 320-detector row volume CT with conventional CT angiography for diagnosis of intracranial aneurysms. Eur J Radiol 2012;81(1):118–22. [10] Yang L, Huang X, Duan S. Clinical application and technique of 64-slice spiral CT subtraction angiography in head and neck. Vasa 2012;41(1):27–33. [11] Sadatomo T, Yuki K, Migita K, Taniguchi E, Kodama Y, Kurisu K. Evaluation of relation among aneurysmal neck, parent artery, and daughter arteries in middle cerebral artery aneurysms, by three-dimensional digital subtraction angiography. Neurosurg Rev 2005;28(3):196–200. [12] Kwok HC, Dirkzwager I, Duncan DS, Gillham MJ, Milne DG. The accuracy of multidetector computed tomography in the diagnosis of non-occlusive mesenteric ischaemia in patients after cardiovascular surgery. Crit Care Resusc 2014;16(2): 90–5. [13] Kropman RH, Zandvoort HJ, Van Den Heuvel DA, Wille J, Moll FL, De Vries JP. CT angiography to evaluate hemodynamic changes in popliteal artery aneurysms during flexion and extension of the knee joint. J Cardiovasc Surg (Torino) 2014;55(2 Suppl 1):249–53. [14] Furlow B. Radiation dose in computed tomography. Radiol Technol 2010;81(5): 437–50. [15] Smith-Bindman R, Lipson J, Marcus R, Kim KP, Mahesh M, Gould R, et al. Radiation dose associated with common computed tomography examinations and the associated lifetime attributable risk of cancer. Arch Intern Med 2009;169(22):2078–86. [16] Mileto A, Marin D, Ramirez-Giraldo JC, Scribano E, Krauss B, Mazziotti S, et al. Accuracy of contrast-enhanced dual-energy MDCT for the assessment of iodine uptake in renal lesions. AJR Am J Roentgenol 2014;202(5):W466–74. [17] Mehmedovic A, Mesihovic R, Saray A, Vanis N. Gastric cancer staging: EUS and CT. Med Arch 2014;68(1):34–6. [18] Ajiki T, Fukumoto T, Ueno K, Okazaki T, Matsumoto I, Ku Y. Three-dimensional computed tomographic cholangiography as a novel diagnostic tool for evaluation of bile duct invasion of perihilar cholangiocarcinoma. Hepatogastroenterology 2013; 60(128):1833–8. [19] Chen W, Yang Y, Xing W, Qiu J, Peng Y. Sixteen-row multislice computed tomography angiography in the diagnosis and characterization of intracranial aneurysms: comparison with conventional angiography and intraoperative findings. J Neurosurg 2008;108(6):1184–91. [20] Lubicz B, Levivier M, Francois O, Thoma P, Sadeghi N, Collignon L, et al. Sixty-fourrow multisection CT angiography for detection and evaluation of ruptured intracranial aneurysms: interobserver and intertechnique reproducibility. AJNR Am J Neuroradiol 2007;28(10):1949–55. [21] Xing W, Chen W, Sheng J, Peng Y, Lu J, Wu X, et al. Sixty-four-row multislice computed tomographic angiography in the diagnosis and characterization of intracranial aneurysms: comparison with 3D rotational angiography. World Neurosurg 2011; 76(1–2):105–13.
203
[22] Lu L, Zhang LJ, Poon CS, Wu SY, Zhou CS, Luo S, et al. Digital subtraction CT angiography for detection of intracranial aneurysms: comparison with three-dimensional digital subtraction angiography. Radiology 2012;262(2):605–12. [23] Amans MR, Cooke DL, Vella M, Dowd CF, Halbach VV, Higashida RT, et al. Contrast staining on CT after DSA in ischemic stroke patients progresses to infarction and rarely hemorrhages. Interv Neuroradiol 2014;20(1):106–15. [24] Milosevic Medenica S, VV V, Prstojevic B. 64-Slice CT angiography in the detection of intracranial aneurysms: comparison with DSA and surgical findings. Neuroradiol J 2010;23(1):55–61. [25] McKinney AM, Palmer CS, Truwit CL, Karagulle A, Teksam M. Detection of aneurysms by 64-section multidetector CT angiography in patients acutely suspected of having an intracranial aneurysm and comparison with digital subtraction and 3D rotational angiography. AJNR Am J Neuroradiol 2008;29(3):594–602. [26] Herzig R, Burval S, Krupka B, Vlachova I, Urbanek K, Mares J. Comparison of ultrasonography, CT angiography, and digital subtraction angiography in severe carotid stenoses. Eur J Neurol 2004;11(11):774–81. [27] Tacher V, Lin M, Desgranges P, Deux JF, Grunhagen T, Becquemin JP, et al. Image guidance for endovascular repair of complex aortic aneurysms: comparison of two-dimensional and three-dimensional angiography and image fusion. J Vasc Interv Radiol 2013;24(11):1698–706. [28] Wang HS, Zhao PL, Wang CQ, Yang ZW, Lv GS, Li QC, et al. Comparative study of 64 rows helical CT angiography and 3D-DSA in diagnosis of intracranial aneurysms. Hebei Med J 2012;34(11):1613–5. [29] Schwartz RB, Tice HM, Hooten SM, Hsu L, Stieg PE. Evaluation of cerebral aneurysms with helical CT: correlation with conventional angiography and MR angiography. Radiology 1994;192(3):717–22. [30] Vermeulen M, van Gijn J. The diagnosis of subarachnoid haemorrhage. J Neurol Neurosurg Psychiatry 1990;53(5):365–72. [31] Whiting PF, Weswood ME, Rutjes AW, Reitsma JB, Bossuyt PN, Kleijnen J. Evaluation of QUADAS, a tool for the quality assessment of diagnostic accuracy studies. BMC Med Res Methodol 2006;6:9. [32] Zintzaras E, Ioannidis JP. HEGESMA: genome search meta-analysis and heterogeneity testing. Bioinformatics 2005;21(18):3672–3. [33] Hanley JA, McNeil BJ. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology 1982;143(1):29–36. [34] Hanley JA, McNeil BJ. A method of comparing the areas under receiver operating characteristic curves derived from the same cases. Radiology 1983;148(3): 839–43. [35] Deeks JJ, Macaskill P, Irwig L. The performance of tests of publication bias and other sample size effects in systematic reviews of diagnostic test accuracy was assessed. J Clin Epidemiol 2005;58(9):882–93. [36] Ye HW, Li HW, Song Y, Zhou YS, Wu LG, Yu HT, et al. Early diagnosis of intracranial aneurysm using 64-slice spiral CT angiography: a clinical analysis. Chin J Neuromed 2009;08(06):578–80. [37] Tian WB, Zhang XW, Fan BS. Comparative study of 64 rows helical CT angiography and DSA in diagnosis of intracranial aneurysms. Chin J Pract Nerv Dis 2013; 16(09):69–70. [38] Westerlaan HE, van Dijk JM, Jansen-van der Weide MC, de Groot JC, Groen RJ, Mooij JJ, et al. Intracranial aneurysms in patients with subarachnoid hemorrhage: CT angiography as a primary examination tool for diagnosis–systematic review and meta-analysis. Radiology 2011;258(1):134–45. [39] Romijn M, Gratama van Andel HA, van Walderveen MA, Sprengers ME, van Rijn JC, van Rooij WJ, et al. Diagnostic accuracy of CT angiography with matched mask bone elimination for detection of intracranial aneurysms: comparison with digital subtraction angiography and 3D rotational angiography. AJNR Am J Neuroradiol 2008;29(1):134–9. [40] Kucukay F, Okten RS, Tekiner A, Dagli M, Gocek C, Bayar MA, et al. Threedimensional volume rendering digital subtraction angiography in comparison with two-dimensional digital subtraction angiography and rotational angiography for detecting aneurysms and their morphological properties in patients with subarachnoid hemorrhage. Eur J Radiol 2012;81(10):2794–800. [41] Toyota S, Iwaisako K, Takimoto H, Yoshimine T. Intravenous 3D digital subtraction angiography in the diagnosis of unruptured intracranial aneurysms. AJNR Am J Neuroradiol 2008;29(1):107–9. [42] Gerardin E, Tollard E, Derrey S, Langlois O, Dacher JN, Douvrin F, et al. Usefulness of multislice computerized tomographic angiography in the postoperative evaluation of patients with clipped aneurysms. Acta Neurochir (Wien) 2010;152(5):793–802. [43] van Rooij WJ, Peluso JP, Sluzewski M, Beute GN. Additional value of 3D rotational angiography in angiographically negative aneurysmal subarachnoid hemorrhage: how negative is negative? AJNR Am J Neuroradiol 2008;29(5):962–6. [44] Gemmete JJ, Elias AE, Chaudhary N, Pandey AS. Endovascular methods for the treatment of intracranial cerebral aneurysms. Neuroimaging Clin N Am 2013;23(4): 563–91. [45] Rieger M, Czermak B, El Attal R, Sumann G, Jaschke W, Freund M. Initial clinical experience with a 64-MDCT whole-body scanner in an emergency department: better time management and diagnostic quality? J Trauma 2009;66(3):648–57. [46] Hand DJ. Evaluating diagnostic tests: the area under the ROC curve and the balance of errors. Stat Med 2010;29(14):1502–10. [47] Mak CM, Lam CW, Tam S. Diagnostic accuracy of serum ceruloplasmin in Wilson disease: determination of sensitivity and specificity by ROC curve analysis among ATP7B-genotyped subjects. Clin Chem 2008;54(8):1356–62. [48] Leeflang MM, Deeks JJ, Gatsonis C, Bossuyt PM, Cochrane Diagnostic Test Accuracy Working Group. Systematic reviews of diagnostic test accuracy. Ann Intern Med 2008;149(12):889–97.
Contents lists available at ScienceDirect
Journal of the Neurological Sciences journal homepage: www.elsevier.com/locate/jns
Meta-analysis of diagnostic significance of sixty-four-row multi-section computed tomography angiography and three-dimensional digital subtraction angiography in patients with cerebral artery aneurysm Wei Guo a,b, Xu-Ying He a, Xi-Feng Li a, Dong-Xiang Qian b, Jian-Quan Yan b, De-Lin Bu b, Chuan-Zhi Duan a,⁎ a b
Department of Neurosurgery, Zhujiang Hospital, Southern Medical University, Guangzhou 510282, PR China Department of Neurosurgery, The Third Affiliated Hospital of Guangzhou Medical University, Guangzhou 510150, PR China
a r t i c l e
i n f o
Article history: Received 2 August 2014 Received in revised form 15 August 2014 Accepted 17 August 2014 Available online 27 August 2014 Keywords: Sixty-four-row multi-section computed tomography angiography Three-dimensional digital subtraction angiography Cerebral artery aneurysm Diagnostic Odds ratio Meta-analysis
a b s t r a c t Objective: Imaging methods are essential in evaluating cerebral artery aneurysms and they have evolved with recent technical advances. Sixty-four-row multi-section computed tomography (64-MSCT) angiography and three-dimensional digital subtraction angiography (3D-DSA) are two of the most popular methods. We sought to systematically explore and find out which one would be better in imaging cerebral artery aneurysm, and try to investigate the potential use and value of 64-MSCT angiography and 3D-DSA in cerebral artery aneurysm. Method: Followed by a predefined comprehensive literature search, we carefully searched both English and Chinese electronic databases for potentially relevant studies following our meta-analysis. Two reviewers independently assessed the methodological quality of the included eligible trials based on quality assessment of studies of diagnostic accuracy studies (QUADAS). Pooled summary statistics for sensitivity, specificity, positive and negative likelihood ratios (positive LR and negative LR), and diagnostic odds ratio (ORs) with their 95% confidence intervals (CIs) were utilized. Results: Final meta-analysis of 923 cerebral artery aneurysm cases were incorporated from eight cohort studies and selected for statistical analysis. Pooled sensitivity and specificity of 64-MSCT angiography in the diagnosis of cerebral artery aneurysm were 0.97 (95% CI, 0.96–0.98) and 0.91 (0.88–0.94), respectively. The pooled positive LR was 7.68 (95% CI, 3.34–17.67); and the pooled negative LR was 0.04 (95% CI, 0.03–0.05). The pooled diagnostic OR was 263.69 (95% CI, 121.19–573.77). The area under the SROC curve was 0.9934 (standard error [SE] = 0.0031). No significant evidence of publication bias was detected (P N 0.05). Conclusion: The main finding of our meta-analysis revealed that 64-MSCT angiography relative to the 3D-DSA may have a high diagnostic accuracy for the cerebral artery aneurysm. Thus, 64-MSCT angiography may be an effective tool for the early detection of cerebral artery aneurysm. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Cerebral artery aneurysm, also named intracranial cerebral aneurysm, is widely known as one of the most common and serious cerebrovascular events [1]. A previous report in this field has demonstrated that chronic inflammation might have a negative influence on degeneration in the cerebral artery aneurysm wall, which will increase the potential risk of rupture rate of an aneurysm [2]. Furthermore, ruptured cerebral aneurysms, which vary with its morphological features and patient's medical history, is one of the leading causes of morbidity and fatality in patients with subarachnoid hemorrhage [3]. In the United States, a recent epidemiologic study has shown that among a population of 100, ⁎ Corresponding author at: Department of Neurosurgery, Zhujiang Hospital, Southern Medical University, No. 253, Industrial Avenue, Haizhu District, Guangzhou 510282, PR China. E-mail address: [email protected] (C.-Z. Duan).
http://dx.doi.org/10.1016/j.jns.2014.08.023 0022-510X/© 2014 Elsevier B.V. All rights reserved.
approximately 2 people might be affected by cerebral artery aneurysms, and approximately 15,000 patients will die of complications from ruptured cerebral artery aneurysms each year [4]. In addition, it has been reported that the incidence of cerebral artery aneurysm varies greatly across gender, with a higher prevalence in women and conversely a lower prevalence in men [5]. Meanwhile, investigations have found that the recurrence rate of cerebral artery aneurysms is higher than expected, which becomes a considerable challenge [6]. As for the etiology of cerebral artery aneurysms, it has been widely accepted to be induced by interaction between ambient and genetic factors [7]. A number of studies have shown that various factors such as chronic inflammation, hypertension, hemodynamic stress, arteriovenous malformations, and excessive alcohol consumption may be significantly related with an increased risk of cerebral artery aneurysms [1,5,8]. As we all know, a rapid radiological imaging evaluation of ruptured cerebral artery aneurysms can play an important role to the successful treatment of cerebral artery aneurysms [9]. To date, researchers have paid more attention to
198
W. Guo et al. / Journal of the Neurological Sciences 346 (2014) 197–203
spiral computed tomography and digital subtraction angiography, two popular and useful imaging methods, which are extensively used in neurovascular imaging [10,11]. Sixty-four-row multi-section computed tomography (64-MSCT) angiography, also named x-ray computed tomography angiography (CTA), can reconstruct tomographic images in a variety of directions in the human body via three-dimensional horizontal scan with the help of computer-processed x-rays [12]. The whole practical process is achieved by digital geometry processing, which forms and produces a three-dimensional image of the scanned object from a series of twodimensional radiographic images [13]. The usage of CTA has dramatically increased over the last two decades, especially in the United States, Europe, and Asia, and the main target crowd is children and adults [14,15]. This excellent technique is prevalent and widely applied in medical imaging for diagnostic and therapeutic goals [16]. For example, with regard to the application, 64-MSCT angiography can provide more accurate and detailed data for analyzing the internal and external structures of organs and tissues of the patients examined, such as patients with gastric cancer and perihilar cholangiocarcinoma [17,18]. Up to now, a great number of researchers have shown that CTA with the remarkable performance of rotational time of 0.6 s and slice thickness of 0.625 mm may be an alternative method for diagnosing and treating intracranial aneurysms [19,20]. To be more specific, imaging acquisition of 64-MSCT angiography requires only 1 minute or even less, and 64MSCT is well tolerated by the majority of patients with high-risk acute subarachnoid hemorrhage [21]. The main characteristics of 64-MSCT angiography are that it can depict intracranial aneurysms at different locations and with high sensitivity and specificity; then, its unique advantages include non-invasivity, simplicity, rapidity, and better visualization, which are the prime determinants for it to become the preferred and first-intention diagnostic technique [22]. On the other hand, three-dimensional digital subtraction angiography (3D-DSA), used in a bony or dense soft tissue environment, is regarded as a fluoroscopy technique assisted by radiology to obtain vascular images [23]. The operating process is to form a pre-contrast image from radiological equipment; then, images are acquired after injecting contrast medium into the blood vessels and removing distracting structures from the first image [24]. Therefore, it was widely accepted that DSA primarily aims to image blood vessels with high image quality (b 3 mm), results that are not possible using CTA imaging [25]. Although 3D-DSA is still considered as the gold standard for arterial imaging, especially for intracranial aneurysms, it is used less and less commonly and may be substituted by 64-MSCT angiography [26]. The reasons are that it is invasive, time-consuming, expensive, and stressful for patients with intracranial aneurysms [27]. Hence, previous large studies demonstrated and concluded that 64-MSCT angiography is a more appropriate and feasible alternative to diagnose intracranial aneurysms with significant advantages in the reduction of contrast medium dose and x-ray exposure time [21]. Yet inconsistent results were also found in other relevant documents [22,28]. In the current meta-analysis, the purpose of this study was to explore the utility of 3D-DSA and 64-MSCT angiography in describing the imaging characteristics of patients with cerebral artery aneurysms and have a potential judgment on which imaging detection method is more available and reliable to the present diagnosis of cerebral artery aneurysm. 2. Materials and methods
(“Multi-slice computed tomography angiography” or “Multi-slice CT angiography” or “MSCTA” or “MS-CTA” or “64-section multidetector CT angiography” or “MSCT” or “multislice computed tomography coronary angiography” or “64-section CT angiography” or “64-MSCT angiography” or “64-section CT scanner”) and (“intracranial aneurysm” or “intracranial aneurysm” or “intracranial aneurysms” or “Anterior Communicating Artery Aneurysm” or “Anterior Communicating Artery Aneurysms” or “Basilar Artery Aneurysms” or “Basilar Artery Aneurysm” or “Middle Cerebral Artery Aneurysms” or “Posterior Cerebral Artery Aneurysm” or “Posterior Cerebral Artery Aneurysms” or “Berry Aneurysm” or “Berry Aneurysms” or “Brain Aneurysm” or “Brain Aneurysms” or “Cerebral Aneurysm” or “Cerebral Aneurysms” or “Giant Intracranial Aneurysm” or “Giant Intracranial Aneurysms” or “Intracranial Mycotic Aneurysm” or “Intracranial Mycotic Aneurysms” or “Anterior Cerebral Artery Aneurysm” or “Anterior Cerebral Artery Aneurysms” or “Posterior Communicating Artery Aneurysm” or “Posterior Communicating Artery Aneurysms”) were entered in the database searches as medical subject heading terms and text words with a highly sensitive search strategy. Manual searches were used to screen other eligible studies. 2.2. Study selection After reading the abstracts, full papers were retrieved and assessed for their suitability with the following inclusion criteria: (1) only those cohort studies conducted within a human population to explore the relationship between 64-MSCT angiography and the diagnostic value of cerebral artery aneurysm were incorporated; (2) the diagnosis and management of patients with cerebral artery aneurysm should be confirmed by its clinical etiology and manifestations, and the combined diagnosis from the laboratory and imaging methods observing the occurrence and extent of subarachnoid hemorrhage (SAH) via CT or cerebral angiography, and magnetic resonance imaging [29,30]; (3) articles must be published in a peer-reviewed journal especially providing with original data. Corresponding major exclusion criteria were those experimental articles or trails that: (1) did not satisfy the above inclusion criteria designed in the current study; (2) abstracts, reviews, case report, letters, meta-analysis, or proceedings; (3) duplication publications or studies with overlapping data; and (4) subgroup analysis of the included trials. 2.3. Data extraction and quality assessment We used a standard reporting form to extract data from each included study, and the following descriptive information were collected: surname and initials of the first author, year of submission, country and racial descent, total numbers of included cases, gender information, mean age and floating range, demographic variables, 2 × 2 table information, and confirmation of diagnosis. Diagnostic results data were also counted in the research including the true positive (TP), false positive (FP), false negative (FN), and true negative (TN) information results in four-fold (2 × 2) tables for cerebral artery aneurysm diagnosis. The quality of involved studies was assessed independently by two authors based on a tool for the quality assessment of studies of diagnostic accuracy studies (QUADAS) [31]. Fourteen assessment items were implicated in the QUADAS criteria. Each of these items was scored as “yes” (2), “no” (0), or “unclear” (1). QUADAS scores ranged from 0 to 28; and a score of ≥22 indicates good quality.
2.1. Literature search 2.4. Statistical analysis The following computerized bibliographic databases were applied to identify relevant articles related to the association of 64-MSCT angiography with cerebral artery aneurysm diagnosis, with no restrictions on language or data collection: PubMed, Embase, CINAHL, and Science Citation Index, the Cochrane Library, Current Contents Index, Chinese Biomedical, the Chinese Journal Full-Text, and the Weipu Journal.
The association of the diagnostic significance of 64-MSCT angiography with cerebral artery aneurysm was judged by the positive likelihood ratio (positive LR) and negative likelihood ratio (negative LR) combined with its 95% confidence interval (95% CI). Data were first extracted to quantify heterogeneity among studies with Cochran's Q-
W. Guo et al. / Journal of the Neurological Sciences 346 (2014) 197–203
Identification
statistic (P b 0.05 was considered significant) and I2 tests in Meta-DiSc1 software [32]. Threshold effects were the main reason contributing to the existence of heterogeneity among all the included diagnostic trails, and P N 0.05 indicated no threshold effect and thus no significant evidence of heterogeneity was shown. Threshold effect was assessed using Spearman correlation coefficients. Meanwhile, with regard to the heterogeneity among different studies, χ2 test was applied (P b 0.05 or I2 test exhibiting N 50% indicated the occurrence of heterogeneity). Additionally, when the included studies were relatively small that they could not continue the exploration of between-studies heterogeneity within meta-analysis, we always chose the random-effects model; otherwise, when the P N 0.05 or I2 test exhibited b 50% (no apparent evidence of heterogeneity), a fixed-effects model may be picked. Furthermore, independent random-effects summary sensitivity and specificity estimations were weighted; diagnostic odds ratios (DOC) were also calculated, and based on the results, summary receiver operating characteristic (sROC) curves were concluded to represent the performance of the present diagnostic test [33]. The area under the receiver operating characteristic (ROC) curve (AUC) was calculated as a performance measure for machine detection significance [34]. P b 0.05 was considered as the difference and was statistically significant. Publication bias was conducted, using a scatter plot, mainly by a regression of diagnostic odds ratio (ORs) against 1/sqrt (effective sample size), weighting by effective sample size, with the usage of slope coefficient indicating asymmetry [35]. All statistical analyses were undertaken using the SAS statistical software version 8.2 (SAS Institute Inc., Cary, NC, United States) and Meta. Disc (version 1.4) statistical software for metaanalysis of testing accuracy data.
Articles identified through electronic database searching (N = 296)
3. Results 3.1. Selection of eligible studies Fig. 1 shows the flow chart of identified publications and the main reason for exclusion. Firstly, 298 potential articles were resulted from the electronic databases. Of the 298 articles, 1 study was a duplicate and was therefore removed. After screening of the titles and abstracts, 134 irrelevant studies were excluded subsequently. Then, 153 studies were excluded after more detailed full text assessment, and 10 studies were selected for the qualitative analysis. An additional 2 studies were removed after a further assessment. Finally, 8 cohort studies published from 2008 to 2013 were analyzed in the current meta-analysis [20–22,24,25,28,36,37].
3.2. Demographic variables The 8 included studies consisted of 923 cerebral artery aneurysm with a sample size ranging from 30 to 512. Of the 8 eligible studies, only two were from Caucasians [Serbia (Milosevic Medenica S.), and Belgium (Lubicz B.)], the remaining six studies were performed in Asians [China (Tian W.B., Wang H.S., Lu L., Xing W., and Ye H.W.) and Turkey (Chang J.W.)]. In addition, all of the included studies provided the baseline characteristics information. The basis of the included trials was divided into two biases, namely, per-patient and per-aneurysm basis. The baseline characteristics in the individual studies were summarized in Table 1.
Additional articles identified through a manual search (N = 2)
Screening
Articles reviewed for duplicates (N = 298)
Articles after duplicates removed (N = 297) Studies were excluded, due to: (N = 41) Letters, reviews, meta-analysis (N = 28) Not human studies (N = 62) Not related to research topics
Eligibility
Full-text articles assessed for eligibility (N = 163)
Studies included in qualitative synthesis (N = 10)
Included
199
Studies were excluded, due to: (N = 32) Not cohort study (N = 63) Not relevant to 64-MSCT and 3D-DSA (N = 58) Not relevant to cerebral artery aneurysm
Studies included in quantitative synthesis (meta-analysis) (N = 8)
Fig. 1. Flow chart of literature search and study selection. Eight cohort studies were included in this meta-analysis.
200
W. Guo et al. / Journal of the Neurological Sciences 346 (2014) 197–203
Table 1 The characteristics and methodological quality of the included studies in this meta-analysis. First author
Year
Country
Ethnicity
Number
Gender (M/F)
Age (years)
Basis
QUADAS score
Tian WB [37] Wang HS [28] Lu L [22]
2013 2012 2012
China China China
Asians Asians Asians
30 39 512
18/12 17/22 220/293
50.0 (38–76) 47.5 (41–71) 52.0 ± 12.0
22 2 27
Xing W [21]
2011
China
Asians
133
63/70
52.0 (9–87)
Milosevic Medenica S [24] Ye HW [36] McKinney AM [25] Lubicz B [20]
2010 2009 2008 2008
Serbia China Turkey Belgium
Caucasians Asians Asians Caucasians
47 42 66 54
40/7 18/24 35/31 37/17
54.3 (13–76) 27–67 54.5 (14–93) 55.0 (18–85)
Per-patient basis Per-aneurysm basis Per-aneurysm basis Per-patient basis Per-aneurysm basis Per-patient basis Per-aneurysm basis Per-patient basis Per-aneurysm basis Per-aneurysm basis
26 23 23 24 24
M, male; F, female; QUADAS, quality assessment of studies of diagnostic accuracy studies.
3.3. Test for heterogeneity In the meta-analysis, the relationship between diagnostic significance of 64-MSCT angiography and cerebral artery aneurysms were assessed for the observed heterogeneity. The random effects model was used due to the obvious existence of heterogeneity among studies. There was no significant association (I2 = 0, P = 0.8272) of sensitivity with specificity, which indicates the absence of the threshold effect. 3.4. Sensitivity and specificity analyses of 64-MSCT angiography Concluded from the present meta-analysis, the sensitivity and specificity analyses results of the diagnostic value of 64-MSCT angiography in cerebral artery aneurysm were 0.97 (95% CI, 0.96–0.98) and 0.91 (0.88–0.94), respectively. There were a total of 8 clinical trials conducted for the analysis, and the results of the meta-analysis based on the combined data indicated that up to five trials in the pooled specificity data were 100%. Random effects model results indicated that the pooled positive LR and negative LR were 7.68 (95% CI, 3.34–17.67) and 0.04 (95% CI, 0.03–0.05), respectively (Fig. 2). 3.5. SROC and AUC analysis The ROC curve was calculated for all observers and is shown in Fig. 3. In the current clinical analyses, the pooled diagnostic ORs of 64-MSCT angiography in cerebral artery aneurysms were 263.69 (95% CI, 121.19– 573.77). The results were plotted as a symmetrical SROC curve, and the corresponding AUC based on the 64-MSCT angiography diagnostic image of observers were 0.9934 (standard error [SE] = 0.0031). The above results indicated that there were significant values in 64-MSCT angiography in the diagnosis of trigeminal neuralgia, 3.6. Publication bias result Publication bias was assessed visually by using a scatter plot. It should hold a symmetrical funnel shape; otherwise, publication bias may be detected. Furthermore, we found no evidence of obvious asymmetry based on the symmetrical funnel shape in the present systemic analyses (P N 0.05; Fig. 4). 4. Discussion In the present meta-analysis, we systematically evaluated the technical performance and accuracy of 64-MSCT angiography and 3D-DSA in diagnosing cerebral artery aneurysms. Our results showed that the pooled sensitivity and specificity of 64-MSCT angiography in diagnosing cerebral artery aneurysms were 97% and 91%, respectively. These were consistent with the previous results, in which 64-MSCT angiography had a high diagnostic accuracy accompanied with a lower rate of missed diagnosis and misdiagnosis for cerebral artery aneurysms, suggesting that 64-MSCT angiography may be an effective tool for cerebral artery
aneurysm diagnosis. As a matter of fact, it has been suggested that ruptured cerebral artery aneurysms were one of the major reasons for the occurrence of spontaneous subarachnoid hemorrhage, and without timely treatment, approximately half of the patients could experience and possibly die from ruptured cerebral artery aneurysms [38]. Previously, DSA was confirmed as the “gold standard” in the diagnosis of cerebral artery aneurysms [39]. Nevertheless, due to the normal 2D imaging presentation, it may not be easily utilized for the observation of cerebral aneurysm, traditional imaging detection actually may result in a potential partial of false negative rate [40]. In recent decades, with the development of imaging detection technology, the updated 3DDSA based on the three-dimensional reconstruction of cerebral blood vessels can, to some extent, lead to a comprehensive observation of cerebral vascular from any respect, and hence would decrease the rate of misdiagnosis [25,41]. On the other side, widely acknowledged as an invasive examination method, and characterized by a long operation and examination times, there is no doubt that 3D-DSA would not be the optimal choice for cerebral artery aneurysm patients [42]. Methods and apparatus for a non-invasive examination of cerebral vascular for cerebral artery aneurysm patients therefore became popular. Our results showed that 64-MSCT angiography had a higher sensitivity and specificity for the detection of cerebral artery aneurysm, and the reasons for this phenomenon could be attributed to the fact that 64-MSCT angiography can itself reconstruct tomographic images in a variety of directions in the human body via 3D horizontal scan with the help of computer-processed x-rays [43]. 64-MSCT angiography and its unique advantages (non-invasivity, simplicity, rapidity, and better visualization as mentioned in the Introduction) have become the dominant determinants making 64-MSCT angiography the more popular and more suitable diagnostic technique for patients and hospital staff [44,45]. Additionally, the ROC curve is a broad overview of diagnostic tests, and is also considered a relative operating characteristic curve; to a large extent, it can be attributed that it is a comparison of two operating characteristics (true positive rate and false positive rate) when the criterion changes over time [46].The best possible prediction method would yield a point in the upper left corner or coordinate (0, 1) of the ROC space, representing 100% sensitivity (no false negatives) and 100% specificity (no false positives) [47]. The present results indicate that the corresponding AUC, based on the 64-MSCT angiography diagnostic image of observers, was 0.9934, revealing a significantly high efficacy of 64MSCT angiography in brain tumors. In addition, the diagnostic ORs measure the effectiveness of a diagnostic test, and the ratio ranges from zero to infinity; higher diagnostic odds ratios are indicative of better test performance; otherwise, it means a poor detected outcome [48]. Our findings demonstrated that the pooled diagnostic ORs of 64-MSCT angiography in cerebral artery aneurysm were 263.69, suggesting that 64-MSCT angiography holds a higher value in the diagnosis of brain tumors. Furthermore, due to the restriction of small sample size, heterogeneity tests failed to continue the following stratified analyses, which, in some way, have an adverse effect on the whole reliability of the outcomes.
W. Guo et al. / Journal of the Neurological Sciences 346 (2014) 197–203
201
Fig. 2. Forest plots for the diagnostic accuracy of 64-MSCT in cerebral artery aneurysm. (A) Sensitivity, (B) specificity, (C) positive likelihood ratio, and (D) negative likelihood ratio.
202
W. Guo et al. / Journal of the Neurological Sciences 346 (2014) 197–203
Log Odds Ratio versus 1/sqrt(Effective Sample Size)(Deeks) (Egger’s test: P > 0.05)
10
100
Study Regression Line
1
Diagnostic Odds Ratio
1000
Fig. 3. Forest plot of SROC curve and DOR on the diagnostic accuracy of 64-MSCT in cerebral artery aneurysm. SROC, summary receiver operator characteristic; DOR, diagnostic odds ratio; AUC, area under the curve; SE, standard error.
0.10
0.15
0.20
0.25
0.30
1/root(ESS) Fig. 4. Funnel plot of publication bias on the pooled DOR of 64-MSCT in cerebral artery aneurysm. No publication bias was detected in this meta-analysis. DOR, diagnostic odds ratio.
Although this is the first meta-analysis focused on the comparison of 64-MSCT angiography and 3D-DSA accuracy in the diagnosis and differential diagnosis of cerebral artery aneurysm, our study still has some limitations that merit a deeper consideration. Firstly, our results lacked sufficient statistical power to evaluate the accuracy of 64-MSCT angiography due to a relatively small sample size. Secondly, meta-analysis is a retrospective study that may lead to subject selection bias. Thirdly, this meta-analysis failed to obtain original data from the included studies, which limited further clinical assessment of 64-MSCT angiography for the diagnosis of cerebral artery aneurysm. Fourthly, we failed to find potential articles with an integrity dataset that focused on the comparison of diagnostic performance between 64-MSCT angiography and 3D-DSA accuracy for cerebral artery aneurysm. Importantly, the inclusion criteria of cases and controls were not well defined in all included studies and thus might have influenced our results. In conclusion, our findings provide empirical evidence that 64-MSCT angiography may have a high diagnostic accuracy for cerebral artery aneurysm. Thus, 64-MSCT angiography may be an effective tool for the early detection of cerebral artery aneurysms. However, due to the limitations mentioned above, further detailed studies are still needed to provide a more representative statistical analysis.
W. Guo et al. / Journal of the Neurological Sciences 346 (2014) 197–203
Conflict of interest We declare that we have no conflicts of interest.
Acknowledgments We would like to acknowledge the helpful comments on this paper received from our reviewers.
References [1] Aoki T, Nishimura M, Matsuoka T, Yamamoto K, Furuyashiki T, Kataoka H, et al. PGE(2)-EP(2) signalling in endothelium is activated by haemodynamic stress and induces cerebral aneurysm through an amplifying loop via NF-kappaB. Br J Pharmacol 2011;163(6):1237–49. [2] Hasan DM, Mahaney KB, Brown Jr RD, Meissner I, Piepgras DG, Huston J, et al. Aspirin as a promising agent for decreasing incidence of cerebral aneurysm rupture. Stroke 2011;42(11):3156–62. [3] Lin N, Cahill KS, Frerichs KU, Friedlander RM, Claus EB. Treatment of ruptured and unruptured cerebral aneurysms in the USA: a paradigm shift. J Neurointerv Surg 2012;4(3):182–9. [4] Villablanca JP, Duckwiler GR, Jahan R, Tateshima S, Martin NA, Frazee J, et al. Natural history of asymptomatic unruptured cerebral aneurysms evaluated at CT angiography: growth and rupture incidence and correlation with epidemiologic risk factors. Radiology 2013;269(1):258–65. [5] Caranci F, Briganti F, Cirillo L, Leonardi M, Muto M. Epidemiology and genetics of intracranial aneurysms. Eur J Radiol 2013;82(10):1598–605. [6] Hasan DM, Nadareyshvili AI, Hoppe AL, Mahaney KB, Kung DK, Raghavan ML. Cerebral aneurysm sac growth as the etiology of recurrence after successful coil embolization. Stroke 2012;43(3):866–8. [7] Bor AS, Rinkel GJ, van Norden J, Wermer MJ. Long-term, serial screening for intracranial aneurysms in individuals with a family history of aneurysmal subarachnoid haemorrhage: a cohort study. Lancet Neurol 2014;13(4):385–92. [8] Kelliny M, Maeder P, Binaghi S, Levivier M, Regli L, Meuli R. Cerebral aneurysm exclusion by CT angiography based on subarachnoid hemorrhage pattern: a retrospective study. BMC Neurol 2011;11:8. [9] Luo Z, Wang D, Sun X, Zhang T, Liu F, Dong D, et al. Comparison of the accuracy of subtraction CT angiography performed on 320-detector row volume CT with conventional CT angiography for diagnosis of intracranial aneurysms. Eur J Radiol 2012;81(1):118–22. [10] Yang L, Huang X, Duan S. Clinical application and technique of 64-slice spiral CT subtraction angiography in head and neck. Vasa 2012;41(1):27–33. [11] Sadatomo T, Yuki K, Migita K, Taniguchi E, Kodama Y, Kurisu K. Evaluation of relation among aneurysmal neck, parent artery, and daughter arteries in middle cerebral artery aneurysms, by three-dimensional digital subtraction angiography. Neurosurg Rev 2005;28(3):196–200. [12] Kwok HC, Dirkzwager I, Duncan DS, Gillham MJ, Milne DG. The accuracy of multidetector computed tomography in the diagnosis of non-occlusive mesenteric ischaemia in patients after cardiovascular surgery. Crit Care Resusc 2014;16(2): 90–5. [13] Kropman RH, Zandvoort HJ, Van Den Heuvel DA, Wille J, Moll FL, De Vries JP. CT angiography to evaluate hemodynamic changes in popliteal artery aneurysms during flexion and extension of the knee joint. J Cardiovasc Surg (Torino) 2014;55(2 Suppl 1):249–53. [14] Furlow B. Radiation dose in computed tomography. Radiol Technol 2010;81(5): 437–50. [15] Smith-Bindman R, Lipson J, Marcus R, Kim KP, Mahesh M, Gould R, et al. Radiation dose associated with common computed tomography examinations and the associated lifetime attributable risk of cancer. Arch Intern Med 2009;169(22):2078–86. [16] Mileto A, Marin D, Ramirez-Giraldo JC, Scribano E, Krauss B, Mazziotti S, et al. Accuracy of contrast-enhanced dual-energy MDCT for the assessment of iodine uptake in renal lesions. AJR Am J Roentgenol 2014;202(5):W466–74. [17] Mehmedovic A, Mesihovic R, Saray A, Vanis N. Gastric cancer staging: EUS and CT. Med Arch 2014;68(1):34–6. [18] Ajiki T, Fukumoto T, Ueno K, Okazaki T, Matsumoto I, Ku Y. Three-dimensional computed tomographic cholangiography as a novel diagnostic tool for evaluation of bile duct invasion of perihilar cholangiocarcinoma. Hepatogastroenterology 2013; 60(128):1833–8. [19] Chen W, Yang Y, Xing W, Qiu J, Peng Y. Sixteen-row multislice computed tomography angiography in the diagnosis and characterization of intracranial aneurysms: comparison with conventional angiography and intraoperative findings. J Neurosurg 2008;108(6):1184–91. [20] Lubicz B, Levivier M, Francois O, Thoma P, Sadeghi N, Collignon L, et al. Sixty-fourrow multisection CT angiography for detection and evaluation of ruptured intracranial aneurysms: interobserver and intertechnique reproducibility. AJNR Am J Neuroradiol 2007;28(10):1949–55. [21] Xing W, Chen W, Sheng J, Peng Y, Lu J, Wu X, et al. Sixty-four-row multislice computed tomographic angiography in the diagnosis and characterization of intracranial aneurysms: comparison with 3D rotational angiography. World Neurosurg 2011; 76(1–2):105–13.
203
[22] Lu L, Zhang LJ, Poon CS, Wu SY, Zhou CS, Luo S, et al. Digital subtraction CT angiography for detection of intracranial aneurysms: comparison with three-dimensional digital subtraction angiography. Radiology 2012;262(2):605–12. [23] Amans MR, Cooke DL, Vella M, Dowd CF, Halbach VV, Higashida RT, et al. Contrast staining on CT after DSA in ischemic stroke patients progresses to infarction and rarely hemorrhages. Interv Neuroradiol 2014;20(1):106–15. [24] Milosevic Medenica S, VV V, Prstojevic B. 64-Slice CT angiography in the detection of intracranial aneurysms: comparison with DSA and surgical findings. Neuroradiol J 2010;23(1):55–61. [25] McKinney AM, Palmer CS, Truwit CL, Karagulle A, Teksam M. Detection of aneurysms by 64-section multidetector CT angiography in patients acutely suspected of having an intracranial aneurysm and comparison with digital subtraction and 3D rotational angiography. AJNR Am J Neuroradiol 2008;29(3):594–602. [26] Herzig R, Burval S, Krupka B, Vlachova I, Urbanek K, Mares J. Comparison of ultrasonography, CT angiography, and digital subtraction angiography in severe carotid stenoses. Eur J Neurol 2004;11(11):774–81. [27] Tacher V, Lin M, Desgranges P, Deux JF, Grunhagen T, Becquemin JP, et al. Image guidance for endovascular repair of complex aortic aneurysms: comparison of two-dimensional and three-dimensional angiography and image fusion. J Vasc Interv Radiol 2013;24(11):1698–706. [28] Wang HS, Zhao PL, Wang CQ, Yang ZW, Lv GS, Li QC, et al. Comparative study of 64 rows helical CT angiography and 3D-DSA in diagnosis of intracranial aneurysms. Hebei Med J 2012;34(11):1613–5. [29] Schwartz RB, Tice HM, Hooten SM, Hsu L, Stieg PE. Evaluation of cerebral aneurysms with helical CT: correlation with conventional angiography and MR angiography. Radiology 1994;192(3):717–22. [30] Vermeulen M, van Gijn J. The diagnosis of subarachnoid haemorrhage. J Neurol Neurosurg Psychiatry 1990;53(5):365–72. [31] Whiting PF, Weswood ME, Rutjes AW, Reitsma JB, Bossuyt PN, Kleijnen J. Evaluation of QUADAS, a tool for the quality assessment of diagnostic accuracy studies. BMC Med Res Methodol 2006;6:9. [32] Zintzaras E, Ioannidis JP. HEGESMA: genome search meta-analysis and heterogeneity testing. Bioinformatics 2005;21(18):3672–3. [33] Hanley JA, McNeil BJ. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology 1982;143(1):29–36. [34] Hanley JA, McNeil BJ. A method of comparing the areas under receiver operating characteristic curves derived from the same cases. Radiology 1983;148(3): 839–43. [35] Deeks JJ, Macaskill P, Irwig L. The performance of tests of publication bias and other sample size effects in systematic reviews of diagnostic test accuracy was assessed. J Clin Epidemiol 2005;58(9):882–93. [36] Ye HW, Li HW, Song Y, Zhou YS, Wu LG, Yu HT, et al. Early diagnosis of intracranial aneurysm using 64-slice spiral CT angiography: a clinical analysis. Chin J Neuromed 2009;08(06):578–80. [37] Tian WB, Zhang XW, Fan BS. Comparative study of 64 rows helical CT angiography and DSA in diagnosis of intracranial aneurysms. Chin J Pract Nerv Dis 2013; 16(09):69–70. [38] Westerlaan HE, van Dijk JM, Jansen-van der Weide MC, de Groot JC, Groen RJ, Mooij JJ, et al. Intracranial aneurysms in patients with subarachnoid hemorrhage: CT angiography as a primary examination tool for diagnosis–systematic review and meta-analysis. Radiology 2011;258(1):134–45. [39] Romijn M, Gratama van Andel HA, van Walderveen MA, Sprengers ME, van Rijn JC, van Rooij WJ, et al. Diagnostic accuracy of CT angiography with matched mask bone elimination for detection of intracranial aneurysms: comparison with digital subtraction angiography and 3D rotational angiography. AJNR Am J Neuroradiol 2008;29(1):134–9. [40] Kucukay F, Okten RS, Tekiner A, Dagli M, Gocek C, Bayar MA, et al. Threedimensional volume rendering digital subtraction angiography in comparison with two-dimensional digital subtraction angiography and rotational angiography for detecting aneurysms and their morphological properties in patients with subarachnoid hemorrhage. Eur J Radiol 2012;81(10):2794–800. [41] Toyota S, Iwaisako K, Takimoto H, Yoshimine T. Intravenous 3D digital subtraction angiography in the diagnosis of unruptured intracranial aneurysms. AJNR Am J Neuroradiol 2008;29(1):107–9. [42] Gerardin E, Tollard E, Derrey S, Langlois O, Dacher JN, Douvrin F, et al. Usefulness of multislice computerized tomographic angiography in the postoperative evaluation of patients with clipped aneurysms. Acta Neurochir (Wien) 2010;152(5):793–802. [43] van Rooij WJ, Peluso JP, Sluzewski M, Beute GN. Additional value of 3D rotational angiography in angiographically negative aneurysmal subarachnoid hemorrhage: how negative is negative? AJNR Am J Neuroradiol 2008;29(5):962–6. [44] Gemmete JJ, Elias AE, Chaudhary N, Pandey AS. Endovascular methods for the treatment of intracranial cerebral aneurysms. Neuroimaging Clin N Am 2013;23(4): 563–91. [45] Rieger M, Czermak B, El Attal R, Sumann G, Jaschke W, Freund M. Initial clinical experience with a 64-MDCT whole-body scanner in an emergency department: better time management and diagnostic quality? J Trauma 2009;66(3):648–57. [46] Hand DJ. Evaluating diagnostic tests: the area under the ROC curve and the balance of errors. Stat Med 2010;29(14):1502–10. [47] Mak CM, Lam CW, Tam S. Diagnostic accuracy of serum ceruloplasmin in Wilson disease: determination of sensitivity and specificity by ROC curve analysis among ATP7B-genotyped subjects. Clin Chem 2008;54(8):1356–62. [48] Leeflang MM, Deeks JJ, Gatsonis C, Bossuyt PM, Cochrane Diagnostic Test Accuracy Working Group. Systematic reviews of diagnostic test accuracy. Ann Intern Med 2008;149(12):889–97.
