Figures
Abstract
Background and Purpose
Accurate measurement of aortic annulus diameter is crucial for choosing suitable prosthetic size for transcatheter aortic valve implantation (TAVI). Several imaging methods are available for the measurement, but significant variability between different modalities has been observed. The purpose of this study was to systematically compare the measurements of aortic annulus diameter between multi-detector computed tomography (MDCT), transthoracic echocardiography (TTE), and transesophegeal echocardiography (TEE).
Methods
PubMed and EMBASE databases between January 2000 and January 2012 were searched. We extracted data from eligible studies evaluating the aortic annulus diameter by MDCT and echocardiography (TTE, TEE, or both). We performed a random-effects meta-analysis to calculate the weighted mean differences of aortic annulus diameter measurement between MDCT, TTE, and TEE.
Results
A total of 10 eligible studies involving 581 subjects with aortic valve stenosis were included. Aortic annulus diameter measured on coronal view by MDCT (25.3±0.52 mm) was respectively larger than that measured on sagittal view by MDCT (22.7±0.37 mm), TTE (22.6±0.28 mm), and TEE (23.1±0.32 mm). The weighted mean difference of aortic annulus diameter between coronal view by MDCT and TTE these two methods was 2.97 mm, followed by the weighted mean difference of 2.53 mm between coronal view and sagittal view by MDCT, and the mean difference of 1.74 mm between coronal view on MDCT and TEE (P<0.0001 for all). The weighted mean difference of aortic annulus diameter measurement between TEE and TTE was significant but somewhat small (0.45 mm, P = 0.007).
Conclusion
Aortic annulus diameter measured on coronal view by MDCT was robustly and significantly larger than that obtained on sagittal view by MDCT, TTE, or TEE. Such variability of aortic annulus diameter measurement by different imaging modalities cannot be ignored when developing optimal strategies for selection of prosthetic valve size in TAVI.
Citation: Zhang R, Song Y, Zhou Y, Sun L (2013) Comparison of Aortic Annulus Diameter Measurement between Multi-Detector Computed Tomography and Echocardiography: A Meta-Analysis. PLoS ONE 8(3): e58729. https://doi.org/10.1371/journal.pone.0058729
Editor: Yiqing Song, Brigham & Women’s Hospital, and Harvard Medical School, United States of America
Received: November 20, 2012; Accepted: February 5, 2013; Published: March 14, 2013
Copyright: © 2013 Zhang et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: The authors have no support or funding to report.
Competing interests: The authors have declared that no competing interest exist.
Introduction
Transcatheter aortic valve implantation (TAVI) has recently become an alternative to surgical aortic replacement in patients with severe aortic valve stenosis (AS), who are at a high surgical risk or contraindication to conventional aortic valve replacement surgery [1], [2], [3]. For TAVI, accurate measurement of aortic annulus diameter is crucial for choosing suitable prosthetic size. Aortic annulus diameter has been evaluated usually by transthoracic echocardiography (TTE) and transesophegeal echocardiography (TEE). By providing multiple-dimensional information of anatomical shape, multi-detector computed tomography (MDCT) has been also applied for accurate assessment of aortic annulus. Previous studies have assessed its performance by conducting simultaneous assessment of aortic annulus by TTE, TEE and MDCT in AS patients [4], [5] and showed substantial variability in the measurements of aortic annulus diameter between those modalities across studies. For instance, the mean difference of aortic annulus diameter could be as large as 4.1 mm between TTE and MDCT [4] and ranged from 1.4 mm to 4.0 mm between TTE and TEE [6]. Furthermore, some studies showed that aortic annulus diameter measured by MDCT was significantly larger than that measured by TTE or TEE [7], [8] whereas some reported the largest aortic annulus detected by TEE [6]. In clinical practice, a slight difference in the measurement of aortic annulus diameter has a significant impact on valve sizing recommendations and even the procedure decisions [9]. Choosing a larger or smaller size of prosthetic valve could eventually lead to paravalvular leakage, valve embolization, and prosthetic mismatch after procedure [10]. In the absence of gold standard for aortic valve sizing, it is important to learn the sources of variability in the measurement of aortic annulus diameter between different imaging modalities. To provide useful information to help therapeutic decision making, we therefore conducted a systematic review and meta-analysis to compare the overall mean differences in aortic annulus diameter measurement between TTE, TEE and MDCT in AS patients.
Materials and Methods
Search Strategy
We searched the PubMed and EMBASE databases between January 2000 and January 2012 with the following keywords: “aortic annulus”, “transthoracic echocardiography”, “transesophegeal echocardiography”, “multi-detector computed tomography”, “transcatheter aortic valve implantation” and “aortic valve stenosis”. The search was restricted to human studies published in English. References of review articles were also searched. Selected studies were individually examined to exclude studies with potentially duplicate and overlapping data. Abstracts without full-text publication, reviews, editorials or letters were excluded.
Study Selection and Data Extraction
Two reviewers (Zhang R and Sun L) independently searched and retrieved the full text of relevant articles. Discrepancy was resolved by group discussion. In each study, the TTE, TEE, and MDCT measurements of aortic annulus diameter were based on the visualization of the most caudal leaflet hinge points. For MDCT, we chose three widely-performed views -sagittal view, coronal view or three chamber view- on which aortic annulus diameter could be directly measured. Our inclusion criteria of this meta-analysis were listed as follows: (a) studies evaluating the aortic annulus diameter by MDCT and echocardiography (TTE, TEE, or both) were included; (b) the use of CT scanners with a minimum of 16 detector rows; and (c) the aortic annulus diameter measured by MDCT were on sagittal view, coronal view or three chamber view. Articles that did not present relevant data for meta-analysis were not included.
Information on the first author, year of publication, study population, sample size, mean age, background diseases, and methods of aortic annulus measurement was extracted from each included study. The means and standard deviations of aortic annulus diameter by TTE, TEE, and on coronal view, sagittal view or three chamber view by MDCT were obtained from the text, tables, or graphs in each article. For studies with unavailable or incomplete information, the corresponding authors were contacted on 3 occasions over 1 month to complete the missing information. Two of the 6 contacted authors provided the additional data.
In addition, only two studies had a standard surgical gauge measurement of aortic annulus during surgical aortic valve replacement. We also extracted surgical measurement data from those studies in secondary analysis.
Data Synthesis and Statistical Analysis
Meta-analysis was performed using DerSimonian and Laird’s random-effects model in which each study is weighted by the inverse of the sum of within-study plus between-study variance [11]. The weighted mean differences and 95% confidence intervals of aortic annulus between TTE, TEE, and MDCT were calculated. Results are presented in figures and tables, with the mean and confidence interval for each study being presented, as well as the summary values obtained from the meta-analysis. Between-study heterogeneity was assessed using Cochrane’s Q test and the I2 test [12]. The percentages of I2 around 25% (I2 = 25), 50% (I2 = 50), and 75% (I2 = 75) indicate low, medium, and high heterogeneity, respectively [12]. We assessed publication bias using visual inspection of Begg’s modified funnel plots, in which the mean difference was plotted against its standard error from each study. Publication bias was also assessed by Begg’s adjusted rank correlation test [13] and Egger’s regression asymmetry test [14]. All analyses were performed using the STATA statistical software (version 10.1, STATA Corp., College Station, Texas). P-values of less than 0.05 were considered statistically significant.
Results
In total, 619 titles were screened for relevance, of which 16 articles were considered eligible, including 10 studies with available data (Figure 1). Hand-search yielded 3 eligible articles. All individuals in these studies (n = 581) were AS patients. Of these 10 studies [5], [9], [15], [16], [17], [18], [19], [20], [21], [22], 8 studies performed both TTE and MDCT [5], [9], [15], [17], [19], [20], [21], [22]and other 8 studies performed both TEE and MDCT [5], [9], [15], [16], [18], [19], [20], [22]. Table 1 shows basic characteristics of the 10 included studies.
As shown in Table 2, aortic annulus diameter measured on coronal view by MDCT (25.3±0.52 mm) was significantly larger than that measured on sagittal view by MDCT (22.7±0.37 mm), TTE (22.6±0.28 mm), or TEE (23.1±0.32 mm). MDCT measurement on coronal view showed the largest aortic annulus diameter and TTE measurement showed the smallest aortic annulus diameter. The weighted mean difference of aortic annulus diameter between these two methods was 2.97 mm (95% CI, 1.98 to 3.96) (Figure 2), followed by that between coronal view and sagittal view by MDCT (2.53 mm; 95% CI, 1.38 to 3.68 mm) (Figure 3), and that between coronal view by MDCT and TEE (1.74 mm; 95% CI, 1.30 to 2.19 mm) (Figure 4) (P<0.0001 for all). The mean difference of aortic annulus measurement between TEE and TTE was slight (0.45 mm, P = 0.007) (Figure 5). The mean differences between TTE and sagittal view, TTE and three chamber view, TEE and sagittal view, TEE and three chamber view were not significant (P>0.05 for all).
Significant between study heterogeneity were observed for the weighted mean difference of aortic annulus diameter measurement between TTE and coronal view on MDCT (I2 = 79%; P for Q test <0.001) and that between coronal view and sagittal view on MDCT (I2 = 90%; P for Q test <0.001) (Table 2). We conducted a sensitivity analysis to assess the extent to which individual studies with extremely large mean difference on the weighted mean difference. As a result, Tzikas’s study contributed to most of between study heterogeneity [21]. The exclusion of this study appreciably reduced between-study heterogeneity and the final meta-analysis results; the weighted mean differences changed from 2.97 mm (1.98 to 3.96) (I2 = 79% and P for Q test <0.001) to 2.50 mm (1.83 to 3.16) (I2 = 44% and P for Q test = 0.13) for MDCT coronal view vs. TTE and from 2.53 mm (1.38 to 3.68) (I2 = 90% and P for Q test <0.001) to 2.06 mm (1.36–2.76) (I2 = 67% and P for Q test = 0.006) for MDCT coronal view vs. sagittal view. Neither did omitting each study from the remaining studies led to almost the same weighted mean differences without substantial changes in between-study heterogeneity.
For aortic annulus diameter measurement between coronal view and sagittal view on MDCT, the Begg’s funnel plot for the visual assessment of publication bias showed that larger mean difference in Tzikas’s study tended to be above the horizontal line, indicating a possibility of publication bias in favor of large difference. For all mean difference measurements, neither the Egger test nor the Begg test showed evidence of publication bias (all p>0.05).
To provide additional information on the accuracy of aortic annulus diameter, we further compared the surgical measurement with those by other imaging modalities, although available data were limited. As a result, the largest mean difference was 2.96 mm between and coronal view on MDCT, followed by the mean difference of 1.40 mm between surgical measurement and sagittal view on MDCT. The smallest mean difference was 1.09 mm between surgical measurement and TEE. Due to small sample sizes, the results should be very suggestive.
Discussion
Our meta-analysis quantitatively assessed the difference of aortic annulus diameter measurement between MDCT, TTE, and TEE. By synthesizing available data from eligible studies, our results showed that aortic annulus diameter measured on coronal view by MDCT was consistently and significantly larger than that measured on sagittal view by MDCT and that by TTE or TEE. The largest mean difference was seen between coronal view and TTE. Our study provided empirical data for the importance of optimizing the assessment of annular measurement for better selection of valve size for TAVI. Such robust evidence for systematic variability of the aortic annulus measurements between TTE, TEE, and MDCT should be taken into account for choosing prosthetic size.
The largest aortic annulus diameter measured on coronal view by MDCT has been reported by several studies [8], [15], [21]. By achieving more power, our meta-analysis confirmed the result by presenting the mean difference of 2.96 mm between coronal view and TTE. Using two dimensional (2D) TTE and TEE, the aortic annulus diameter is derived from a single diameter measurement on the basis of the assumption of a circular geometry. MDCT could provide three dimensional information of aortic annulus, but it is limited by the radiation exposure and the administration of a contrast agent. From anatomical view, the structure of aortic annulus is not circular but oval. The ellipsoid nature of the aortic annulus is a cause of measurement bias of its diameter assessment and mainly leads to a larger diameter on coronal view than that on sagittal view. The oval shape of the aortic annulus also largely explains the fact that the annulus diameter on the coronal view was consistently larger (from 1.3 mm to 4.0 mm) than that on the sagittal view [8], [22] and that on TEE, because the sagittal view on MDCT is a somewhat similar view with parasternal long-axis view on TTE and long-axis view of the aortic annulus on TEE. The three-dimensional reconstruction of the MDCT images may also contribute to the source of the measured difference. The most important advantage of MDCT is to reconstruct images at any level or plane; however, given the oval shape of the aortic annulus, a minor change of orientation also could lead to significant difference. This process could lead to disparate results between different views on MDCT, and between echocardiography and MDCT. In addition, we could not compare the aortic annulus diameter between the coronal view and three chamber view, due to the absence of relevant data. It seems reasonable to speculate that aortic annulus diameter on coronal view should be larger than that on three chamber view, because three chamber view has the exact same orientation as the parasternal long-axis view on TTE.
Although a 4 mm difference in the annulus diameter measurement between TTE and TEE was shown by one study [6], most studies described the excellent agreement between TTE and TEE [5], [9]. In the present study, we observed no significant differences between TTE and TEE, TTE and sagittal view, TTE and three chamber view, TEE and sagittal view, TEE and three chamber view. The absence of significant difference between those modalities could come from either similar reproducibility of MDCT with TTE and TEE [23], or the small variation of different imaging plane. Both of them can be the potential influencing factor for accurate aortic annulus diameter measurement. In our meta-analysis, between-study heterogeneity in mean differences of aortic annulus diameter was observed but seemed to be influenced by the results from one single study. After excluding this study, the results remained unchanged although between-study heterogeneity disappeared.
Obviously, the differences in aortic annulus measurement between the modalities can cause confusion among operators as to choosing appropriate valve size for TAVI. From clinical perspectives, the method allowing a safe procedure with a low rate of complications should be the best. Koos et al [18] reported the different measurements between TEE and MDCT changed TAVI strategy in a relevant number of patients (22% to 24%). However, it remains uncertain which diameter is more accurate for predicting prosthetic size for TAVI. The results on the agreement of single measurement with final prosthetic size have been mixed [16], [17], [19], [24]. For example, recent studies indicated that TTE measurements might result in undersizing of prosthesis, which is an important determinant for paravalvular aortic regurgitation after TAVI [25], [26]. With the controversial results, Gurvitch R et al [23] used the mean difference between the MDCT measurements and TEE measurements to calculate an “adjusted” sizing model, which reduced the alteration of TAVI by 10% to 12% when unadjusted aortic annulus diameter by MDCT altered the TAVI strategy by 42% to 44%. They also showed that in the cases with residual discrepancies between MDCT and TEE that exceeded these criteria, the frequency of moderate paravalvular aortic regurgitation following TAVI was very high, although future prospective studies are needed to assess the importance and usefulness of the “adjusted” criteria.
In our study, aortic annulus diameter on TEE had the smallest difference with the intra-operative aortic annulus diameter measured by surgical standard method, and coronal view on MDCT showed the biggest mean difference with surgical measurement. However, surgical measurement is made in arrested heart and after excision of the native aortic valve. In percutaneous procedures, the replacement valve is implanted over the native annulus, which is highly calcified in AS patients. Therefore, we could not completely rely on surgical result to evaluate the accuracy of different methods for choosing prosthetic size in TAVI. Although aortic annulus diameter measured by surgical method could not be used as a gold standard, it still is a reflection of the true size of aortic annulus to some degrees.
This study had some limitations. First, aortic annulus diameter measurement on three chamber view by MDCT only was available in 2 studies. At this point, the study power is relatively inadequate to assess the differences between this view and other imaging methods. Second, we could not evaluate the accuracy of TTE, TEE, and MDCT, because the data of the prosthesis size based on good clinical outcomes after TAVI were often absent. Third, measurements of aortic annulus diameter were not made in the same cardiac phase. However, it seems feasible because previous studies showed no significant differences in the diameter of aortic annulus between diastole and systole [27]. Finally, publication bias is evitable in any meta-analysis. Nevertheless, the consistency of the results in almost all included studies indicates that the lack of unpublished data may not substantially affect the pooled results.
In conclusion, aortic annulus diameter measured on coronal view by MDCT was significantly larger than that measured on sagittal view by MDCT and by TTE, TEE. The apparent discrepancy of aortic annulus diameter measurement exists between TTE, TEE and MDCT, and the differences should be taken in to account for choosing prosthetic valve size in TAVI.
Author Contributions
Conceived and designed the experiments: RZ YS YZ LS. Performed the experiments: RZ YS YZ LS. Analyzed the data: RZ YS YZ LS. Contributed reagents/materials/analysis tools: RZ YS YZ LS. Wrote the paper: RZ YS YZ LS.
References
- 1. Cribier A, Eltchaninoff H, Bash A, Borenstein N, Tron C, et al. (2002) Percutaneous transcatheter implantation of an aortic valve prosthesis for calcific aortic stenosis: first human case description. Circulation 106: 3006–3008.
- 2. Leon MB, Smith CR, Mack M, Miller DC, Moses JW, et al. (2010) Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery. N Engl J Med 363: 1597–1607.
- 3. Helton TJ, Kapadia SR, Tuzcu EM (2011) Clinical trial experience with transcatheter aortic valve insertion. Int J Cardiovasc Imaging 27: 1143–1154.
- 4. Jabbour A, Ismail TF, Moat N, Gulati A, Roussin I, et al. (2011) Multimodality imaging in transcatheter aortic valve implantation and post-procedural aortic regurgitation: comparison among cardiovascular magnetic resonance, cardiac computed tomography, and echocardiography. J Am Coll Cardiol 58: 2165–2173.
- 5. Messika-Zeitoun D, Serfaty JM, Brochet E, Ducrocq G, Lepage L, et al. (2010) Multimodal assessment of the aortic annulus diameter: implications for transcatheter aortic valve implantation. J Am Coll Cardiol 55: 186–194.
- 6. Moss RR, Ivens E, Pasupati S, Humphries K, Thompson CR, et al. (2008) Role of echocardiography in percutaneous aortic valve implantation. JACC Cardiovasc Imaging 1: 15–24.
- 7. Ng AC, Delgado V, van der Kley F, Shanks M, van de Veire NR, et al. (2010) Comparison of aortic root dimensions and geometries before and after transcatheter aortic valve implantation by 2- and 3-dimensional transesophageal echocardiography and multislice computed tomography. Circ Cardiovasc Imaging 3: 94–102.
- 8. Tops LF, Wood DA, Delgado V, Schuijf JD, Mayo JR, et al. (2008) Noninvasive evaluation of the aortic root with multislice computed tomography implications for transcatheter aortic valve replacement. JACC Cardiovasc Imaging 1: 321–330.
- 9. Hutter A, Opitz A, Bleiziffer S, Ruge H, Hettich I, et al. (2010) Aortic annulus evaluation in transcatheter aortic valve implantation. Catheter Cardiovasc Interv 76: 1009–1019.
- 10. Webb JG, Pasupati S, Humphries K, Thompson C, Altwegg L, et al. (2007) Percutaneous transarterial aortic valve replacement in selected high-risk patients with aortic stenosis. Circulation 116: 755–763.
- 11. DerSimonian R, Laird NM (1986) Meta-analysis in clinical trials. Control Clin Trials 7: 177–188.
- 12. Hardy RJ, Thompson SG (1998) Detecting and describing heterogeneity in meta-analysis. Stat Med 17: 841–856.
- 13. Begg CB, Mazumdar M (1994) Operating characteristics of a rank correlation test for publication bias. Biometrics 50: 1088–1101.
- 14. Egger M, Davey Smith G, Schneider M, Minder C (1997) Bias in meta-analysis detected by a simple, graphical test. BMJ 315: 629–634.
- 15. Altiok E, Koos R, Schroder J, Brehmer K, Hamada S, et al. (2011) Comparison of two-dimensional and three-dimensional imaging techniques for measurement of aortic annulus diameters before transcatheter aortic valve implantation. Heart 97: 1578–1584.
- 16. Dashkevich A, Blanke P, Siepe M, Pache G, Langer M, et al. (2011) Preoperative assessment of aortic annulus dimensions: comparison of noninvasive and intraoperative measurement. Ann Thorac Surg 91: 709–714.
- 17. Delgado V, Ng AC, van de Veire NR, van der Kley F, Schuijf JD, et al. (2010) Transcatheter aortic valve implantation: role of multi-detector row computed tomography to evaluate prosthesis positioning and deployment in relation to valve function. Eur Heart J 31: 1114–1123.
- 18. Koos R, Altiok E, Mahnken AH, Neizel M, Dohmen G, et al. (2011) Evaluation of aortic root for definition of prosthesis size by magnetic resonance imaging and cardiac computed tomography: Implications for transcatheter aortic valve implantation. Int J Cardiol 158: 353–358.
- 19. Mesa Rubio D, Suarez de Lezo Cruz Conde J, Alvarez-Osorio MP, Ruiz Ortiz M, Delgado Ortega M, et al. (2011) Measurement of aortic valve annulus using different cardiac imaging techniques in transcatheter aortic valve implantation: agreement with finally implanted prosthesis size. Echocardiography 28: 388–396.
- 20. Mizia-Stec K, Pysz P, Jasinski M, Adamczyk T, Drzewiecka-Gerber A, et al. (2012) Preoperative quantification of aortic valve stenosis: comparison of 64-slice computed tomography with transesophageal and transthoracic echocardiography and size of implanted prosthesis. Int J Cardiovasc Imaging 28: 343–352.
- 21. Tzikas A, Schultz CJ, Piazza N, Moelker A, Van Mieghem NM, et al. (2011) Assessment of the aortic annulus by multislice computed tomography, contrast aortography, and trans-thoracic echocardiography in patients referred for transcatheter aortic valve implantation. Catheter Cardiovasc Interv 77: 868–875.
- 22. Wood DA, Tops LF, Mayo JR, Pasupati S, Schalij MJ, et al. (2009) Role of multislice computed tomography in transcatheter aortic valve replacement. Am J Cardiol 103: 1295–1301.
- 23. Gurvitch R, Webb JG, Yuan R, Johnson M, Hague C, et al. (2011) Aortic annulus diameter determination by multidetector computed tomography: reproducibility, applicability, and implications for transcatheter aortic valve implantation. JACC Cardiovasc Interv 4: 1235–1245.
- 24. Sherif MA, Abdel-Wahab M, Stocker B, Geist V, Richardt D, et al. (2010) Anatomic and procedural predictors of paravalvular aortic regurgitation after implantation of the Medtronic CoreValve bioprosthesis. J Am Coll Cardiol 56: 1623–1629.
- 25. Detaint D, Lepage L, Himbert D, Brochet E, Messika-Zeitoun D, et al. (2009) Determinants of significant paravalvular regurgitation after transcatheter aortic valve: implantation impact of device and annulus discongruence. JACC Cardiovasc Interv 2: 821–827.
- 26. Zajarias A, Cribier AG (2009) Outcomes and safety of percutaneous aortic valve replacement. J Am Coll Cardiol 53: 1829–1836.
- 27. LaBounty TM, Sundaram B, Agarwal P, Armstrong WA, Kazerooni EA, et al. (2008) Aortic valve area on 64-MDCT correlates with transesophageal echocardiography in aortic stenosis. AJR Am J Roentgenol 191: 1652–1658.