- Email: [email protected]
Effect of ascending aortic dimension on acute procedural success following self-expanding transcatheter aortic valve replacement
IJCA-25078; No of Pages 6 International Journal of Cardiology xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Effect of ascending aortic dimension on acute procedural success following self-expanding transcatheter aortic valve replacement A multicenter retrospective analysis Yoshio Maeno a, Sung-Han Yoon a, Yigal Abramowitz a, Yusuke Watanabe b, Hasan Jilaihawi c, Mao-Shin Lin d, Jason Chan e, Rahul Sharma a, Hideyuki Kawashima b, Sharjeel Israr a, Hiroyuki Kawamori a, Masaki Miyasaka a, Tanya Rami a, Yoshio Kazuno a, Geeteshwar Mangat a, Mohammad Kashif a, Tarun Chakravarty a, Hsien-Li Kao d, Michael Kang-yin Lee e, Mamoo Nakamura a, Ken Kozuma b, Wen Cheng a, Raj R. Makkar a,⁎ a
Heart Institute, Cedars-Sinai Medical Center, Los Angeles, United States Division of Cardiology, Department of Internal Medicine, Teikyo University Hospital, Tokyo, Japan Department of Medicine, Cardiothoracic Surgery, New York University Langone Medical Center, New York, United States d Division of Cardiology, Heart Center, National Taiwan University Hospital, Taipei, Taiwan e Division of Cardiology, Queen Elizabeth Hospital, Kowloon, Hong Kong b c
a r t i c l e
i n f o
Article history: Received 2 December 2016 Received in revised form 13 May 2017 Accepted 25 May 2017 Available online xxxx Keywords: Aortic dimensions Device success Oversizing Predictors TAVI TAVR
a b s t r a c t Aims: Self–expanding (SE) valves are characterized with long stent frame design and the radial force of the device exists both in the inflow and outflow level. Therefore, we hypothesized that device success of SE-valves may be influenced by ascending aortic dimensions (AAD). The aim of this study was to determine the influence of AAD on acute device success rates following SE transcatheter aortic valve replacement (TAVR). Methods & Results: In 4 centers in the United States and Asia, 214 consecutive patients underwent SE-TAVR. Outcomes were assessed in line with Valve Academic Research Consortium criteria. AAD was defined as the sum of the short and long axis aortic diameter divided by 2. Overall, device success rate was 85.0%. Multivariate analysis revealed that increased AAD (Odds ratio 1.27) and % oversizing (Odds ratio 0.88) were found to be independent predictors of unsuccessful device implantation. The c-statistic of the model for device success was area under the curve 0.79, sensitivity 81.3% and specificity 44.0%. Co-existence of several risk factors was associated with an exponential fall to 64.2% in device success rate. For a large AAD, however, optimally oversized SE-valves (threshold 16.2%) resulted with high device success rates compared to suboptimal oversizing (88.6% vs. 64.2%, p = 0.005). Conclusions: Larger AAD and smaller degrees of oversizing were confirmed to be the most relevant predictors of unsuccessful device implantation following SE-valve implantations. Optimal oversizing of great significance was noted, particularly that with a large AAD. © 2017 Elsevier B.V. All rights reserved.
1. Introduction Transcatheter aortic valve replacement (TAVR) is a well-established alternative to surgical aortic valve replacement for high-risk patients with severe aortic valve stenosis [1,2]. Previous prospective studies have shown that self-expanding (SE) TAVR results in high rate of paravalvular leak (PVL), high incidence of pacemaker, and low device success rates compared to balloon-expandable TAVR [3–5]. Previous studies have well established that the predictors of PVL after SE-TAVR
⁎ Corresponding author at: Advanced Health Sciences Pavilion, Cedars-Sinai Heart Institute, 127 S. San Vicente Blvd, Third Floor, Suite A3600, Los Angeles, CA 90048, United States. E-mail addresses: [email protected] (Y. Maeno), [email protected] (R.R. Makkar).
[6–8]. Device success is a well-defined outcome, but only a few studies have reported the frequency of device success for SE-devices in accordance with the Valve Academic Research Consortium (VARC)-2 definitions [5,9]. Unlike balloon-expandable valves, SE-valves have long stent frame designs with radial forces existing at both the inflow and outflow level. Therefore, it generates a hypothesis that SE-valve may be influenced by the ascending aortic diameter (AAD). The Society of Cardiovascular Computed Tomography (SCCT) guidelines recommend the assessment of AAD before SE-TAVR procedure [10]. Furthermore, for SE-valve, SCCT guidelines recommend proximal AAD should not exceed 40-43 mm by multidetector CT (MDCT) measurements [10,11]. However, there is scarce data available regarding the interaction between device success rates following SE-valve and AAD. The aim of this study was to determine the influence of AAD on acute device success rates following SE-valve implantations.
http://dx.doi.org/10.1016/j.ijcard.2017.05.120 0167-5273/© 2017 Elsevier B.V. All rights reserved.
Please cite this article as: Y. Maeno, et al., Effect of ascending aortic dimension on acute procedural success following self-expanding transcatheter aortic valve replacement, Int J Cardiol (2017), http://dx.doi.org/10.1016/j.ijcard.2017.05.120
2
Y. Maeno et al. / International Journal of Cardiology xxx (2017) xxx–xxx
2. Methods
3. Results
2.1. Study population and procedure A cohort of 214 consecutive patients with aortic stenosis were treated with SE-valves (Evolut R./CoreValve, Medtronic, Minneapolis, MN, USA) in 4 centers (Cedars-Sinai Medical Center, Los Angeles, United States; Teikyo University Hospital, Tokyo, Japan; National Taiwan University Hospital, Taipei, Taiwan; Queen Elizabeth Hospital, Kowloon, Hong Kong). Patients with previous bioprosthesis, and with non-contrast or poor MDCT imaging quality were excluded from the initial analysis of this cohort. The decision to proceed with TAVR was with the consensus of the individual in the dedicated heart team. TAVR endpoints, device success, and adverse events were considered according to VARC-2 definitions [9]. Device success was defined as following; Absence of procedural mortality; correct positioning of a single prosthetic heart valve into the proper anatomical location; and intended performance of the prosthetic heart valve [9]. Post-procedural PVL was assessed in line with VARC-2 criteria with periprocedural transesophageal echocardiography examinations reviewed retrospectively [9]. Clinical data, patient characteristics, echocardiographic data, and procedural variables were retrospectively recorded. All aortic root dimensions were retrospectively analyzed at core laboratory Cedars-Sinai Medical Center. 2.2. MDCT image acquisition and aortic annulus analysis MDCT images were performed using several different scanners, 64-slice or higher. Multiple CT scan protocols incorporating electrocardiography gating were used according to site-specific practices. For aortic annular and aortic-valvular complex dimensions, curved multiplanar reconstruction analyses were performed using 3-mensio valves software™ (version 8.2, Pie Medical Imaging, Maastricht, the Netherlands) [12]. We measured AAD by the following method: Average AAD = (short axis aortic diameter + long axis aortic diameter)/2. The AAD height was considered according to SCCT guidelines [10]. Therefore, the AAD height was defined as the cross-sectional region 40 mm superior to the annular plane [10] (Fig. 1). For reconstruction, mid-systolic data was used [13]. Cover index was determined as follows: calculated perimeter oversizing (%) = (prosthesis perimeter/annulus perimeter − 1) × 100. The nominal perimeter of a fully expanded valve is 72.2 mm for the 23-mm valve, 81.6 mm for the 26-mm valve, 91.1 mm for the 29-mm valve, and 97.3 mm for the 31-mm valve. A recently validated 850-Hounsfield unit threshold was used to detect areas of calcium in the region of interest [14]. Angulation of the aorta was calculated from a coronal projection at the level of the aortic annulus and was defined as the angle between the horizontal plane and the plane of the aortic annulus [15,16]. 2.3. Statistical analysis Continuous variables were tested for a normality of distribution using the ShapiroWilk test. These variables were then reported and analyzed accordingly. Mann-Whitney U tests were used in cases of abnormal distribution. Categorical variables were compared by chi square statistics or the Fisher exact test. Pearson bivariate analysis with a two-tailed test for significance was used for parametric variables. The variables associated with preprocedurally unsuccessful device implantation with a p-value b0.10 on univariate analysis were entered in a multivariate logistic regression model to determine the independent predictors of unsuccessful device implantation which was further evaluated using c-statistics of the receiver operator characteristic curve. Sensitivity, specificity were calculated using specific cutoffs using the Youden index generated from the receiver operator characteristic (ROC) curve based on the predictive probability for unsuccessful device implantation. All of the analyses were considered significant at a two-tailed p-value of b0.05. SPSS statistics software 22.0 (SPSS, Chicago, Illinois) was used to perform all statistical evaluation.
Baseline clinical and pre-procedural characteristics of the 214 patients are shown in Table 1. The mean age of the patients was 83.0 (77.0–88.0) years and the mean logistic EuroSCORE was 18.0 (10.1–31.2) %. Overall, device success occurred in 182 of 214 patients (85.0%). Prevalence of females was higher among the group with device success compared to the group with unsuccessful device implantation (56.6% vs. 25.0%, p = 0.001). Patients with history of peripheral artery disease trended lower in device unsuccessful implantation group (6.3% vs. 20.3%, p = 0.06). Other clinical baseline parameters were comparable between both groups. There were no differences in the severity of aortic stenosis and the presence of left ventricular outflow tract calcification between both groups (Table 1). For aortic root dimensions, patients with unsuccessful device implantation had large aortic annulus or large AAD (perimeter 83.4 (78.5–86.6) mm vs. 74.9 (69.8–77.8) mm; AAD 34.3 ± 3.5 mm vs. 31.6 ± 3.1 mm; p b 0.001 for both) (Table 1). However, annulus perimeter was not correlating with AAD (r = 0.39). Patients with unsuccessful device implantation also had higher angulation of the aorta (49.9 ± 7.7° vs. 46.4 ± 9.2°, p = 0.041) and much leaflet calcium volume [178.4 (76.5–657.5) mm3 vs. 121.4 (47.0–259.1) mm3, p = 0.019]. Details of the post-procedural outcomes are summarized in Table 1. Overall, Mortality at discharge was occurred for 1.4% (3 patients). The most common valve size used was 29 mm regardless of device success. Smaller degrees of oversizing were significantly associated with unsuccessful device implantation [10.4 (6.7–15.7) % vs. 17.0 (12.5– 21.0) %, p b 0.001]. Post-dilatation was performed more frequently in patients that had unsuccessful device implantation (59.4% vs. 26.9%, p b 0.001). In patients with unsuccessful device implantation, 56.3% (n = 18) of those patients had moderate to severe PVL, and 31.3% (n = 10) of those patients needed a 2nd valve. 8 patients with more than mean aortic valve gradients of 20 mm Hg and peak velocities of 3 m/s or with prosthesis-patient mismatch were observed.
3.1. Independent predictors of unsuccessful device implantation Independent predictors of unsuccessful device implantation included the degree of oversizing (odds ratio [OR] 0.88, 95% confidence interval [CI] 0.82 to 0.94, p b 0.001) and AAD (OR 1.27, 95% CI 1.12 to 1.45, p b 0.001) (Table 2). Predictive probabilities were generated using this logistic regression model and the c-statistics for the model for unsuccessful device implantation was area under the curve (AUC) 0.79 (95% CI 0.70 to 0.89, p b 0.001, sensitivity 81.3%, specificity 44.0%) (Supplementary Fig. 1). Using ROC curve analysis, an AAD of 32.1 mm was identified as the threshold for increased risk for unsuccessful device
Fig. 1. Ascending aortic dimensions According to the society of cardiovascular computed tomography guidelines, the AAD height was defined as the cross-sectional region 40 mm superior to the annular plane. We measured AAD by the following method: Average AAD = (short axis aortic diameter + long axis aortic diameter)/2. Abbreviation: AAD = ascending aortic dimensions.
Please cite this article as: Y. Maeno, et al., Effect of ascending aortic dimension on acute procedural success following self-expanding transcatheter aortic valve replacement, Int J Cardiol (2017), http://dx.doi.org/10.1016/j.ijcard.2017.05.120
Y. Maeno et al. / International Journal of Cardiology xxx (2017) xxx–xxx
3
Table 1 Baseline clinical and pre/post-procedural characteristics.
Baseline clinical and pre-procedural characteristics Age, yrs Female Body mass index, kg/m2 Hypertension Dyslipidemia Diabetes Chronic obstructive pulmonary disease Coronary artery disease Cerebrovascular disease Peripheral artery disease Previous pacemaker eGFR, ml/min Logistic EURO score, % AV area, cm2 Ejection fraction, % AV mean gradient, mm Hg CT aortic annulus dimensions Mean annulus diameter, mm Annulus ration (minor/major) Annulus perimeter, mm Aortic root angulation, 0 Leaflet calcium volume (HU-850), mm3 The presence of LVOT calcification % oversizing by perimeter, % Mean AAD, mm Post-procedural characteristics and outcome at discharge Valve type Evolute.R CoreValve Valve size, mm 23 26 29 31 Alternative approach Pre-dilatation Post-dilatation Need for a 2nd valve Aortic root injury Peri-procedural TEE None PVL Mild PVL Moderate or severe PVL AV area N1.2 cm2 (26–31 mm); AVA N0.9 cm2 (23 mm), % AV gradient b20 mm Hg or peak velocity b3 m/s, % Mortality Cerebrovascular accident/TIA Myocardial infarction Acute kidney injury stage 2 or 3 New permanent pacemaker implantationa Major vascular complication
Overall (n = 214)
Unsuccessful device implantation (n = 32)
Device success (n = 182)
p-Value
83.0 (77.0–88.0) 111 (51.9%) 24.7 (21.6–28.3) 182 (85.0%) 150 (70.1%) 62 (29.0%) 44 (20.6%) 91 (42.5%) 42 (19.6%) 39 (18.2%) 24 (11.2%) 45.6 (36.8–57.3) 18.0 (10.1–31.2) 0.68 (0.50–0.80) 62.0 (50.0–68.0) 44.0 (37.0–52.0)
81.0 (75.3–85.8) 8 (25.0%) 26.1 (21.7–29.2) 27 (84.4%) 25 (78.1%) 11 (34.4%) 6 (18.8%) 14 (43.8%) 6 (18.8%) 2 (6.3%) 2 (6.3%) 48.4 (36.4–57.8) 15.0 (9.5–29.2) 0.60 (0.49–0.70) 56.6 (41.3–68.0) 42.0 (37.0–47.8)
84.0 (78.0–88.0) 103 (56.6%) 24.6 (21.6–28.2) 155 (85.2%) 125 (68.7%) 51 (28.0%) 38 (20.9%) 77 (42.3%) 36 (19.8%) 37 (20.3%) 22 (12.1%) 45.1 (36.8–57.3) 18.0 (10.1–31.7) 0.70 (0.50–0.80) 63.0 (53.0–68.0) 44.0 (37.0–53.1)
0.90 0.001 0.50 0.54 0.28 0.47 0.78 0.88 0.89 0.06 0.27 0.33 0.34 0.13 0.11 0.36
24.0 ± 2.4 0.81 ± 0.06 75.7 (70.4–81.5) 46.9 ± 9.0 130.1 (51.4–270.1) 95 (44.4%) 16.2 (11.6–20.4) 32.0 ± 3.3
25.7 ± 2.7 0.82 ± 0.07 83.4 (78.5–86.6) 49.9 ± 7.7 178.4 (76.5–657.5) 17 (53.1%) 10.4 (6.7–15.7) 34.3 ± 3.5
23.7 ± 2.3 0.81 ± 0.06 74.9 (69.8–79.8) 46.4 ± 9.2 121.4 (47.0–259.1) 78 (42.9%) 17.0 (12.5–21.0) 31.6 ± 3.1
b0.001 0.24 b0.001 0.041 0.019 0.28 b0.001 b0.001
53 (24.8%) 161 (75.2%)
7 (21.9%) 25 (78.1%)
46 (25.3%) 136 (74.7%)
8 (3.7%) 66 (30.8%) 98 (45.8%) 40 (18.7%) 9 (5.2%) 53 (24.8%) 68 (31.8%) 10 (4.7%) 2 (0.9%)
0 5 (15.6%) 15 (46.9%) 12 (37.5%) 1 (3.1%) 9 (28.1%) 19 (59.4%) 10 (31.3%) 1 (3.1%)
8 (4.4%) 61 (33.5%) 83 (45.6%) 28 (15.4%) 8 (4.4%) 44 (24.2%) 49 (26.9%) 0 1 (0.5%)
0.60 0.63 b0.001 b0.001 0.28
140 (65.4%) 56 (26.2%) 18 (8.4%) 207 (96.7%) 211 (98.6%) 3 (1.4%) 2 (0.9%) 1 (0.5%) 3 (1.4%) 54 (28.4%) 15 (7.0%)
10 (31.3%) 4 (12.5%) 18 (56.3%) 25 (78.1%) 29 (90.6%) 3 (9.4%) 1 (3.1%) 0 0 8 (26.7%) 2 (6.3%)
130 (71.4%) 52 (28.6%) 0 182 (100%) 182 (100%) 0 1 (0.5%) 1 (0.5%) 3 (1.6%) 46 (28.7%) 13 (7.1%)
b0.001 0.056 b0.001 b0.001 0.030 0.030 0.28 0.85 0.61 0.82 0.61
0.68
0.001
Values are mean ± SD, median (interquartile range), or n (%). Abbreviations: AAD = ascending aorta dimensions; AV = aortic valve; LVOT = left ventricular outflow tract; PVL = paravalvular leak; TEE = transesophageal echocardiography; TIA = transit ischemic attack. a Patients with history of permanent pacemaker were excluded.
Table 2 Multivariate regression for unsuccessful device implantation following Self-expanding valve.
AAD % oversizing by perimeter Female Peripheral artery disease Aortic root angulation Mean annulus diameter Annulus perimeter Leaflet calcium volume Valve size
Univariate odds ratio (95% CI)
p-Value
Multivariate odds ratioa (95% CI)
p-Value
1.28 (1.13–1.44) 0.87 (0.82–0.93) 0.26 (0.11–0.60) 0.26 (0.06–1.14) 1.05 (1.001–1.09) 1.42 (1.20–1.69) 1.14 (1.08–1.21) 1.002 (1.001–1.003) 1.43 (1.14–1.79)
b0.001 b0.001 0.001 0.057 0.043 b0.001 b0.001 0.004 0.002
1.27 (1.12–1.45) 0.88 (0.82–0.94) Dropped Dropped Dropped Dropped Dropped Dropped Dropped
b0.001 b0.001
Abbreviations: AAD = ascending aortic dimensions; CI = confidence interval. a All parameters (p b 0.10) were entered in a multivariate logistic regression.
Please cite this article as: Y. Maeno, et al., Effect of ascending aortic dimension on acute procedural success following self-expanding transcatheter aortic valve replacement, Int J Cardiol (2017), http://dx.doi.org/10.1016/j.ijcard.2017.05.120
4
Y. Maeno et al. / International Journal of Cardiology xxx (2017) xxx–xxx
Fig. 2. Importance optimally oversizing for large AAD Optimally oversized SE-valves (N16.2%) had high device success rates for large AAD. Abbreviations: AAD = ascending aorta dimensions; SE = self-expanding.
implantation (AUC 0.72, sensitivity 75.0%, specificity 40.1%). Overall, 45.3% (97 patients) of the patients had an AAD N32.1 mm. For each valve size, sub-analysis for the thresholds of AAD using ROC curve analysis are as following; no patients with unsuccessful device implantation for 23 mm valve, 30.8 mm for 26 mm valve (AUC 0.75), 32.3 mm for 29 mm valve (AUC 0.67), and 34.8 mm for 31 mm valve (AUC 0.61). On the other hand, % oversizing below 16.2 was identified as suboptimal oversizing for unsuccessful device implantation (AUC 0.74, sensitivity 78.1%, specificity 45.1%). Overall 107 of 214 patients (50.0%) were below this threshold (under 16.2% oversizing). Combination of these two risk factors was associated with an exponential fall in device success rate, ranging from only 75.3% (large AAD alone) or 76.6% (small % oversizing alone) to 64.2% (both); this observation was robust for SE-devices. For patients with large AAD (N32.1 mm), optimal oversizing was associated with significantly higher device success rates than suboptimal oversizing (88.6% vs. 64.2%, p = 0.005) (Fig. 2). Of patients with optimally oversized SE-TAVR with large AAD, we experienced 5 cases with unsuccessful SE-device implantation. Two of them had need for a 2nd valve, and the remaining 3 patients had moderate PVL (2 patients) or death related to procedure (1 patient who had also moderate PVL). 4. Discussion Our findings can be summarized as follows: (i) large AAD and small degree of oversizing were confirmed to be the most relevant predictors of unsuccessful device implantation following SE-TAVR. (ii) a dramatic importance for optimal oversizing was noted, particularly with large AAD. Very few systematic data is available regarding acute device success after SE-TAVR. Device success is an important endpoint that has been associated with decreased mortality and TAVR-related complications in comparison to patients with unsuccessful device implantation [17]. Previous multicenter prospective the CHOICE trials reported device success rates following SE-TAVR as 77.5% [5]. On the other hand, a recently published the US CoreValve trial reported that acute devices success was 86.8% [16]. The present study (85.0%) was similar to that reported in the US CoreValve trial. The effect of increased AAD on acute device success following SE-TAVR has not previously been systematically examined. Nonetheless, a large AAD is a relative contraindication for CoreValve implantation according to the manufacturer's instructions. Moreover, Leipsic et al. suggested that AAD should be below 43 mm following SE-TAVR
[11]. Our findings demonstrated that lower degrees of oversizing and larger AAD were found to be independent predictors of acute procedural unsuccessful device implantation for SE- valve implantations. Lower degree of oversizing is a well-established independent predictor of PVL in prior studies and results in an incomplete attachment of the device to the aortic valvular complex wall [5,8,17]. This might be a cause of malaposition during valve deployment. On the basis of a hypothesis that the radial force of the outflow frame of SE-valves may be affected by AAD, our findings demonstrated that larger AAD was associated with decreased acute device success following SE-TAVR. Although the present study did not have sufficient power to show each threshold for different valve sizes due to limited number of this adverse event, the AAD threshold for unsuccessful each-sized device implantation increased linearly. Our findings imply that each labeled valve size has an optimal AAD. Therefore, future studies will be needed to validate this hypothesis and determine optimal AAD. There are several possible explanations for our findings. It is of extreme importance that the valve frame of SE-prosthesis, both at the outflow and inflow level needs to be closely attached to the wall of the aortic-valvular complex in order to acquire a stable device position. When the outflow stent frame is weakly attached to the aortic wall, valve stability is solely dependent on the inflow level. Furthermore, malaposition is affected by cardiac rhythm during valve deployment, particularly when the outflow frame is not closely attached to the aortic wall and especially in the presence of a lower degree of oversizing (Supplementary Fig. 2 shows a representative case). As a result, the need for a second valve, PVL, and/or valve dysfunction due to inadequate expansion may occur. The CHOICE trial showed that the use of SE-valves resulted in lower device success rates than the use of balloon-expandable (BE) valves [5]. In the setting of large AAD, the reduced anchoring due to larger aortic dimensions of the long stented SE-valve may contribute to increased risk of unsuccessful device implantation. In addition, AAD is less relevant for the shorter stent frame of balloon-expandable valves. Our findings also revealed that the combination of large AAD and lower degree of oversizing was associated with an exponential fall in device success rate. This latter indicates that the close attachment between valve frame and aortic valvular complex, particularly in both annulus- and AAD level, may be essential to increase anchoring. Tzamtzis et al. demonstrated that radial force of the SE-valve frame proportionately drops with the increasing diameter of the aorto-valvular complex [18]. Our findings are consistent with this study. In the present study, overall 6 of 10 patients who needed for a 2nd valve implantation had a large AAD and sub-optimally oversized SE-valve. On the other hand, overall
Please cite this article as: Y. Maeno, et al., Effect of ascending aortic dimension on acute procedural success following self-expanding transcatheter aortic valve replacement, Int J Cardiol (2017), http://dx.doi.org/10.1016/j.ijcard.2017.05.120
Y. Maeno et al. / International Journal of Cardiology xxx (2017) xxx–xxx
12 of 18 patients with moderate to severe PVL had a large AAD and suboptimally oversized SE-valve. Therefore, in patients with a large AAD, the optimal device sizing is essential to reduce PVL and avoid a second valve implantation. In a logistic regression analysis, the model demonstrates an important interaction of risk factors for device success, which may help to identify patients at risk for unsuccessful device implantation due to non-modifiable factors such as large AAD. It also emphasizes the importance of the modifiable factor of valve selection, which is oversizing with the dramatic difference of device success rates between suboptimal- and optimal oversizing. Optimal oversizing was especially important in the presence of a large AAD to increase acute device success rates following SE-valve implantations. Moreover, it becomes generally acceptable that SE-valves need larger oversizing compared to BE-valves in order to prevent PVL or unsuccessful device implantations [5,8]. Our findings support the results of these studies. In the present study, for patients with a large AAD, we experienced 5 cases of unsuccessful SE-valve implantation, despite an optimal oversizing. This may be attributable to other factors such as depth of valve implantation, aortic root angulation, and calcification of the aortic-valvular complex – all which have been identified as predictors of PVL in prior studies [7,14,15,19]. Previous studies have demonstrated that higher grades of leaflet calcification predicted increased PVL following SE-valve implantations [7,20]. On the other hand, lower degree of leaflet calcification has been associated with valve embolization and/ or need for a 2nd valve following SE-TAVR [21]. These contradictive effect of calcification grades on device success may explain why increased leaflet calcification, was not found to be an independent predictor for unsuccessful device implantation in the present study. Nonetheless, we did show that patients with unsuccessful device implantation had higher calcium volume in a univariable analysis mainly due to the more dominant effect of increased calcification on higher PVL grades. Finally, in the setting of a large AAD, optimally oversized SE-valve is an important factor contributing to a high rate of device success. In addition, in order to achieve good clinical outcomes, our findings suggest that AAD should be assessed pre-TAVR along with aortic annulus measurements. In clinical practice, for patients with large AAD, upsizing of SE valve should be considered, especially in cases of borderline sizing.
4.1. Study limitations Several limitations of the present study should be addressed. The study represents a retrospective, multi-center. The findings are subject to selection bias and confounders. The relatively low number of events (32 patients with unsuccessful device implantation) mandated a method of multivariable analysis that may exclude relevant variables with a weaker predictive value; certain previously identified predictor of PVL or prosthesis patient mismatch after SE-TAVR [6,19], such as depth of valve implantation, is not available in this dataset. However, given the present study based on pre-procedural assessment, we believe that our findings provide the importance of AAD assessment for successful SE-TAVR. In addition, they once more take a look at the interaction between AAD and SE-valve to achieve device success.
5. Conclusions A large AAD and smaller degrees of oversizing were confirmed to be relevant predictors for unsuccessful device implantation following SE-TAVR. Although a large AAD is an unmodifiable risk factor, optimal oversizing (N 16.2) appears to be important in increasing acute device success rates for these patients. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.ijcard.2017.05.120.
5
Conflict of interest Dr. Watanabe and Dr. Sharma is a proctor for Edwards Lifesciences. Dr. Jilaihawi is a consultant for Edwards Lifesciences, St. Jude Medical, and Venus MedTech. Dr. Makkar has received grant support from Edwards Lifesciences and St. Jude Medical; and is a consultant for Abbott Vascular, Cordis, and Medtronic; and holds equity in Entourage Medical. All other authors declare having no potential conflict of interests. Acknowledgement of grant support Dr. Makkar has received grant support from Edwards Lifesciences (Irvine, California, USA) and St. Jude Medical (St. Paul, Minnesota, USA). References [1] D.H. Adams, J.J. Popma, M.J. Reardon, S.J. Yakubov, J.S. Coselli, G.M. Deeb, et al., Transcatheter aortic-valve replacement with a self-expanding prosthesis, N. Engl. J. Med. 370 (19) (May 8 2014) 1790–1798. [2] J.J. Popma, D.H. Adams, M.J. Reardon, S.J. Yakubov, N.S. Kleiman, D. Heimansohn, et al., Transcatheter aortic valve replacement using a self-expanding bioprosthesis in patients with severe aortic stenosis at extreme risk for surgery, J. Am. Coll. Cardiol. 63 (19) (May 20 2014) 1972–1981. [3] N.E. Moat, P. Ludman, M.A. de Belder, B. Bridgewater, A.D. Cunningham, C.P. Young, et al., Long-term outcomes after transcatheter aortic valve implantation in high-risk patients with severe aortic stenosis: the U.K. TAVI (United Kingdom Transcatheter Aortic Valve Implantation) Registry, J. Am. Coll. Cardiol. 58 (20) (Nov 8 2011) 2130–2138. [4] M. Gilard, H. Eltchaninoff, B. Iung, P. Donzeau-Gouge, K. Chevreul, J. Fajadet, et al., Registry of transcatheter aortic-valve implantation in high-risk patients, N. Engl. J. Med. 366 (18) (May 3 2012) 1705–1715. [5] M. Abdel-Wahab, J. Mehilli, C. Frerker, F.-J. Neumann, T. Kurz, R. Tölg, et al., Comparison of balloon-expandable vs self-expandable valves in patients undergoing transcatheter aortic valve replacement: the CHOICE randomized clinical trial, JAMA 311 (15) (Apr 16 2014) 1503–1514. [6] M.A. Sherif, M. Abdel-Wahab, B. Stöcker, V. Geist, D. Richardt, R. Tölg, et al., Anatomic and procedural predictors of paravalvular aortic regurgitation after implantation of the Medtronic CoreValve bioprosthesis, J. Am. Coll. Cardiol. 56 (20) (Nov 9 2010) 1623–1629. [7] M. Seiffert, B. Fujita, M. Avanesov, C. Lunau, G. Schön, L. Conradi, et al., Device landing zone calcification and its impact on residual regurgitation after transcatheter aortic valve implantation with different devices, Eur. Heart J. Cardiovasc. Imaging 17 (5) (May 2016) 576–584. [8] D. Dvir, J.G. Webb, N. Piazza, P. Blanke, M. Barbanti, S. Bleiziffer, et al., Multicenter evaluation of transcatheter aortic valve replacement using either SAPIEN XT or CoreValve: degree of device oversizing by computed-tomography and clinical outcomes, Catheter. Cardiovasc. Interv. 86 (3) (Sep 2015) 508–515. [9] A.P. Kappetein, S.J. Head, P. Généreux, N. Piazza, N.M. van Mieghem, E.H. Blackstone, et al., Updated standardized endpoint definitions for transcatheter aortic valve implantation: the Valve Academic Research Consortium-2 consensus document, J. Am. Coll. Cardiol. 60 (15) (Oct 9 2012) 1438–1454. [10] S. Achenbach, V. Delgado, J. Hausleiter, P. Schoenhagen, J.K. Min, J.A. Leipsic, SCCT expert consensus document on computed tomography imaging before transcatheter aortic valve implantation (TAVI)/transcatheter aortic valve replacement (TAVR), J. Cardiovasc. Comput. Tomogr. 6 (6) (Dec 2012) 366–380. [11] J. Leipsic, R. Gurvitch, T.M. Labounty, J.K. Min, D. Wood, M. Johnson, et al., Multidetector computed tomography in transcatheter aortic valve implantation, JACC Cardiovasc. Imaging 4 (4) (Apr 2011) 416–429. [12] Y. Watanabe, M.-C. Morice, E. Bouvier, T. Leong, K. Hayashida, T. Lefèvre, et al., Automated 3-dimensional aortic annular assessment by multidetector computed tomography in transcatheter aortic valve implantation, JACC Cardiovasc. Interv. 6 (9) (Sep 2013) 955–964. [13] H. Jilaihawi, M. Kashif, G. Fontana, A. Furugen, T. Shiota, G. Friede, et al., Cross-sectional computed tomographic assessment improves accuracy of aortic annular sizing for transcatheter aortic valve replacement and reduces the incidence of paravalvular aortic regurgitation, J. Am. Coll. Cardiol. 59 (14) (Apr 3 2012) 1275–1286. [14] H. Jilaihawi, R.R. Makkar, M. Kashif, K. Okuyama, T. Chakravarty, T. Shiota, et al., A revised methodology for aortic-valvar complex calcium quantification for transcatheter aortic valve implantation, Eur. Heart J. Cardiovasc. Imaging 15 (12) (Dec 2014) 1324–1332. [15] Y. Abramowitz, Y. Maeno, T. Chakravarty, Y. Kazuno, N. Takahashi, H. Kawamori, et al., Aortic angulation attenuates procedural success following self-expandable but not balloon-expandable TAVR, JACC Cardiovasc. Imaging 9 (8) (Aug 2016) 964–972. [16] J.J. Popma, M.J. Reardon, S.J. Yakubov, J.B. Hermiller, J.K. Harrison, T.G. Gleason, et al., Safety and efficacy of self-expanding TAVR in patients with aortoventricular angulation, JACC Cardiovasc. Imaging 9 (8) (Aug 2016) 973–981. [17] M. Abdel-Wahab, T. Comberg, H.J. Büttner, M. El-Mawardy, K. Chatani, M. Gick, et al., Aortic regurgitation after transcatheter aortic valve implantation with balloon- and self-expandable prostheses: a pooled analysis from a 2-center experience, JACC Cardiovasc. Interv. 7 (3) (Mar 2014) 284–292.
Please cite this article as: Y. Maeno, et al., Effect of ascending aortic dimension on acute procedural success following self-expanding transcatheter aortic valve replacement, Int J Cardiol (2017), http://dx.doi.org/10.1016/j.ijcard.2017.05.120
6
Y. Maeno et al. / International Journal of Cardiology xxx (2017) xxx–xxx
[18] S. Tzamtzis, J. Viquerat, J. Yap, M.J. Mullen, G. Burriesci, Numerical analysis of the radial force produced by the Medtronic-CoreValve and Edwards-SAPIEN after transcatheter aortic valve implantation (TAVI), Med. Eng. Phys. 35 (1) (Jan 2013) 125–130. [19] H. Jilaihawi, D. Chin, T. Spyt, M. Jeilan, M. Vasa-Nicotera, J. Bence, et al., Prosthesispatient mismatch after transcatheter aortic valve implantation with the Medtronic-Corevalve bioprosthesis, Eur. Heart J. 31 (7) (Apr 2010) 857–864. [20] G. Tarantini, V. Gasparetto, M. Napodano, C. Fraccaro, G. Gerosa, G. Isabella, Valvular leak after transcatheter aortic valve implantation: a clinician updateon
epidemiology, pathophysiology and clinical implications, Am. J. Cardiovasc. Dis. 1 (3) (2011) 312–320. [21] N.M. Van Mieghem, C.J. Schultz, R.M.A. van der Boon, R.-J. Nuis, A. Tzikas, M.L. Geleijnse, et al., Incidence, timing, and predictors of valve dislodgment during TAVI with the Medtronic Corevalve system, Catheter. Cardiovasc. Interv. 79 (2012) 726–732.
Please cite this article as: Y. Maeno, et al., Effect of ascending aortic dimension on acute procedural success following self-expanding transcatheter aortic valve replacement, Int J Cardiol (2017), http://dx.doi.org/10.1016/j.ijcard.2017.05.120
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Effect of ascending aortic dimension on acute procedural success following self-expanding transcatheter aortic valve replacement A multicenter retrospective analysis Yoshio Maeno a, Sung-Han Yoon a, Yigal Abramowitz a, Yusuke Watanabe b, Hasan Jilaihawi c, Mao-Shin Lin d, Jason Chan e, Rahul Sharma a, Hideyuki Kawashima b, Sharjeel Israr a, Hiroyuki Kawamori a, Masaki Miyasaka a, Tanya Rami a, Yoshio Kazuno a, Geeteshwar Mangat a, Mohammad Kashif a, Tarun Chakravarty a, Hsien-Li Kao d, Michael Kang-yin Lee e, Mamoo Nakamura a, Ken Kozuma b, Wen Cheng a, Raj R. Makkar a,⁎ a
Heart Institute, Cedars-Sinai Medical Center, Los Angeles, United States Division of Cardiology, Department of Internal Medicine, Teikyo University Hospital, Tokyo, Japan Department of Medicine, Cardiothoracic Surgery, New York University Langone Medical Center, New York, United States d Division of Cardiology, Heart Center, National Taiwan University Hospital, Taipei, Taiwan e Division of Cardiology, Queen Elizabeth Hospital, Kowloon, Hong Kong b c
a r t i c l e
i n f o
Article history: Received 2 December 2016 Received in revised form 13 May 2017 Accepted 25 May 2017 Available online xxxx Keywords: Aortic dimensions Device success Oversizing Predictors TAVI TAVR
a b s t r a c t Aims: Self–expanding (SE) valves are characterized with long stent frame design and the radial force of the device exists both in the inflow and outflow level. Therefore, we hypothesized that device success of SE-valves may be influenced by ascending aortic dimensions (AAD). The aim of this study was to determine the influence of AAD on acute device success rates following SE transcatheter aortic valve replacement (TAVR). Methods & Results: In 4 centers in the United States and Asia, 214 consecutive patients underwent SE-TAVR. Outcomes were assessed in line with Valve Academic Research Consortium criteria. AAD was defined as the sum of the short and long axis aortic diameter divided by 2. Overall, device success rate was 85.0%. Multivariate analysis revealed that increased AAD (Odds ratio 1.27) and % oversizing (Odds ratio 0.88) were found to be independent predictors of unsuccessful device implantation. The c-statistic of the model for device success was area under the curve 0.79, sensitivity 81.3% and specificity 44.0%. Co-existence of several risk factors was associated with an exponential fall to 64.2% in device success rate. For a large AAD, however, optimally oversized SE-valves (threshold 16.2%) resulted with high device success rates compared to suboptimal oversizing (88.6% vs. 64.2%, p = 0.005). Conclusions: Larger AAD and smaller degrees of oversizing were confirmed to be the most relevant predictors of unsuccessful device implantation following SE-valve implantations. Optimal oversizing of great significance was noted, particularly that with a large AAD. © 2017 Elsevier B.V. All rights reserved.
1. Introduction Transcatheter aortic valve replacement (TAVR) is a well-established alternative to surgical aortic valve replacement for high-risk patients with severe aortic valve stenosis [1,2]. Previous prospective studies have shown that self-expanding (SE) TAVR results in high rate of paravalvular leak (PVL), high incidence of pacemaker, and low device success rates compared to balloon-expandable TAVR [3–5]. Previous studies have well established that the predictors of PVL after SE-TAVR
⁎ Corresponding author at: Advanced Health Sciences Pavilion, Cedars-Sinai Heart Institute, 127 S. San Vicente Blvd, Third Floor, Suite A3600, Los Angeles, CA 90048, United States. E-mail addresses: [email protected] (Y. Maeno), [email protected] (R.R. Makkar).
[6–8]. Device success is a well-defined outcome, but only a few studies have reported the frequency of device success for SE-devices in accordance with the Valve Academic Research Consortium (VARC)-2 definitions [5,9]. Unlike balloon-expandable valves, SE-valves have long stent frame designs with radial forces existing at both the inflow and outflow level. Therefore, it generates a hypothesis that SE-valve may be influenced by the ascending aortic diameter (AAD). The Society of Cardiovascular Computed Tomography (SCCT) guidelines recommend the assessment of AAD before SE-TAVR procedure [10]. Furthermore, for SE-valve, SCCT guidelines recommend proximal AAD should not exceed 40-43 mm by multidetector CT (MDCT) measurements [10,11]. However, there is scarce data available regarding the interaction between device success rates following SE-valve and AAD. The aim of this study was to determine the influence of AAD on acute device success rates following SE-valve implantations.
http://dx.doi.org/10.1016/j.ijcard.2017.05.120 0167-5273/© 2017 Elsevier B.V. All rights reserved.
Please cite this article as: Y. Maeno, et al., Effect of ascending aortic dimension on acute procedural success following self-expanding transcatheter aortic valve replacement, Int J Cardiol (2017), http://dx.doi.org/10.1016/j.ijcard.2017.05.120
2
Y. Maeno et al. / International Journal of Cardiology xxx (2017) xxx–xxx
2. Methods
3. Results
2.1. Study population and procedure A cohort of 214 consecutive patients with aortic stenosis were treated with SE-valves (Evolut R./CoreValve, Medtronic, Minneapolis, MN, USA) in 4 centers (Cedars-Sinai Medical Center, Los Angeles, United States; Teikyo University Hospital, Tokyo, Japan; National Taiwan University Hospital, Taipei, Taiwan; Queen Elizabeth Hospital, Kowloon, Hong Kong). Patients with previous bioprosthesis, and with non-contrast or poor MDCT imaging quality were excluded from the initial analysis of this cohort. The decision to proceed with TAVR was with the consensus of the individual in the dedicated heart team. TAVR endpoints, device success, and adverse events were considered according to VARC-2 definitions [9]. Device success was defined as following; Absence of procedural mortality; correct positioning of a single prosthetic heart valve into the proper anatomical location; and intended performance of the prosthetic heart valve [9]. Post-procedural PVL was assessed in line with VARC-2 criteria with periprocedural transesophageal echocardiography examinations reviewed retrospectively [9]. Clinical data, patient characteristics, echocardiographic data, and procedural variables were retrospectively recorded. All aortic root dimensions were retrospectively analyzed at core laboratory Cedars-Sinai Medical Center. 2.2. MDCT image acquisition and aortic annulus analysis MDCT images were performed using several different scanners, 64-slice or higher. Multiple CT scan protocols incorporating electrocardiography gating were used according to site-specific practices. For aortic annular and aortic-valvular complex dimensions, curved multiplanar reconstruction analyses were performed using 3-mensio valves software™ (version 8.2, Pie Medical Imaging, Maastricht, the Netherlands) [12]. We measured AAD by the following method: Average AAD = (short axis aortic diameter + long axis aortic diameter)/2. The AAD height was considered according to SCCT guidelines [10]. Therefore, the AAD height was defined as the cross-sectional region 40 mm superior to the annular plane [10] (Fig. 1). For reconstruction, mid-systolic data was used [13]. Cover index was determined as follows: calculated perimeter oversizing (%) = (prosthesis perimeter/annulus perimeter − 1) × 100. The nominal perimeter of a fully expanded valve is 72.2 mm for the 23-mm valve, 81.6 mm for the 26-mm valve, 91.1 mm for the 29-mm valve, and 97.3 mm for the 31-mm valve. A recently validated 850-Hounsfield unit threshold was used to detect areas of calcium in the region of interest [14]. Angulation of the aorta was calculated from a coronal projection at the level of the aortic annulus and was defined as the angle between the horizontal plane and the plane of the aortic annulus [15,16]. 2.3. Statistical analysis Continuous variables were tested for a normality of distribution using the ShapiroWilk test. These variables were then reported and analyzed accordingly. Mann-Whitney U tests were used in cases of abnormal distribution. Categorical variables were compared by chi square statistics or the Fisher exact test. Pearson bivariate analysis with a two-tailed test for significance was used for parametric variables. The variables associated with preprocedurally unsuccessful device implantation with a p-value b0.10 on univariate analysis were entered in a multivariate logistic regression model to determine the independent predictors of unsuccessful device implantation which was further evaluated using c-statistics of the receiver operator characteristic curve. Sensitivity, specificity were calculated using specific cutoffs using the Youden index generated from the receiver operator characteristic (ROC) curve based on the predictive probability for unsuccessful device implantation. All of the analyses were considered significant at a two-tailed p-value of b0.05. SPSS statistics software 22.0 (SPSS, Chicago, Illinois) was used to perform all statistical evaluation.
Baseline clinical and pre-procedural characteristics of the 214 patients are shown in Table 1. The mean age of the patients was 83.0 (77.0–88.0) years and the mean logistic EuroSCORE was 18.0 (10.1–31.2) %. Overall, device success occurred in 182 of 214 patients (85.0%). Prevalence of females was higher among the group with device success compared to the group with unsuccessful device implantation (56.6% vs. 25.0%, p = 0.001). Patients with history of peripheral artery disease trended lower in device unsuccessful implantation group (6.3% vs. 20.3%, p = 0.06). Other clinical baseline parameters were comparable between both groups. There were no differences in the severity of aortic stenosis and the presence of left ventricular outflow tract calcification between both groups (Table 1). For aortic root dimensions, patients with unsuccessful device implantation had large aortic annulus or large AAD (perimeter 83.4 (78.5–86.6) mm vs. 74.9 (69.8–77.8) mm; AAD 34.3 ± 3.5 mm vs. 31.6 ± 3.1 mm; p b 0.001 for both) (Table 1). However, annulus perimeter was not correlating with AAD (r = 0.39). Patients with unsuccessful device implantation also had higher angulation of the aorta (49.9 ± 7.7° vs. 46.4 ± 9.2°, p = 0.041) and much leaflet calcium volume [178.4 (76.5–657.5) mm3 vs. 121.4 (47.0–259.1) mm3, p = 0.019]. Details of the post-procedural outcomes are summarized in Table 1. Overall, Mortality at discharge was occurred for 1.4% (3 patients). The most common valve size used was 29 mm regardless of device success. Smaller degrees of oversizing were significantly associated with unsuccessful device implantation [10.4 (6.7–15.7) % vs. 17.0 (12.5– 21.0) %, p b 0.001]. Post-dilatation was performed more frequently in patients that had unsuccessful device implantation (59.4% vs. 26.9%, p b 0.001). In patients with unsuccessful device implantation, 56.3% (n = 18) of those patients had moderate to severe PVL, and 31.3% (n = 10) of those patients needed a 2nd valve. 8 patients with more than mean aortic valve gradients of 20 mm Hg and peak velocities of 3 m/s or with prosthesis-patient mismatch were observed.
3.1. Independent predictors of unsuccessful device implantation Independent predictors of unsuccessful device implantation included the degree of oversizing (odds ratio [OR] 0.88, 95% confidence interval [CI] 0.82 to 0.94, p b 0.001) and AAD (OR 1.27, 95% CI 1.12 to 1.45, p b 0.001) (Table 2). Predictive probabilities were generated using this logistic regression model and the c-statistics for the model for unsuccessful device implantation was area under the curve (AUC) 0.79 (95% CI 0.70 to 0.89, p b 0.001, sensitivity 81.3%, specificity 44.0%) (Supplementary Fig. 1). Using ROC curve analysis, an AAD of 32.1 mm was identified as the threshold for increased risk for unsuccessful device
Fig. 1. Ascending aortic dimensions According to the society of cardiovascular computed tomography guidelines, the AAD height was defined as the cross-sectional region 40 mm superior to the annular plane. We measured AAD by the following method: Average AAD = (short axis aortic diameter + long axis aortic diameter)/2. Abbreviation: AAD = ascending aortic dimensions.
Please cite this article as: Y. Maeno, et al., Effect of ascending aortic dimension on acute procedural success following self-expanding transcatheter aortic valve replacement, Int J Cardiol (2017), http://dx.doi.org/10.1016/j.ijcard.2017.05.120
Y. Maeno et al. / International Journal of Cardiology xxx (2017) xxx–xxx
3
Table 1 Baseline clinical and pre/post-procedural characteristics.
Baseline clinical and pre-procedural characteristics Age, yrs Female Body mass index, kg/m2 Hypertension Dyslipidemia Diabetes Chronic obstructive pulmonary disease Coronary artery disease Cerebrovascular disease Peripheral artery disease Previous pacemaker eGFR, ml/min Logistic EURO score, % AV area, cm2 Ejection fraction, % AV mean gradient, mm Hg CT aortic annulus dimensions Mean annulus diameter, mm Annulus ration (minor/major) Annulus perimeter, mm Aortic root angulation, 0 Leaflet calcium volume (HU-850), mm3 The presence of LVOT calcification % oversizing by perimeter, % Mean AAD, mm Post-procedural characteristics and outcome at discharge Valve type Evolute.R CoreValve Valve size, mm 23 26 29 31 Alternative approach Pre-dilatation Post-dilatation Need for a 2nd valve Aortic root injury Peri-procedural TEE None PVL Mild PVL Moderate or severe PVL AV area N1.2 cm2 (26–31 mm); AVA N0.9 cm2 (23 mm), % AV gradient b20 mm Hg or peak velocity b3 m/s, % Mortality Cerebrovascular accident/TIA Myocardial infarction Acute kidney injury stage 2 or 3 New permanent pacemaker implantationa Major vascular complication
Overall (n = 214)
Unsuccessful device implantation (n = 32)
Device success (n = 182)
p-Value
83.0 (77.0–88.0) 111 (51.9%) 24.7 (21.6–28.3) 182 (85.0%) 150 (70.1%) 62 (29.0%) 44 (20.6%) 91 (42.5%) 42 (19.6%) 39 (18.2%) 24 (11.2%) 45.6 (36.8–57.3) 18.0 (10.1–31.2) 0.68 (0.50–0.80) 62.0 (50.0–68.0) 44.0 (37.0–52.0)
81.0 (75.3–85.8) 8 (25.0%) 26.1 (21.7–29.2) 27 (84.4%) 25 (78.1%) 11 (34.4%) 6 (18.8%) 14 (43.8%) 6 (18.8%) 2 (6.3%) 2 (6.3%) 48.4 (36.4–57.8) 15.0 (9.5–29.2) 0.60 (0.49–0.70) 56.6 (41.3–68.0) 42.0 (37.0–47.8)
84.0 (78.0–88.0) 103 (56.6%) 24.6 (21.6–28.2) 155 (85.2%) 125 (68.7%) 51 (28.0%) 38 (20.9%) 77 (42.3%) 36 (19.8%) 37 (20.3%) 22 (12.1%) 45.1 (36.8–57.3) 18.0 (10.1–31.7) 0.70 (0.50–0.80) 63.0 (53.0–68.0) 44.0 (37.0–53.1)
0.90 0.001 0.50 0.54 0.28 0.47 0.78 0.88 0.89 0.06 0.27 0.33 0.34 0.13 0.11 0.36
24.0 ± 2.4 0.81 ± 0.06 75.7 (70.4–81.5) 46.9 ± 9.0 130.1 (51.4–270.1) 95 (44.4%) 16.2 (11.6–20.4) 32.0 ± 3.3
25.7 ± 2.7 0.82 ± 0.07 83.4 (78.5–86.6) 49.9 ± 7.7 178.4 (76.5–657.5) 17 (53.1%) 10.4 (6.7–15.7) 34.3 ± 3.5
23.7 ± 2.3 0.81 ± 0.06 74.9 (69.8–79.8) 46.4 ± 9.2 121.4 (47.0–259.1) 78 (42.9%) 17.0 (12.5–21.0) 31.6 ± 3.1
b0.001 0.24 b0.001 0.041 0.019 0.28 b0.001 b0.001
53 (24.8%) 161 (75.2%)
7 (21.9%) 25 (78.1%)
46 (25.3%) 136 (74.7%)
8 (3.7%) 66 (30.8%) 98 (45.8%) 40 (18.7%) 9 (5.2%) 53 (24.8%) 68 (31.8%) 10 (4.7%) 2 (0.9%)
0 5 (15.6%) 15 (46.9%) 12 (37.5%) 1 (3.1%) 9 (28.1%) 19 (59.4%) 10 (31.3%) 1 (3.1%)
8 (4.4%) 61 (33.5%) 83 (45.6%) 28 (15.4%) 8 (4.4%) 44 (24.2%) 49 (26.9%) 0 1 (0.5%)
0.60 0.63 b0.001 b0.001 0.28
140 (65.4%) 56 (26.2%) 18 (8.4%) 207 (96.7%) 211 (98.6%) 3 (1.4%) 2 (0.9%) 1 (0.5%) 3 (1.4%) 54 (28.4%) 15 (7.0%)
10 (31.3%) 4 (12.5%) 18 (56.3%) 25 (78.1%) 29 (90.6%) 3 (9.4%) 1 (3.1%) 0 0 8 (26.7%) 2 (6.3%)
130 (71.4%) 52 (28.6%) 0 182 (100%) 182 (100%) 0 1 (0.5%) 1 (0.5%) 3 (1.6%) 46 (28.7%) 13 (7.1%)
b0.001 0.056 b0.001 b0.001 0.030 0.030 0.28 0.85 0.61 0.82 0.61
0.68
0.001
Values are mean ± SD, median (interquartile range), or n (%). Abbreviations: AAD = ascending aorta dimensions; AV = aortic valve; LVOT = left ventricular outflow tract; PVL = paravalvular leak; TEE = transesophageal echocardiography; TIA = transit ischemic attack. a Patients with history of permanent pacemaker were excluded.
Table 2 Multivariate regression for unsuccessful device implantation following Self-expanding valve.
AAD % oversizing by perimeter Female Peripheral artery disease Aortic root angulation Mean annulus diameter Annulus perimeter Leaflet calcium volume Valve size
Univariate odds ratio (95% CI)
p-Value
Multivariate odds ratioa (95% CI)
p-Value
1.28 (1.13–1.44) 0.87 (0.82–0.93) 0.26 (0.11–0.60) 0.26 (0.06–1.14) 1.05 (1.001–1.09) 1.42 (1.20–1.69) 1.14 (1.08–1.21) 1.002 (1.001–1.003) 1.43 (1.14–1.79)
b0.001 b0.001 0.001 0.057 0.043 b0.001 b0.001 0.004 0.002
1.27 (1.12–1.45) 0.88 (0.82–0.94) Dropped Dropped Dropped Dropped Dropped Dropped Dropped
b0.001 b0.001
Abbreviations: AAD = ascending aortic dimensions; CI = confidence interval. a All parameters (p b 0.10) were entered in a multivariate logistic regression.
Please cite this article as: Y. Maeno, et al., Effect of ascending aortic dimension on acute procedural success following self-expanding transcatheter aortic valve replacement, Int J Cardiol (2017), http://dx.doi.org/10.1016/j.ijcard.2017.05.120
4
Y. Maeno et al. / International Journal of Cardiology xxx (2017) xxx–xxx
Fig. 2. Importance optimally oversizing for large AAD Optimally oversized SE-valves (N16.2%) had high device success rates for large AAD. Abbreviations: AAD = ascending aorta dimensions; SE = self-expanding.
implantation (AUC 0.72, sensitivity 75.0%, specificity 40.1%). Overall, 45.3% (97 patients) of the patients had an AAD N32.1 mm. For each valve size, sub-analysis for the thresholds of AAD using ROC curve analysis are as following; no patients with unsuccessful device implantation for 23 mm valve, 30.8 mm for 26 mm valve (AUC 0.75), 32.3 mm for 29 mm valve (AUC 0.67), and 34.8 mm for 31 mm valve (AUC 0.61). On the other hand, % oversizing below 16.2 was identified as suboptimal oversizing for unsuccessful device implantation (AUC 0.74, sensitivity 78.1%, specificity 45.1%). Overall 107 of 214 patients (50.0%) were below this threshold (under 16.2% oversizing). Combination of these two risk factors was associated with an exponential fall in device success rate, ranging from only 75.3% (large AAD alone) or 76.6% (small % oversizing alone) to 64.2% (both); this observation was robust for SE-devices. For patients with large AAD (N32.1 mm), optimal oversizing was associated with significantly higher device success rates than suboptimal oversizing (88.6% vs. 64.2%, p = 0.005) (Fig. 2). Of patients with optimally oversized SE-TAVR with large AAD, we experienced 5 cases with unsuccessful SE-device implantation. Two of them had need for a 2nd valve, and the remaining 3 patients had moderate PVL (2 patients) or death related to procedure (1 patient who had also moderate PVL). 4. Discussion Our findings can be summarized as follows: (i) large AAD and small degree of oversizing were confirmed to be the most relevant predictors of unsuccessful device implantation following SE-TAVR. (ii) a dramatic importance for optimal oversizing was noted, particularly with large AAD. Very few systematic data is available regarding acute device success after SE-TAVR. Device success is an important endpoint that has been associated with decreased mortality and TAVR-related complications in comparison to patients with unsuccessful device implantation [17]. Previous multicenter prospective the CHOICE trials reported device success rates following SE-TAVR as 77.5% [5]. On the other hand, a recently published the US CoreValve trial reported that acute devices success was 86.8% [16]. The present study (85.0%) was similar to that reported in the US CoreValve trial. The effect of increased AAD on acute device success following SE-TAVR has not previously been systematically examined. Nonetheless, a large AAD is a relative contraindication for CoreValve implantation according to the manufacturer's instructions. Moreover, Leipsic et al. suggested that AAD should be below 43 mm following SE-TAVR
[11]. Our findings demonstrated that lower degrees of oversizing and larger AAD were found to be independent predictors of acute procedural unsuccessful device implantation for SE- valve implantations. Lower degree of oversizing is a well-established independent predictor of PVL in prior studies and results in an incomplete attachment of the device to the aortic valvular complex wall [5,8,17]. This might be a cause of malaposition during valve deployment. On the basis of a hypothesis that the radial force of the outflow frame of SE-valves may be affected by AAD, our findings demonstrated that larger AAD was associated with decreased acute device success following SE-TAVR. Although the present study did not have sufficient power to show each threshold for different valve sizes due to limited number of this adverse event, the AAD threshold for unsuccessful each-sized device implantation increased linearly. Our findings imply that each labeled valve size has an optimal AAD. Therefore, future studies will be needed to validate this hypothesis and determine optimal AAD. There are several possible explanations for our findings. It is of extreme importance that the valve frame of SE-prosthesis, both at the outflow and inflow level needs to be closely attached to the wall of the aortic-valvular complex in order to acquire a stable device position. When the outflow stent frame is weakly attached to the aortic wall, valve stability is solely dependent on the inflow level. Furthermore, malaposition is affected by cardiac rhythm during valve deployment, particularly when the outflow frame is not closely attached to the aortic wall and especially in the presence of a lower degree of oversizing (Supplementary Fig. 2 shows a representative case). As a result, the need for a second valve, PVL, and/or valve dysfunction due to inadequate expansion may occur. The CHOICE trial showed that the use of SE-valves resulted in lower device success rates than the use of balloon-expandable (BE) valves [5]. In the setting of large AAD, the reduced anchoring due to larger aortic dimensions of the long stented SE-valve may contribute to increased risk of unsuccessful device implantation. In addition, AAD is less relevant for the shorter stent frame of balloon-expandable valves. Our findings also revealed that the combination of large AAD and lower degree of oversizing was associated with an exponential fall in device success rate. This latter indicates that the close attachment between valve frame and aortic valvular complex, particularly in both annulus- and AAD level, may be essential to increase anchoring. Tzamtzis et al. demonstrated that radial force of the SE-valve frame proportionately drops with the increasing diameter of the aorto-valvular complex [18]. Our findings are consistent with this study. In the present study, overall 6 of 10 patients who needed for a 2nd valve implantation had a large AAD and sub-optimally oversized SE-valve. On the other hand, overall
Please cite this article as: Y. Maeno, et al., Effect of ascending aortic dimension on acute procedural success following self-expanding transcatheter aortic valve replacement, Int J Cardiol (2017), http://dx.doi.org/10.1016/j.ijcard.2017.05.120
Y. Maeno et al. / International Journal of Cardiology xxx (2017) xxx–xxx
12 of 18 patients with moderate to severe PVL had a large AAD and suboptimally oversized SE-valve. Therefore, in patients with a large AAD, the optimal device sizing is essential to reduce PVL and avoid a second valve implantation. In a logistic regression analysis, the model demonstrates an important interaction of risk factors for device success, which may help to identify patients at risk for unsuccessful device implantation due to non-modifiable factors such as large AAD. It also emphasizes the importance of the modifiable factor of valve selection, which is oversizing with the dramatic difference of device success rates between suboptimal- and optimal oversizing. Optimal oversizing was especially important in the presence of a large AAD to increase acute device success rates following SE-valve implantations. Moreover, it becomes generally acceptable that SE-valves need larger oversizing compared to BE-valves in order to prevent PVL or unsuccessful device implantations [5,8]. Our findings support the results of these studies. In the present study, for patients with a large AAD, we experienced 5 cases of unsuccessful SE-valve implantation, despite an optimal oversizing. This may be attributable to other factors such as depth of valve implantation, aortic root angulation, and calcification of the aortic-valvular complex – all which have been identified as predictors of PVL in prior studies [7,14,15,19]. Previous studies have demonstrated that higher grades of leaflet calcification predicted increased PVL following SE-valve implantations [7,20]. On the other hand, lower degree of leaflet calcification has been associated with valve embolization and/ or need for a 2nd valve following SE-TAVR [21]. These contradictive effect of calcification grades on device success may explain why increased leaflet calcification, was not found to be an independent predictor for unsuccessful device implantation in the present study. Nonetheless, we did show that patients with unsuccessful device implantation had higher calcium volume in a univariable analysis mainly due to the more dominant effect of increased calcification on higher PVL grades. Finally, in the setting of a large AAD, optimally oversized SE-valve is an important factor contributing to a high rate of device success. In addition, in order to achieve good clinical outcomes, our findings suggest that AAD should be assessed pre-TAVR along with aortic annulus measurements. In clinical practice, for patients with large AAD, upsizing of SE valve should be considered, especially in cases of borderline sizing.
4.1. Study limitations Several limitations of the present study should be addressed. The study represents a retrospective, multi-center. The findings are subject to selection bias and confounders. The relatively low number of events (32 patients with unsuccessful device implantation) mandated a method of multivariable analysis that may exclude relevant variables with a weaker predictive value; certain previously identified predictor of PVL or prosthesis patient mismatch after SE-TAVR [6,19], such as depth of valve implantation, is not available in this dataset. However, given the present study based on pre-procedural assessment, we believe that our findings provide the importance of AAD assessment for successful SE-TAVR. In addition, they once more take a look at the interaction between AAD and SE-valve to achieve device success.
5. Conclusions A large AAD and smaller degrees of oversizing were confirmed to be relevant predictors for unsuccessful device implantation following SE-TAVR. Although a large AAD is an unmodifiable risk factor, optimal oversizing (N 16.2) appears to be important in increasing acute device success rates for these patients. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.ijcard.2017.05.120.
5
Conflict of interest Dr. Watanabe and Dr. Sharma is a proctor for Edwards Lifesciences. Dr. Jilaihawi is a consultant for Edwards Lifesciences, St. Jude Medical, and Venus MedTech. Dr. Makkar has received grant support from Edwards Lifesciences and St. Jude Medical; and is a consultant for Abbott Vascular, Cordis, and Medtronic; and holds equity in Entourage Medical. All other authors declare having no potential conflict of interests. Acknowledgement of grant support Dr. Makkar has received grant support from Edwards Lifesciences (Irvine, California, USA) and St. Jude Medical (St. Paul, Minnesota, USA). References [1] D.H. Adams, J.J. Popma, M.J. Reardon, S.J. Yakubov, J.S. Coselli, G.M. Deeb, et al., Transcatheter aortic-valve replacement with a self-expanding prosthesis, N. Engl. J. Med. 370 (19) (May 8 2014) 1790–1798. [2] J.J. Popma, D.H. Adams, M.J. Reardon, S.J. Yakubov, N.S. Kleiman, D. Heimansohn, et al., Transcatheter aortic valve replacement using a self-expanding bioprosthesis in patients with severe aortic stenosis at extreme risk for surgery, J. Am. Coll. Cardiol. 63 (19) (May 20 2014) 1972–1981. [3] N.E. Moat, P. Ludman, M.A. de Belder, B. Bridgewater, A.D. Cunningham, C.P. Young, et al., Long-term outcomes after transcatheter aortic valve implantation in high-risk patients with severe aortic stenosis: the U.K. TAVI (United Kingdom Transcatheter Aortic Valve Implantation) Registry, J. Am. Coll. Cardiol. 58 (20) (Nov 8 2011) 2130–2138. [4] M. Gilard, H. Eltchaninoff, B. Iung, P. Donzeau-Gouge, K. Chevreul, J. Fajadet, et al., Registry of transcatheter aortic-valve implantation in high-risk patients, N. Engl. J. Med. 366 (18) (May 3 2012) 1705–1715. [5] M. Abdel-Wahab, J. Mehilli, C. Frerker, F.-J. Neumann, T. Kurz, R. Tölg, et al., Comparison of balloon-expandable vs self-expandable valves in patients undergoing transcatheter aortic valve replacement: the CHOICE randomized clinical trial, JAMA 311 (15) (Apr 16 2014) 1503–1514. [6] M.A. Sherif, M. Abdel-Wahab, B. Stöcker, V. Geist, D. Richardt, R. Tölg, et al., Anatomic and procedural predictors of paravalvular aortic regurgitation after implantation of the Medtronic CoreValve bioprosthesis, J. Am. Coll. Cardiol. 56 (20) (Nov 9 2010) 1623–1629. [7] M. Seiffert, B. Fujita, M. Avanesov, C. Lunau, G. Schön, L. Conradi, et al., Device landing zone calcification and its impact on residual regurgitation after transcatheter aortic valve implantation with different devices, Eur. Heart J. Cardiovasc. Imaging 17 (5) (May 2016) 576–584. [8] D. Dvir, J.G. Webb, N. Piazza, P. Blanke, M. Barbanti, S. Bleiziffer, et al., Multicenter evaluation of transcatheter aortic valve replacement using either SAPIEN XT or CoreValve: degree of device oversizing by computed-tomography and clinical outcomes, Catheter. Cardiovasc. Interv. 86 (3) (Sep 2015) 508–515. [9] A.P. Kappetein, S.J. Head, P. Généreux, N. Piazza, N.M. van Mieghem, E.H. Blackstone, et al., Updated standardized endpoint definitions for transcatheter aortic valve implantation: the Valve Academic Research Consortium-2 consensus document, J. Am. Coll. Cardiol. 60 (15) (Oct 9 2012) 1438–1454. [10] S. Achenbach, V. Delgado, J. Hausleiter, P. Schoenhagen, J.K. Min, J.A. Leipsic, SCCT expert consensus document on computed tomography imaging before transcatheter aortic valve implantation (TAVI)/transcatheter aortic valve replacement (TAVR), J. Cardiovasc. Comput. Tomogr. 6 (6) (Dec 2012) 366–380. [11] J. Leipsic, R. Gurvitch, T.M. Labounty, J.K. Min, D. Wood, M. Johnson, et al., Multidetector computed tomography in transcatheter aortic valve implantation, JACC Cardiovasc. Imaging 4 (4) (Apr 2011) 416–429. [12] Y. Watanabe, M.-C. Morice, E. Bouvier, T. Leong, K. Hayashida, T. Lefèvre, et al., Automated 3-dimensional aortic annular assessment by multidetector computed tomography in transcatheter aortic valve implantation, JACC Cardiovasc. Interv. 6 (9) (Sep 2013) 955–964. [13] H. Jilaihawi, M. Kashif, G. Fontana, A. Furugen, T. Shiota, G. Friede, et al., Cross-sectional computed tomographic assessment improves accuracy of aortic annular sizing for transcatheter aortic valve replacement and reduces the incidence of paravalvular aortic regurgitation, J. Am. Coll. Cardiol. 59 (14) (Apr 3 2012) 1275–1286. [14] H. Jilaihawi, R.R. Makkar, M. Kashif, K. Okuyama, T. Chakravarty, T. Shiota, et al., A revised methodology for aortic-valvar complex calcium quantification for transcatheter aortic valve implantation, Eur. Heart J. Cardiovasc. Imaging 15 (12) (Dec 2014) 1324–1332. [15] Y. Abramowitz, Y. Maeno, T. Chakravarty, Y. Kazuno, N. Takahashi, H. Kawamori, et al., Aortic angulation attenuates procedural success following self-expandable but not balloon-expandable TAVR, JACC Cardiovasc. Imaging 9 (8) (Aug 2016) 964–972. [16] J.J. Popma, M.J. Reardon, S.J. Yakubov, J.B. Hermiller, J.K. Harrison, T.G. Gleason, et al., Safety and efficacy of self-expanding TAVR in patients with aortoventricular angulation, JACC Cardiovasc. Imaging 9 (8) (Aug 2016) 973–981. [17] M. Abdel-Wahab, T. Comberg, H.J. Büttner, M. El-Mawardy, K. Chatani, M. Gick, et al., Aortic regurgitation after transcatheter aortic valve implantation with balloon- and self-expandable prostheses: a pooled analysis from a 2-center experience, JACC Cardiovasc. Interv. 7 (3) (Mar 2014) 284–292.
Please cite this article as: Y. Maeno, et al., Effect of ascending aortic dimension on acute procedural success following self-expanding transcatheter aortic valve replacement, Int J Cardiol (2017), http://dx.doi.org/10.1016/j.ijcard.2017.05.120
6
Y. Maeno et al. / International Journal of Cardiology xxx (2017) xxx–xxx
[18] S. Tzamtzis, J. Viquerat, J. Yap, M.J. Mullen, G. Burriesci, Numerical analysis of the radial force produced by the Medtronic-CoreValve and Edwards-SAPIEN after transcatheter aortic valve implantation (TAVI), Med. Eng. Phys. 35 (1) (Jan 2013) 125–130. [19] H. Jilaihawi, D. Chin, T. Spyt, M. Jeilan, M. Vasa-Nicotera, J. Bence, et al., Prosthesispatient mismatch after transcatheter aortic valve implantation with the Medtronic-Corevalve bioprosthesis, Eur. Heart J. 31 (7) (Apr 2010) 857–864. [20] G. Tarantini, V. Gasparetto, M. Napodano, C. Fraccaro, G. Gerosa, G. Isabella, Valvular leak after transcatheter aortic valve implantation: a clinician updateon
epidemiology, pathophysiology and clinical implications, Am. J. Cardiovasc. Dis. 1 (3) (2011) 312–320. [21] N.M. Van Mieghem, C.J. Schultz, R.M.A. van der Boon, R.-J. Nuis, A. Tzikas, M.L. Geleijnse, et al., Incidence, timing, and predictors of valve dislodgment during TAVI with the Medtronic Corevalve system, Catheter. Cardiovasc. Interv. 79 (2012) 726–732.
Please cite this article as: Y. Maeno, et al., Effect of ascending aortic dimension on acute procedural success following self-expanding transcatheter aortic valve replacement, Int J Cardiol (2017), http://dx.doi.org/10.1016/j.ijcard.2017.05.120