The Impact of Aortic Valvular Calcium on Transcatheter Heart Valve Distortion

Objectives To investigate the relationship between the eccentric calcification of aortic valve and transcatheter heart valve (THV) distortion and the impact of THV distortion on echo parameters and clinical outcomes. Background The effects of eccentric calcification of the aortic valve on the THV distortion and the relationship between THV distortion and clinical impact were not fully understood. Methods Patients with symptomatic severe aortic stenosis who were undergoing THV implantation were enrolled. Patients underwent preprocedural, postprocedural multislice computed tomography (MSCT), and follow-up transthoracic echocardiogram (TTE). Delta calcium score (ΔCS) is defined as the difference between the maximum and minimal calcium scores of the three cusps, while valve distortion score (VDS) is defined as the difference between the longest and shortest stent frame, as obtained using MSCT. Patients were divided into two groups according to ΔCS: “noneccentric calcification group” and “eccentric calcification group.” Results A total of 118 patients were enrolled (59 patients in noneccentric and 59 in eccentric calcification groups). VDS was significantly lower in the noneccentric calcification group than in the eccentric calcification group (1.31 ± 0.82 mm vs. 1.73 ± 0.76 mm, p=0.004). VDS was not associated with the degree of paravalvular leak (PVL) and aortic valvular mean pressure gradient (AVPG) at 30-day and 1-year follow-up TTE and the cumulative rates of all-cause death and rehospitalization at 2-year clinical follow-up. Conclusions Eccentric valvular calcification was associated with longitudinal THV distortion. However, THV distortion was not associated with PVL, AVPG, and adverse clinical events during midterm follow-up.


Introduction
Transcatheter aortic valve implantation (TAVI) is a wellestablished treatment for intermediate-to high-risk patients with severe aortic stenosis (AS) [1][2][3]. However, previous studies reported that high burden of aortic valve calcification is a risk factor for TAVI complications, such as annulus rupture and paravalvular leak (PVL) [4,5]. Calcification is also associated with the following factors: higher rates of adverse events during percutaneous coronary interventions (PCI), peripheral vascular interventions, and endovascular therapy for abdominal aortic aneurysms [6][7][8][9][10]. Some previous small studies investigated the distortion of transcatheter heart valve (THV) and its impact on THV functions [11,12]. However, the effects of eccentric calcification of the aortic valve on the distortion of THV and the impact of THV distortion on THV functions and clinical impact were not fully understood. is study aimed to investigate the relationship between the degree and eccentricity of aortic valve calcification and THV distortion and the impact of THV distortion on follow-up echo parameters and clinical outcomes after TAVI.

Study Population and Procedures.
Symptomatic patients diagnosed with severe AS who were undergoing TAVI were enrolled in this study. AS severity was defined according to the Valve Academic Research Consortium-2 criteria [13]. All TAVI procedures were performed using the third-generation balloon expandable Edwards SAPIEN3 heart valve system (ES3, Edwards Lifesciences, Irvine, California) with fluoroscopic and echo guidance (transthoracic or transesophageal). All patients underwent transfemoral placement of the THVs, sized 20, 23, 26, and 29 mm. THV size was selected using multislice computed tomography (MSCT) [14]. Heparin was administered to all patients during the procedure. Moreover, antiplatelet therapy (aspirin or clopidogrel or both) or anticoagulant therapy was administered before and after the procedure.

Computed Tomography Analysis and Definition.
All patients underwent pre-and post-TAVI MSCT using the SOMATOM Definition AS + CT system with 64 detectors (SIEMENS Healthcare, Erlangen, Germany).
e MSCTs performed before TAVI were contrast-enhanced. Scans were performed using electrocardiography synchronized data during acquisition with a systolic phase requirement. Aortic valve calcium scores were measured for each leaflet using the pre-TAVI MSCT image. To calculate aortic valve calcium scores, calcium volume score method was used [15,16]. We applied the calcium volume score (not the Agatston score) because calcium volume score does not rely on lesion density, and instead seeking aims to estimate the true calcium volume. us, we believed the calcium volume score was more appropriate for our study. Delta calcium score (ΔCS) is defined as the difference between the maximum and minimal three-cusp calcium scores. Patients were divided into two groups according to the ΔCS: those with lowerthan-median ΔCS were assigned to the "noneccentric calcification group," whereas those with a higher-than-median ΔCS were assigned to the "eccentric calcification group" (median ΔCS � 148.5 mm 3 ).
e MSCTs of post-TAVI procedures were usually performed before discharge and without use of contrast medium. Post-TAVI MSCT was used to create a 3D-volumerendered view of the THV. e stent frame length was measured at each cusp. Valve distortion score (VDS) refers to the difference between the longest and shortest stent frame lengths (Figures 1(a) and 1(b)). e nominal ES3 stent frame lengths are 15.5 mm (20 mm THV), 18 mm (23 mm THV), 20 mm (26 mm THV), and 22.5 mm (29 mm THV). All THV stent frames were geometrically measured for the maximal and minimal orthogonal diameters at the four levels of THV: (a) THV inflow, (b) native annulus, (c) mid-THV, and (d) THV outflow [17]. Cross-sectional eccentricity index (THV) was defined as follows: (1 − short diameter/long diameter) × 100 (Figure 1(c)) [17,18]. Valve oversizing ratio was calculated as follows: (observed THV external area at the native annulus level/native annular area − 1) × 100 [17]. All imaging constitutions and measurements were performed using a dedicated software (3mensio software; Pie Medical Imaging, Maastricht, Netherlands).

Study Objectives.
is study aimed to evaluate (1) the relationship between eccentric aortic valve calcification and THV distortion using 3D-MSCT imaging, (2) the relationship between THV distortion and THV functions such as PVL and aortic valvular mean pressure gradient (AVPG) using follow-up transthoracic echocardiogram (TTE), and (3) the impact of THV distortion on midterm (2-year) clinical outcomes.

Statistical Analysis.
Categorical data are presented as counts and percentages and are compared using the chisquare test or Fisher exact tests. All continuous variables were tested for normality using the Kolmogorov-Smirnov test. Continuous variables are presented as mean ± SD or median (interquartile range) for normally and nonnormally distributed variables. Between-group differences in continuous variables were compared using Student's t-test or Mann-Whitney U test. All data were analyzed using SPSS (version 25 for Windows; SPSS, Inc., Chicago Illinois) and R (version 3.0.2; e R Foundation for Statistical Computing Vienna, Austria).

Baseline Characteristics and Clinical Outcomes.
Between June 2016 and December 2018, 241 patients underwent TAVI. Among them, 139 patients underwent SA-PIEN3 implantation. Of these, 19 patients who did not undergo postprocedural MSCT and two patients who died before discharge (1 patient died from acute respiratory distress syndrome and one patient died from colitis) were excluded in this study. In total, 118 patients were included. Patients' baseline data are shown in Table 1.
is study included 42 males (35.6%) with mean age of 83.6 ± 5.1 years. e mean STS and Logistic Euro scores were 8.2 ± 7.5 and 21.0 ± 12.9, respectively. e prevalence of chronic obstructive pulmonary disease was significantly higher in the eccentric calcification group, whereas the rate of prior PCI was higher in the noneccentric calcification group. e prevalence of diabetes mellitus was shown to be higher in the eccentric calcification group. Table 2. Among the three cusps in the 118 patients, the highest calcium score was observed in the noncoronary cusps (255.4 ± 159.8 mm 3 ).

MSCT Parameters. Preprocedural MSCT parameters are shown in
Of these, 59 (50%) were assigned to the noneccentric calcification group (ΔCS < 148.5 mm 3 ), while 59 patients (50%) were assigned to the eccentric calcification group (ΔCS ≥ 148.5 mm 3 ). e total calcium score of all the cusps was significantly lower in the noneccentric calcification group compared with those of the eccentric group (429.6 ± 333.3 mm 3 vs. 733.9 ± 335.8 mm 3 , p < 0.001). Furthermore, the noncoronary cusp and the right-coronary cusp calcium scores were  Figure 1: Transcatheter heart valve distortion and multislice computed tomography analysis. A normal expanded transcatheter heart valve (THV) and distorted THV in a multislice computed tomography (MSCT) volume-rendered view (a). Using the postprocedural MSCT, a 3D-volume-rendered view of THV was constituted. e length of the stent frame on the center of each cusp was measured. Valve distortion score (VDS) was defined as the difference between the longest and shortest stent frame lengths (b). e THV stent frame was geometrically evaluated for maximal and minimal orthogonal diameters at the four levels of THV: THV inflow, native annulus, mid-THV, and THV outflow. e cross-sectional eccentricity index was defined as follows: (1 − short diameter/long diameter) × 100 (c). also significantly lower in the noneccentric calcification group compared with those of the eccentric calcification group. On average, patients underwent MSCT 5.7 days after TAVI. Table 2 shows the parameters of THV measured using postprocedural MSCT. VDS was significantly lower for the noneccentric calcification group than those in the eccentric calcification group (1.31 ± 0.82 mm vs. 1.73 ± 0.76 mm, p < 0.004) (Figure 2). e longest stent frames were most frequently observed in the noncoronary cusp, while the shortest stent frames were most often observed in the leftcoronary cusp. No significant between-group differences were observed in cross-sectional eccentricity indexes at any part of the THV levels (THV inflow, native annulus, mid-THV, and THV outflow). Table 3. No significant difference was observed in the THV size between noneccentric and eccentric calcification groups. One patient needed a second valve due to valve migration. e most frequently used THVs were 23 mm in both groups. Predilatation and postdilatation were performed in 63 (53.4%) and 31 (26.3%) patients, respectively. e rate of predilatation was lower in the noneccentric group, compared with those of the eccentric calcification group. Nominal delivery balloon volume, overfilling volume (≥1 mL additional volume), and underfilling volume (≥1 mL less volume) were used in 63 (53.4%), two (1.7%), and 53 (44.9%) patients, respectively. No significant difference was observed in the delivery balloon volume between the two groups. Table 4. No significant difference was observed in the aortic valve area (AVA) and AVPG between noneccentric and eccentric calcification groups using the preprocedural TTE (AVA; 0.66 ± 0.15 cm 2 vs. 0.64 ± 0.14 cm 2 , p � 0.458, AVPG; 46.6 ± 20.2 mmHg vs. 49.7 ± 17.2 mmHg, p � 0.376). e noneccentric calcification group had a less maximum aortic gradient than the eccentric calcification group (75.4 ± 26.0 mmHg vs. 87.1 ± 26.9 mmHg, p � 0.018).

TTE Parameters. TTE parameters are shown in
No significant difference was observed in the degree of PVL and AVPG between noneccentric calcification group and eccentric calcification group at 30-day follow-up TTE (Table 4). In addition, VDS was not associated with the degree of PVL and AVPG (Figures 3(a) and 3(c)).

Journal of Interventional Cardiology
One-year follow-up TTE data were obtained in 89 (75.4%) patients. No significant difference was observed in the degree of PVL and AVPG between noneccentric calcification group and eccentric calcification group, and VDS was not associated with the degree of PVL and AVPG at 1-year follow-up TTE (Table 4 and Figures 3(b) and 3(d)).   Figure 4 shows Kaplan-Meier curves of all-cause death free and rehospitalization free survival rates.

Discussion
Findings showed that (1) eccentric calcification was associated with greater longitudinal THV distortion than noneccentric calcification of the aortic valve ( Figure 5). On the other hand, (2) eccentric calcification was not associated with the cross-sectional THV eccentricity. In addition, (3) longitudinal THV distortion was not associated with PVL grade and AVPG at 30-day and 1-year TTE. Moreover, (4) longitudinal THV distortion did not affect the cumulative rates of adverse clinical events (all-cause death, rehospitalization, and valve thrombosis) at 2-year clinical follow-up.

Eccentric Calcification and THV Distortion.
Studies have revealed that eccentric calcification caused device deformation and adverse events in procedures other than TAVI. During PCI procedures, stent implantation in severely calcified lesions can result in less optimal stent expansion, mispositioning, and stent fracture [6][7][8]19]. In addition, calcified plaques are associated with side branch stenosis after main branch stenting due to stent migration to the side opposite to the side where there is severe calcification [20]. Eccentric calcification is also associated with increased risk of endoleak in the procedure of endovascular therapies for abdominal aortic aneurysms and hemodynamic instability in the procedure of carotid angioplasty [21,22]. us, we evaluated the effects of eccentric calcification of the aortic 0.0~0. 9 1.0~1.9 2.0~2. valve on the distortion of THV and the impact of THV distortion on THV functions and clinical impact in current study. Some potential mechanisms of these THV distortions were considered in this study. Severe calcification sites push the THV toward the mild calcification sites. Furthermore, because the THV delivery balloon is semicompliant, the delivery balloon may expand in a nonuniform manner within the eccentric calcification site. e cells of the THV stent frame are fully opened in mildly calcified sites, but not in sites with severe calcification. us, longitudinal THV distortions can possibly occur in aortic valves with eccentric calcification.
No significant between-group difference was observed in the cross-sectional eccentricity index between the noneccentric and eccentric calcification groups. is might be due to improvements in radial strength as devices used evolved (ES3 has larger cells and wider strut angles than the SAPIEN XT; Edwards Lifesciences, Irvine, California) [23].

Impacts of THV Distortion on TTE Parameters and Clinical
Outcomes. In spite of these adverse effects related to device deformation due to severe eccentric calcification, THV distortion did not affect the severity of PVL and AVPG, as measured during the 30-day follow-up TTE in this study. is significant PVL (>moderate) low rate may be due to the improvements made to the THV device. e S3 has an outer polyethylene terephthalate cuff, which prevents significant PVL [23].
Additionally, the THV distortion was not associated with TTE parameters (degree of PVL and AVPG) at 1-year follow-up. Previous studies reported that THV has good durability in mid-to long-term follow-ups [24][25][26][27]. However, the association between THV distortion due to eccentric calcification and THV durability is not well known.
is study revealed that THV maintains excellent functions regardless of longitudinal THV distortion in midterm follow-up TTE.
Moreover, this study revealed that the degree of THV distortion was not associated with the following adverse events: cumulative rates of all-cause death, rehospitalization for valve-related symptoms or worsening congestive heart failure, and valve thrombosis at 2-year clinical follow-up. e fact that THV kept good functions regardless of THV  distortion might contribute to these positive clinical outcomes. ere are some reports which evaluated the durability of THV [28]. Most of them showed the favorable durability of THV. However, the relationship between THV function (durability) and severe eccentric calcification or THV distortion has not been fully elucidated. In addition, Blackman et al. reported that median duration of moderate structural valve degeneration was about 6 years following TAVI [24]. us, we cannot easily conclude that THV distortion does not affect the THV function, although current study added the novel information that severe eccentric calcification was associated with longitudinal THV distortion, but longitudinal THV distortion was not associated with THV dysfunction and adverse clinical events during midterm follow-up. Studies with large number of patients and long-term follow-up are warranted.

Study Limitations.
is study has several limitations. First, this was an observational single-center study. Second, this study was conducted with a relatively small patient cohort. ird, patients without postprocedure MSCT scans were excluded in this study. erefore, selection bias cannot be excluded. Fourth, this study only evaluated the relationship between the eccentricity of calcification in aortic valvular leaflet and THV distortion. us, we cannot exclude the possibility that calcium outside the valve (such as left ventricular outflow tract) affected the THV distortion. Fifth, we focused on the relationship between degree of VDS and PVL in the current study. However, there is a possibility that other confounding factors might affect the degree of PVL. Sixth, the number of patients with VDS larger than 2 mm was relatively small (25.4%). us, we cannot exclude the possibility that the results were affected by the small number of them. To solve this problem, further multicenter registry with large number of patients is needed. Finally, this study was conducted with only midterm follow-up. erefore, whether this distortion of THV impacts clinical outcomes or not is unknown within the context of long-term follow-up. Large numbers of patients and longer follow-up studies are needed. Several prospective randomized trials (the PART-NER 3 trial, the Medtronic Evolut low-risk trial) showed excellent short-term outcomes after TAVI in low-risk patients [29,30]. Based on these trials, indication of TAVI is expected to expand. Before that, long-term results need to be evaluated whether THV distortions are associated with deterioration or adverse clinical outcomes.

Conclusions
Eccentricity of aortic valvular calcification was associated with longitudinal, but not cross-sectional, distortion of the THV stent frame. However, this longitudinal THV distortion was not associated with the degree of PVL and AVPG, and clinical adverse events in midterm follow-up.

Data Availability
e data used to support the findings of this study are included within the article.

Conflicts of Interest
Dr. Naganuma is a clinical proctor for Edwards Lifesciences and Medtronic. All other authors have no relationships relevant to the contents of this paper to disclose.