Subclinical valve thrombosis, radiographically defined as hypo-attenuated leaflet thickening (HALT) on computed tomography angiography (CTA), may impact bioprosthetic valve durability and outcomes following transcatheter aortic valve replacement (TAVR). This study aimed to evaluate HALT incidence, identify distinct predictors in bicuspid (BAV) versus tricuspid (TAV) aortic valves, and assess its clinical and hemodynamic implications. A total of 508 patients (285 BAV, 223 TAV) were included in the analysis. HALT incidence was comparable between groups (13.6% [39/285] vs 13.0% [29/223], p = 0.823), though severity and predictors differed. TAV patients showed greater leaflet thickening (2.7 ± 1.5 mm vs 1.9 ± 1.1 mm, p = 0.04). In BAV, independent predictors included elevated D-dimer, deeper transcatheter heart valve (THV) implantation, higher THV nadir eccentricity index, and asymmetric leaflet expansion. In TAV, predictors included elevated D-dimer, lower early peak transvalvular velocity, and asymmetric leaflet expansion. HALT was not associated with 1-year transvalvular gradients or left ventricular ejection fraction in either group. Clinically, HALT was associated with higher cerebrovascular event rates (2.9% vs 0.5%, p = 0.031) and an increased risk of major adverse cardiac and cerebrovascular events (MACCE) (13.2% vs 5.7%, p = 0.036; HR = 2.24). In conclusion, HALT incidence is comparable between BAV and TAV patients, but predictors differ: THV expansion characteristics are dominant in BAV, while hemodynamic parameters prevail in TAV. Elevated D-dimer is a consistent predictor across both groups. Although HALT does not affect mid-term hemodynamics, its potential association with higher MACCE risk warrants long-term surveillance and individualized management strategies.
Transcatheter aortic valve replacement (TAVR) has emerged as an effective alternative for the treatment of severe aortic stenosis (AS). While advancements in device technology and procedural techniques have significantly reduced perioperative complications, recent studies highlight the frequent occurrence of subclinical valve thrombosis post-TAVR, radiographically characterized as hypo-attenuated leaflet thickening (HALT). This phenomenon has raised considerable clinical concern. Longitudinal data suggest that HALT may accelerate bioprosthetic valve degeneration and compromise durability. HALT often coexists with reduced leaflet motion (RELM), which can decrease the effective orifice area and increase transvalvular pressure gradients. Moreover, HALT may serve as a precursor to clinically overt valve thrombosis and thromboembolic events. As TAVR is increasingly applied to low-risk and relatively younger patients, the number of BAV patients receiving TAVR has been growing. Emerging evidence indicates that transcatheter heart valves (THVs) in BAV patients frequently assume an elliptical configuration or incomplete stent expansion, potentially altering HALT risk profiles compared to tricuspid aortic valve (TAV) patients. This study aims to evaluate HALT incidence patterns, identify independent HALT predictors in BAV versus TAV populations, and assess HALT’s impact on hemodynamic and clinical outcomes.
Methods
This was a single-center study conducted at West China Hospital, Sichuan University. All data were prospectively collected from July 1, 2019 to August 31 2023. After clinical evaluation by the heart team, patients aged 18-100 years with severe aortic stenosis and suitable anatomy for self-expanding TAVR valves were included. Patients unable to complete CTA due to medical (e.g., risk of contrast-induced nephropathy) or nonmedical reasons (e.g., refusal), as well as those with poor image quality or rare type-2 BAV, were excluded. All patients provided informed consent for the use of their medical records for research purposes. They underwent TAVR with self-expanding valves featuring a supra-annular design. The majority received VenusA series (Venus MedTech, Hangzhou, China; 71.1%), followed by Taurus series (Peijia Medical, Suzhou, China; 13.8%), VitaFlow series (MicroPort, Shanghai, China; 11.0%) and Evolut series (Medtronic, Minneapolis, MN, USA; 4.1%). Procedures were performed predominantly via transfemoral access under conscious sedation and local anesthesia. After aortic valve crossing and left ventricular wire placement, the delivery system was advanced under fluoroscopic guidance. The decision of postdilation at the discretion of operators, typically in the presence of significant paravalvular leak and/or residual gradient due to incomplete apposition and/or under-expansion. Antithrombotic therapy followed the relevant guidelines and expert consensus. ,
CTA image analysis
Preoperative CTA were done within 30 days before the TAVR procedure, the main items that be evaluated in this study included ( Figure 1 ): (1) Anatomical classification of valve type, firstly determine the valve type was BAV or TAV, if it was BAV, next classifying BAV into type-0, type-1 based on Sievers classification (rare Type-2 BAV excluded). (2) Annulus calcification volume, used the standard of 850Hu to calculate the annulus level calcification degree. (3) Aortic root assessment: assessed the native valve thickening degree (heavy thicken or not) and measured the eccentricity index (EI, EI = minimum diameter/ maximum diameter), perimeter and area of the plane of annulus, left ventricular outflow tract (LVOT), sinus of Valsalva (SOV), Sino-tubular junction (STJ) and ascending aorta (AAO, at 40mm). (4) Annulus angle, measured the angle between annulus and the horizontal line. (5) Left and right coronary artery height, measured the perpendicular distance from the coronary orifice to annulus plane.
Aortic root geometric parameters. (A: Overview of anatomical planes of the aortic root; B: Aortic valve annulus; C: Left ventricular outflow tract; D: sinus of Valsalva; E: Sino-tubular junction; F: Ascending aorta; G: Annular angulation; H: Heights of left and right coronary ostia)
Postoperative CTA were all completed within 30 days after TAVR. Morphologic assessment of the THV focused on 4 anatomical landmarks: frame inflow plane, native annular plane, leaflet nadir plane, and leaflet coaptation plane ( Figures 2 and 3 ). Quantitative analysis included: (1) THV expansion ratio and deformation degree at these 4 planes: expansion ratio: calculated as internal cross-sectional area/nominal area at each anatomical level. Deformation degree: evaluated through EI using the formula: EI = minimum diameter/maximum diameter). (2) Implant depth: defined as the mean perpendicular distance between the basal hinge points of native aortic valve cusps and the THV inflow plane. (3) Coronary alignment assessment: coronary ostial-commissural mismatch was measured as the acute angle formed between the coronary ostium centerline and its closest THV commissural post. (4) Asymmetrical leaflet expansion: represented by the angular disparity between maximum and minimum leaflet opening angles at the coaptation plane. (5) Sinus Flow Channel Evaluation: Neo-sinus width was measured as the maximum transverse diameter between the THV stent frame and native aortic wall within the Valsalva sinus.
Postimplantation status of transcatheter heart valve (THV). (A: Overview of THV anatomical planes; B: THV inflow plane; C: Native annular plane; D: THV leaflet nadir plane; E: THV leaflet coaptation plane; F: Implantation depth)
Postimplantation characteristics of transcatheter heart valves. (A: Bioprosthetic leaflet opening angle; B: Coronary alignment degree; C: Neo-sinus width)
Endpoints
The primary effectiveness endpoint was the occurrence of HALT ( Figure 4 ). Diagnostic criteria included localized or diffuse leaflet thickening with hypoattenuation on short-axis views, typically presenting as wedge-shaped or crescentic morphology. On long-axis views, thickening extended from the leaflet base to the free edge. After finding the classic HALT appearance, we recorded mean leaflet thickness and total hypo-attenuated area ( Figure 5 ). THV hemodynamics was the second endpoint. Noninvasive TTE was performed at 1 month, 6 months, 1 year after TAVR to assess the hemodynamic changes. 1-year incidence of all-cause mortality, heart failure rehospitalization, stroke or TIA, myocardial infarction and the composite endpoint (major adverse cardiac and cerebrovascular events, MACCE) was the third effectiveness endpoints.
Identification of hypo-attenuated leaflet thickening (HALT) on Computed Tomography Angiography (CTA) imaging. ( A–C: Transverse CTA views of HALT showing hallmark wedge-shaped or crescentic morphologies; D–F: Longitudinal CTA slices reveal HALT lesions spanning from the base to the free margin of the leaflets.)
Hypo-attenuated leaflet thickening (HALT) Area and Thickness Measurement. (A: HALT area measured in transverse plane; HALT thickness measured in longitudinal view.)
Statistical methods
Continuous variables were expressed as mean ± SD or median and tested by unpaired t test or analysis of variance. Categorical variables were expressed in frequency and percentage, evaluated by Chi-square test or Fisher test appropriately. HALT predictors were identified using univariate regression, with significant variables (p <0.05) entered into multivariable models. Cox regression assessed associations between HALT and 1-year outcomes, reported as hazard ratios with 95% CIs. Analyses used SPSS v27.0.
Results
Study population and occurrence of HALT
A total of 508 patients were included in this study ( Figure 6 ), 223 patients with TAV and 285 patients with BAV. In our cohort, HALT occurred in 39/285 BAV patients (13.6%). Among them, 19 were type-1 BAV (14.1%) and 20 were type-0 BAV (13.2%). TAV patients had HALT in 29/223 cases (13.0%). HALT incidence did not differ between BAV and TAV patients (13.6% vs 13.0%, p = 0.823). HALT was more severe in TAV patients ( Table 1 ), with greater mean thickness (2.7 ± 1.5mm vs 1.9 ± 1.1mm, p = 0.04) and larger total area (31.9 ± 17.2mm² vs 26.1 ± 11.6mm², p = 0.14).
Study flowchart. Between 2019 and 2023, a total of 746 patients underwent transcatheter aortic valve replacement for aortic stenosis and completed 1-year follow-up. Among them, 238 were excluded based on predefined exclusion criteria, leaving 508 patients included in the final analysis.
Table 1
HALT severity
| TAV (n = 29) | BAV (n = 39) | p-value | |
|---|---|---|---|
| Total thickness, mm | 2.7 ± 1.5 | 1.9 ± 1.1 | 0.04 |
| Total area, mm² | 31.9 ± 17.2 | 26.1 ± 11.6 | 0.14 |
BAV= bicuspid aortic valve; HALT= hypo-attenuated leaflet thickening; TAV= tricuspid aortic valve.
Baseline and procedural characteristics
Table 2 delineates baseline and procedural characteristics stratified by HALT status. In BAV cohorts, HALT patients demonstrated a higher prevalence of hypertension (51.2% [20/39] vs 32.5% [80/246], p = 0.023) and lower preoperative left ventricular ejection fraction (LVEF) (53.9 ± 16.1% vs 58.2 ± 14.0%, p = 0.076). In TAV cohorts, HALT patients were older (75.7 ± 9.1 years vs 73.0 ± 7.4 years, p = 0.082), and had lower mean pressure gradients (46.1 ± 15.4 mmHg vs 52.8 ± 17.2 mmHg, p = 0.047) and slower peak transaortic velocities (4.3 ± 0.7 m/s vs 4.6 ± 0.7 m/s, p = 0.083). Although anticoagulant use was higher in non-HALT patients overall (16.8 % vs 7.3 %, p = 0.045), this difference disappeared after stratification by valve morphology.
Table 2
Baseline and procedural characteristics
| BAV (= 285) | TAV (n=223) | ALL(n=508) | |||||||
|---|---|---|---|---|---|---|---|---|---|
| HALT (n = 39) | Non-HALT (n = 246) | p-value | HALT (n = 29) | non-HALT (n = 194) | p-value | HALT (n = 68) | non-HALT (n = 440) | p-value | |
| Age, y | 72.3 ± 7.7 | 70.6 ± 7.5 | 0.192 | 75.7 ± 9.1 | 73.0 ± 7.4 | 0.082 | 73.7 ± 8.4 | 71.6 ± 7.5 | 0.039 * |
| Male, n (%) | 19 (48.7) | 149 (60.5) | 0.162 | 17 (58.6) | 103 (53) | 0.578 | 36 (52.9) | 252 (57.2) | 0.502 |
| BMI, kg/m² | 22.9 ± 3.3 | 23.2 ± 3.3 | 0.587 | 22.7 ± 2.9 | 22.7 ± 3.6 | 0.995 | 22.8 ± 3.1 | 23.0 ± 3.4 | 0.704 |
| Hypertension, n (%) | 20 (51.2) | 80 (32.5) | 0.023* | 15 (51.7) | 95 (48.9) | 0.782 | 35 (51.4) | 175 (39.7) | 0.068 |
| Diabetes mellitus, n (%) | 6 (15.3) | 53 (21.5) | 0.378 | 2 (6.8) | 32 (16.4) | 0.18 | 8 (11.7) | 85 (19.3) | 0.134 |
| Hyperlipidemia, n (%) | 7 (17.9) | 51 (20.7) | 0.688 | 3 (10.3) | 33 (17) | 0.363 | 10 (14.7) | 84 (19) | 0.386 |
| COPD, n (%) | 6 (15.3) | 23 (9.3) | 0.247 | 3 (10.3) | 25 (12.8) | 0.7 | 9 (13.2) | 48 (10.9) | 0.572 |
| Atrial fibrillation, n (%) | 2 (5.1) | 27 (10.9) | 0.262 | 5 (17.2) | 41 (21.1) | 0.629 | 7 (10.2) | 68 (15.4) | 0.264 |
| CAD, n (%) | 9 (23) | 53 (21.5) | 0.829 | 7 (24.1) | 44 (22.6) | 0.862 | 16 (23.5) | 97 (22) | 0.784 |
| Previous PCI, n (%) | 6 (15.3) | 23 (9.3) | 0.247 | 1 (3.4) | 21 (10.8) | 0.214 | 7 (10.2) | 44 (10) | 0.94 |
| PAD, n (%) | 4 (10.2) | 15 (6) | 0.333 | 2 (6.8) | 14 (7.2) | 0.95 | 6 (8.8) | 29 (6.5) | 0.499 |
| Stroke or TIA, n (%) | 4 (10.2) | 31 (12.6) | 0.678 | 6 (20.6) | 27 (13.9) | 0.338 | 10 (14.7) | 58 (13.1) | 0.731 |
| Renal insufficiency, n (%) | 3 (7.6) | 11 (4.4) | 0.387 | 1 (3.4) | 19 (9.7) | 0.265 | 4 (5.8) | 30 (6.8) | 0.774 |
| Baseline platelet, 10 9 | 162.3 ± 59.6 | 161.8 ± 57.6 | 0.958 | 188.3 ± 55.3 | 168.2 ± 59.7 | 0.089 | 173.4 ± 58.8 | 164.6 ± 58.5 | 0.251 |
| NYHA Class III or IV, n (%) | 14 (35.8) | 75 (30.4) | 0.498 | 14 (48.2) | 75 (38.6) | 0.324 | 28 (41.1) | 150 (34) | 0.254 |
| STS-PROM score, % | 3.2 ± 2.5 | 2.9 ± 2.2 | 0.365 | 4.7 ± 4.5 | 3.6 ± 2.9 | 0.126 | 3.8 ± 3.4 | 3.2 ± 2.4 | 0.193 |
| Pre-TTE data | |||||||||
| Severe AR, n (%) | 6 (15.3) | 34 (13.8) | 0.794 | 15 (51.7) | 88 (45.3) | 0.521 | 21 (30.8) | 122 (27.7) | 0.59 |
| Severe MR, n (%) | 3 (7.6) | 31 (12.6) | 0.38 | 8 (27.5) | 38 (19.5) | 0.321 | 11 (16.1) | 69 (15.6) | 0.917 |
| Pre-TTE gradient, mmHg | 57.2 ± 23 | 60.1 ± 21.8 | 0.443 | 46.1 ± 15.4 | 52.8 ± 17.2 | 0.047* | 52.5 ± 20.8 | 56.9 ± 20.2 | 0.093 |
| Pre-TTE velocity, m/s | 4.7 ± 0.8 | 4.8 ± 0.8 | 0.596 | 4.3 ± 0.7 | 4.6 ± 0.7 | 0.083 | 4.6 ± 0.8 | 4.7 ± 0.8 | 0.163 |
| Pre-TTE LVEF, % | 53.9 ± 16.1 | 58.2 ± 14.0 | 0.076 | 54.9 ± 14.8 | 57.4 ± 13.2 | 0.335 | 54.3 ± 15.5 | 57.9 ± 13.6 | 0.073 |
| Procedure details | |||||||||
| Transfemoral access, n (%) | 37 (94.8) | 241 (97.9) | 0.202 | 27 (93.1) | 191 (98.4) | 0.387 | 64 (94.1) | 432 (98.1) | 0.062 |
| Pre-dilation, n (%) | 37 (94.8) | 233 (94.7) | 0.968 | 23 (79.3) | 171 (88.1) | 0.187 | 60 (88.2) | 404 (91.8) | 0.328 |
| THV size | 0.892 | 0.153 | 0.404 | ||||||
| 23-26, n (%) | 28 (71.7) | 174 (70.7) | 17 (58.6) | 139 (71.6) | 45 (66.1) | 313 (71.1) | |||
| 29-32, n (%) | 11 (28.2) | 72 (29.2) | 12 (41.3) | 55 (28.3) | 23 (33.8) | 127 (28.8) | |||
| Postdilation, n (%) | 20 (51.2) | 151 (61.3) | 0.232 | 7 (24.1) | 70 (36) | 0.207 | 27 (39.7) | 221 (50.2) | 0.106 |
| Valve-in-Valve, n (%) | 3 (7.6) | 10 (4) | 0.313 | 2 (6.8) | 10 (5.1) | 0.698 | 5 (7.3) | 20 (4.5) | 0.319 |
| Anticoagulation, n (%) | 2 (5.1) | 30 (12.2) | 0.194 | 3 (10.3) | 44 (22.6) | 0.129 | 5 (7.3) | 74 (16.8) | 0.045 * |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
