Comparison of postoperative outcomes following multidetector computed tomography based vs transesophageal echocardiography based annulus sizing for transcatheter aortic valve replacement: A systematic review and meta‐analysis

Abstract Background The purpose of this paper was to evaluate the difference in postoperative outcomes following multidetector computed tomography (MDCT) and transesophageal echocardiography (TEE)‐based annulus sizing for transcatheter aortic valve replacement (TAVR). Methods Electronic search of PubMed, Biomed Central, Scopus, and Google Scholar databases was conducted until August 15, 2019. We included all types of studies comparing MDCT‐based annulus sizing with TEE‐based annulus sizing and assessing paravalvular regurgitation (PVR). Data were summarized using the Mantel‐Haenszel odds ratio (OR) with 95% confidence intervals (CI). Results A total of six studies were included. Pooled analysis of 431 participants in the MDCT group and 509 participants in the TEE group demonstrated that MDCT‐based annulus sizing is associated with a significantly lower incidence of more than moderate PVR as compared to 2DTEE‐based sizing (OR: 0.31, 95% CI: 0.18‐0.54, P < .0001; I 2 = 0%). There was no statistical difference in annulus rupture (OR: 0.57, 95% CI: 0.12‐2.66, P = .91; I 2 = 0%), procedural mortality (OR: 0.97, 95% CI: 0.19‐4.86, P = .97; I2 = 0%), and 30‐day mortality (OR: 0.63, 95% CI: 0.26‐1.50, P = .29; I 2 = 0%) with MDCT or 2DTEE‐based annulus sizing. Compared with 3DTEE, the incidence of PVR in the MDCT group was lower, but there was no statistical difference in 30‐day mortality. Conclusion Use of MDCT in comparison with 2DTEE is associated with significantly lower incidence of more than moderate PVR after TAVR. There seems to be no difference in annulus rupture and 30‐day mortality with either imaging modality.


| INTRODUC TI ON
Transcatheter aortic valve replacement (TAVR) is an effective therapeutic modality in managing patients with severe aortic stenosis. 1 Though a highly successful procedure, complications like paravalvular aortic regurgitation (PVR) can be seen in up to 38% of patients undergoing TAVR. 1,2 The occurrence of PVR consequently results in poor clinical outcomes and a significant increase in mortality. Tamburino et al 3 reported PVR to be an independent predictor of mortality between 30 days and 1 year, in a sample of 663 patients. The authors observed a fourfold increased risk of mortality in patients demonstrating more than moderate postprocedural PVR. 3 Incongruous sizing of the aortic annulus resulting in inappropriate valve selection is a major reason for postoperative PVR. The junctional nadirs of the aortic leaflets at the distal part of the left ventricular outflow tract form a virtual ring that is regarded as the aortic annulus during TAVR. 4 In the absence of a discrete anatomical structure, accurate assessment of the annulus via appropriate imaging is critical in preventing PVR. On the other hand, oversizing of the prosthetic valve can lead to significant complications like annulus rupture, coronary obstruction, and conduction disturbances. 5 Traditionally, two-dimensional (2D) transesophageal echocardiography (TEE) has been used for evaluating annulus size for TAVR. 6 However, it is increasingly recognized that 2DTEE may not accurately measure the oval three-dimensional (3D) annulus structure and considerable sizing variations may occur depending upon the axis of orientation. 4,7 The use of 3DTEE has been described to overcome the limitations of 2DTEE with significantly higher annulus diameters achieved with exclusive use of 3DTEE for valvular sizing. 8 Over the last decade, multidetector computed tomography (MDCT) has been increasingly used for annulus sizing before TAVR, as it provides a detailed understanding of the valvular anatomy with a superior spatial resolution. 9 Studies have demonstrated that annulus measurements with 2DTEE frequently result in valve undersizing as compared to MDCT-based measurements. 10 On the other hand, a recent meta-analysis by Rong et al 11 has shown that measurements by 3DTEE may be comparable to that of MDCT and may lead to reduced contrast exposure. While multiple studies have compared differences in annulus sizing with TEE and MDCT, 10,12,13 evidence on the effect of imaging modality on the postoperative outcomes has not been summarized to date. Therefore, the purpose of this systematic review and meta-analysis was to evaluate the difference in postoperative outcomes following MDCT and TEE-based annulus sizing for TAVR.

| ME THODS
The guidelines of the PRISMA statement (Preferred Reporting Items for Systematic Reviews and Meta-analyses) 14 and the Cochrane Handbook for Systematic Reviews of Intervention were followed during the conduct of this review. 15 The research question to be answered was the following: Does using MDCT-based annulus sizing in TAVR associate with a lower incidence of PVR and improved clinical outcomes as compared to TEE-based measurements?

| Search strategy
A computerized literature search of PubMed, Biomed Central, Scopus, and Google Scholar databases was carried out. The last literature search was conducted on August 15, 2019. Two independent reviewers performed the electronic search using the following keywords: "Multidetector Computed Tomography," "Computed Tomography," "MDCT," "Transesophageal Echocardiography," "Echocardiography," "TEE," "transcatheter aortic valve replacement," "transcatheter aortic valve implantation," "paravalvular regurgitation," "paravalvular leak," and "clinical outcomes." The search strategy and results of the PubMed search are presented in Table S1. We also performed a manual search of references of included studies and review articles on the subject for identification of any additional studies. After assessing the studies by their titles and abstracts, full texts of selected articles were retrieved. Both the reviewers assessed individual studies based on inclusion criteria. Disagreements, if any, were resolved by mutual agreement.

| Inclusion criteria and outcomes
Utilizing the PICOS (Population, Intervention, Comparison, Outcome, and Study design) outline, we included all types of studies conducted on patients undergoing TAVR (Population), comparing MDCT-based annulus sizing (Intervention) with TEE-based annulus sizing (Comparison) and assessing PVR and other clinical outcomes (Outcomes). At the protocol stage, we aimed to include studies comparing both 2DTEE and 3DTEE with MDCT for annulus valve sizing in TAVR patients. Studies comparing MDCT and TEE-based annulus measurements on the same group of patients were excluded. We also excluded single-arm studies, case reports, review articles, and non-English language studies.
Using an abstraction form, two reviewers retrieved data from selected studies. The following details were sourced: Authors, publication year, sample size, inclusion/exclusion criteria, baseline characteristics, MDCT and TEE protocol, PVR, and any other clinical outcomes. The primary outcome was the incidence of moderate-severe PVR. Secondary outcomes were the incidence of annulus rupture, procedural mortality, and 30-day mortality.

| Risk of bias assessment
Retrospective cohort studies were analyzed using the risk of bias assessment tool for nonrandomized studies (RoBANS). 16 Studies were rated as low risk, high risk, or unclear risk of bias for the following: selection of participants, confounding variables, intervention measurements, blinding of outcome assessment, incomplete outcome data, selective outcome reporting. Quality of randomized control trials (RCTs) was assessed using the "Cochrane Collaboration risk assessment tool". 17 Studies were rated as low risk, high risk, or unclear risk of bias for the following: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting, and other biases.

| Statistical analysis
Because of significant heterogeneity among studies, a random-effects model was used to calculate the pooled effect size. Categorical data were summarized using the Mantel-Haenszel odds ratio (OR) with 95% confidence intervals (CI). Heterogeneity was calculated using the I 2 statistic. I 2 values of 25%-50% represented low, values of 50%-75% represented medium, and more than 75% represented substantial heterogeneity. A sensitivity analysis was carried out to assess the influence of each study on the pooled effect size. The software "Review Manager" (RevMan, version 5.3; Nordic Cochrane Centre [Cochrane Collaboration], Copenhagen, Denmark; 2014) was used for the meta-analysis. Publication bias was not assessed using funnel plots as there were less than 10 studies in our analysis. 15

| RE SULTS
The study flowchart is presented in Figure 1. Four studies were excluded after full-text evaluation. 10,12,13,18 In all four studies, MDCT and TEE-based annulus measurements were compared in the same group of patients. A total of six studies met the inclusion criteria. [19][20][21][22][23][24] Five studies compared MDCT and 2DTEE for annulus sizing, [19][20][21][22][23] while one study compared MDCT with 3DTEE-based annulus sizing. 24 The characteristics of the included studies are presented in Table 1. All studies had obtained informed written consent from study participants and were approved by the institutional ethical committee.

F I G U R E 1 Systematic review and meta-analysis flow diagram
Echocardiogram-based and MDCT-based annulus sizing was done at different time intervals in all studies, and data were analyzed retrospectively, except for one trial. Casset et al 20 conducted a prospective randomized trial evaluating the addition of MDCT to TEE and TTE-based annulus measurements on postoperative outcomes. Measurements were recorded in the systolic phase for both groups in all studies. Except for two studies, 22,24 both MDCT and TEE-based measurements were available to the operator during the procedure. Valves implanted were exclusively Edward SAPIEN or SAPIEN XT in four studies, [20][21][22][23] Edward Sapien and CoreValve in one study 19 and Edward Sapien and Evolut R in another study. 24 The risk of bias assessment of included studies is presented in Table 2, and the baseline characteristics of the participants of all six studies are presented in Table 3.
Meta-analysis was carried out for five studies comparing outcomes following MDCT and 2DTEE-based annulus measurements. 19

| MDCT vs 3DTEE
In the retrospective study of Wystub et al, 24 MDCT was used for annulus sizing in 116 patients and 3DTEE was utilized in 111 patients.
There was no significant difference in the baseline characteristics of the two groups. Significantly larger valves were used in the MDCT group as compared to the 3DTEE group (

| D ISCUSS I ON
Of the two types of aortic valvular regurgitation, central regurgitation is usually seen in diseased native valves whereas PVR is a complication seen only after TAVR. 25 Since the native valve is still in situ when the prosthesis is placed over the biological tissue, an incomplete seal may remain, thereby resulting in PVR. 25 Despite a technological improvement in devices to provide an efficient seal between the aortic annulus and the implanted prosthesis, the incidence of PVR is as high as 23.8% post TAVR. 26 The PARTNER trial has demonstrated that even mild PVR is associated with an increased risk of late mortality. 27  Despite MDCT becoming the gold standard imaging for annulus sizing, the requirement of contrast media is a significant limitation especially in patients with severe renal impairment. 32 An estimated 7%-10.5% of TAVR patients have been found to have MDCT contrast-related kidney injury. 33 With around 70% of the TAVR population having preoperative renal disease, TEE may still be an alternative imaging modality for such patients. 34 It may also be useful in individuals with iodine allergy, centers with high patient load or due to economic constraints. 20 In the absence of dynamic information by MDCT, TEE also yields better temporal resolution that aids in tracing calcified nodules and identification of mobile components. 20 In the face of such differences, it is important to analyze the differences in clinical outcomes following MDCT and TEE-based annulus measurements for TAVR. However, to date, only one study has compared the incidence of complications following MDCT vs 3DTEE-based annulus sizing.
Wystub et al, 24 comparing two cohorts of TAVR patients treated at different time intervals, found a reduced incidence of PVR in the MDCT group. Similar to 2DTEE, underestimation of annulus size resulting in smaller prosthesis was described as the probable reason for the difference in PVR. 24 The results of our review are to be interpreted with the following  This is the first systematic review and meta-analysis evaluating outcomes after MDCT vs TEE-based annulus sizing for TAVR.
After the pooling of data of more than 800 patients, our results indicate that the use of MDCT against 2DTEE is associated with a F I G U R E 5 Forest plot of 30-d mortality F I G U R E 4 Forest plot of procedural mortality F I G U R E 3 Forest plot of annulus rupture F I G U R E 2 Forest plot of more than moderate PVR significantly reduced incidence of more than moderate PVR after TAVR. However, there seems to be no difference in annulus rupture, procedural, and 30-day mortality with either imaging modality.
Further studies are required to provide evidence on postoperative outcomes following MDCT or 3DTEE-based annulus sizing.

S U PP O RTI N G I N FO R M ATI O N
Additional supporting information may be found online in the Supporting Information section.