Valvular Heart Disease
Rate of Stenotic Bicuspid Aortic Valve Aortic Dilatation After Aortic Valve Replacement, Calculated Using a 3-Dimensional Reconstruction Tool
2017 Volume 81 Issue 8 Pages 1207-1212
Details
2017 Volume 81 Issue 8 Pages 1207-1212
Background: Progression of asymmetric dilated aorta associated with bicuspid aortic valve (BAV) is difficult to evaluate conventionally. The aim of the study was to calculate the rate of progression of the dilated BAV aorta after aortic valve replacement (AVR) using a 3-dimensional (3-D) reconstruction tool.
Methods and Results: Fourteen stenotic BAV and 14 stenotic tricuspid aortic valve (TAV) patients with mildly dilated ascending aorta were reviewed. A patient-specific 3-D aortic model was reconstructed from preoperative and postoperative computed tomography data (BAV, 2.5±1.9 years after AVR; TAV, 2.2±1.8 years after AVR). Aortic diameter, including the longest and shortest, was measured on the maximum perpendicular cross-section tangential to the 3-D centerline of the reconstructed model. The longest diameter was defined as that passing through the distal point of the greater curvature of the aorta. The shortest diameter was defined as perpendicular to the longest. The progression rates were compared between the BAV and TAV groups. The progression rate of ascending aortic diameter was greater for BAV (longest diameter, 1.02±1.03 vs. −0.075±0.78 mm/year, P<0.001; shortest diameter, 0.41±0.62 vs. −0.016±0.59 mm/year, P=0.003). The longest diameter of the proximal arch also grew more rapidly in the BAV group (P<0.001).
Conclusions: Ascending aortic dilatation with stenotic BAV progresses after AVR at a maximum rate of 1.02±1.03 mm/year. Expansion toward the greater curvature frequently progresses to the proximal arch.
Bicuspid aortic valve (BAV) is a common congenital heart malformation that occurs in 0.5–2% of the population.1,2 Patients with BAV are at increased risk of developing serious aortic complications, including aortic dilatation, dissection and rupture after reaching adulthood.2–4 The valve-related aortic complications, called BAV-related aortopathy, occur in 10–35%, and aortic dissection occurs in approximately 4% of patients with BAV.3 Patients with BAV and dilated aorta have a 9-fold increased risk of aortic dissection.5 On histology, aortas with BAV frequently show cystic medial degeneration, a characteristic histopathological defect of the aortic media.3,5
Ascending aortic dilatation associated with BAV frequently has an asymmetric configuration that bulges toward the right-anterior aspect of the aorta.6 The asymmetric enlargement might be caused by valve-related hemodynamic abnormalities, which lead to the asymmetric distribution of wall shear stress (WSS).7,8
Yasuda et al reported that ascending aortic dilatation associated with BAV sometimes progresses even after aortic valve replacement (AVR).9 They reported that the ascending aortic diameter increased at an annual rate of 0.18±0.08 mm/m2 after AVR.9 The annual progression rate of BAV aortas, however, remains controversial, because the diameter of asymmetric dilated aorta is difficult to measure accurately on conventional computed tomography (CT). We postulated that the progression of ascending aortic diameter in BAV patients depends on its direction, because the dilated ascending aorta with BAV is expected to progress more rapidly toward the right-anterior quadrant, where WSS should be greater.8 The aim of the present study was therefore to measure aortic diameter in the right-anterolateral direction in order to calculate an accurate progression rate of the asymmetric dilated ascending aorta after AVR. Aortic diameter was measured using 3-dimensional (3-D) reconstruction and an automated analysis tool, Mimics (Materialise, Leuven, Belgium), and the annual progression rate was compared between BAV and tricuspid aortic valve (TAV) patients.
Data derived from 28 patients (male, n=18; 64%; female, n=10; 36%; mean age, 66±9 years; range, 45–82 years) with aortic stenosis (AS) who presented with mild ascending aortic dilatation, defined as aortic diameter 33–45 mm, were retrospectively reviewed. All patients underwent AVR without the graft replacement of the ascending aorta at the present institution between January 2002 and March 2014. Twenty (71%) of the patients were treated only with AVR, and 8 (29%) were treated with concomitant surgery (single concomitant, n=6; triple concomitant, n=2) including coronary artery bypass grafting (n=4), mitral valve surgery (n=2), annuloplasty of the tricuspid valve (n=4), and resection of the left arterial appendage (n=1). Patients with infective endocarditis and suboptimal CT for reconstruction of 3-D models using Mimics were excluded from this study. Data from 14 patients with BAV and 14 with TAV were compared. The morphological BAV type was defined according to the Sievers’ classification. The cusp configuration was type 1 (R/L) in 5 (35.7%) of the patients, type 1 (R/N) in 4 (28.6%), type 1 (L/N) in 3 (21.4%), and type 0 (lat) cusp fusion in 2 (14.3%). In the BAV group, 9 patients had bioprosthetic valves, including the Carpentier-Edwards Perimount valve (CEP; Edwards Lifesciences, Irvine, CA, USA) in 7, the St. Jude Medical Epic Stented Tissue valve (Epic; St. Jude Medical, Little Canada, MN, USA) in 1, and the Medtronic Mosaic valve (Mosaic; Medtronic, Dublin, Ireland) in 1. The St. Jude medical bileaflet mechanical valves (SJM) were used for 5 patients with BAV. In the TAV group, there were 11 bioprosthetic valves (7 CEP, 2 Epic, and 2 Mosaic) and 3 bileaflet mechanical valves (SJM).
The following parameters to calculate the progression rate of the proximal aortic diameter were compared: longest diameter (mm), shortest diameter (mm), transvalvular pressure gradient (mmHg) of the aortic valve, area of the aortic valve, degree of aortic regurgitation, left ventricular ejection fraction, body surface area, and cardiovascular risk factors including age, sex, smoking history, hypertension, diabetes mellitus, and dyslipidemia, defined as serum total cholesterol >219 mg/dL.
This retrospective clinical study received Institutional Review Board approval at Hirosaki University, Graduate School of Medicine.
Imaging Protocols and 3-D ReconstructionThe thoracic section of the aorta was reconstructed from axial CT data converted into Digital Imaging and Communications in Medicine (DICOM) files. All 28 patients were assessed before and long after operation, at follow-up (BAV, 2.5±1.9 years after AVR; TAV, 2.2±1.8 years after AVR) on whole body spiral CT (model TSX-310 B; Toshiba Medical Systems, Tochigi, Japan) according to standard procedures in the present institution or its affiliated hospitals. All CT slices measured 512×512 pixels, and the mean pixel size was 0.68±0.035 mm. Iopamiron contrast material (Bayer Healthcare, Leverkusen, Germany), 60–100 mL, was injected at a rate of 2.0–3.5 mL/s. Images were acquired during a single sustained breath-hold by the patient to reduce respiratory-induced motion and associated artifacts. The slice thickness was between 1 and 5 mm. Images >1 mm thick were further sliced to 1 mm using the image processing tools provided with Mimics.
The CT DICOM data were imported into Mimics v16 to reconstruct patient-specific 3-D models of the thoracic aorta using the algorithm introduced by Doyle et al.10 Three steps were required to reconstruct 3-D models from CT data using Mimics. First, a thresholding technique was applied to the imported CT data to highlight the areas of interest, which is called rudimentary segmentation. Thresholding is one of the most objective approaches to highlight the region to be converted to 3-D: specifying the threshold of the gray value of the target region, the part within the threshold range is highlighted. Second, the 3-D models of the highlighted region were calculated automatically. Finally, the new 3-D masks were manually edited using the tools provided with Mimics to smooth the surface and remove any non-physiological bulges, and then the 3-D reconstruction was complete (Figure 1).
(A) Region of interest and (B) 3-dimensional (3-D) reconstruction of regions. DICOM computed tomography (CT) data were imported into Mimics. Regions of interest on CT were selected and reconstructed in 3-D.
The 3-D centerlines and aortic diameters were measured on the 3-D reconstructed models in Mimics. The 3-D centerlines, equivalent to the central axis of the aorta, were automatically calculated. Aortic diameters included the longest and shortest diameters (mm) of the maximum perpendicular cross-section tangential to the centerline. The longest diameter was equal to that passing through the most distal point of the greater curvature. The shortest diameter was perpendicular to the longest. The center of a cross-section was defined as the point at which it passes through the 3-D centerline (Figure 2).
Definitions of the longest and shortest diameters. These diameters were measured on a perpendicular cross-section tangential to the 3-dimensional (3-D) centerline of the aorta. The longest diameter is defined as that passing through the most distal point of the greater curvature. The shortest diameter is perpendicular to the longest. The center of the cross-section is defined as the point at which it passes through the 3-D centerline.
All data were analyzed statistically using SPSS version 20 (SPSS, Chicago, IL, USA). The results are expressed as n (%) or as mean±SD. Categorical data between 2 groups were compared using Person’s chi-squared test. Continuous numeric variables including age, pressure gradient, aortic valve area, ejection fraction, aortic diameter, and progression rate of the aorta were compared using Mann-Whitney test. P<0.05 was considered significant.
Table 1 lists preoperative patient characteristics. There were no significant differences between the BAV and TAV groups. In both groups, the postoperative morbidity and severity of hypertension were similar to preoperative data.
| BAV (n=14) |
TAV (n=14) |
P value | |
|---|---|---|---|
| Age (years) | 63.4±10.1 | 69.9±7.1 | 0.612 |
| Men/Women | 10/4 | 8/6 | 0.347 |
| BSA (m2) | 1.64±0.14 | 1.61±0.17 | 0.520 |
| Smoking | 6 (42.9) | 1 (7.1) | 0.077 |
| Hypertension | 8 (57.1) | 10 (71.4) | 0.695 |
| Diabetes | 0 (0) | 3 (21.4) | 0.222 |
| Dyslipidemia | 2 (14.3) | 5 (35.7) | 0.385 |
| Max PG (mmHg) | 89.2±34.9 | 69.7±26.1 | 0.150 |
| AVA (cm2) | 0.77±0.39 | 0.91±0.37 | 0.277 |
| LVEF (%) | 60.3±10.1 | 66.4±11.2 | 0.316 |
| Concomitant AR (moderate-severe) | 7 (53.8) | 6 (46.2) | 1.000 |
Data given as mean±SD or n (%). AR, aortic regurgitation; AVA, aortic valve area; BAV, bicuspid aortic valve; BSA, body surface area; LVEF, left ventricular ejection fraction; PG, transvalvular pressure gradient; TAV, tricuspid aortic valve.
The time interval between preoperative and follow-up CT assessment ranged from 1 to 8 years in the BAV group and from 1 to 6 years in the TAV group. The average follow-up period was 2.5±1.9 years in the BAV group and 2.2±1.8 years in the TAV group. Tables 2 and 3 lists preoperative and follow-up proximal aortic diameters (mm) measured on the 3-D reconstructed Mimics model. Although the longest and shortest diameters measured preoperatively tended to be larger in the BAV group, the difference between the 2 groups was not significant. Ascending aortic diameter was significantly larger in the BAV group than in the TAV group on follow-up (longest diameter, 39.1±2.73 vs. 35.0±2.63 mm, P=0.002; shortest diameter, 36.8±3.46 vs. 34.2±2.89 mm, P=0.039). Table 4 lists the annual progression rate of the proximal aortic diameter measured on the reconstructed model. The longest diameter of both the ascending aorta and the proximal arch progressed much more rapidly in the BAV group than the TAV group (ascending aorta, 1.02±1.03 vs. −0.075±0.78 mm/year, P<0.001; proximal aortic arch, 0.63±0.65 vs. 0.035±0.062 mm/year, P<0.001). The progression rate of the shortest diameter was significantly larger in the BAV group only in the ascending aorta (0.41±0.62 vs. −0.016±0.59 mm/year, P=0.003). This shows that the ascending aortic dilatation with BAV progressed even after AVR, and that the expansion toward the convexity of the aorta frequently progressed to the proximal aortic arch.
| BAV (n=14) (follow-up, 2.5±1.9 years) |
TAV (n=14) (follow-up, 2.2±1.8 years) |
P value | |
|---|---|---|---|
| Longest diameter (mm) | |||
| Preoperative | 37.4±3.17 | 35.3±2.38 | 0.050 |
| Follow-up | 39.1±2.73 | 35.0±2.63 | 0.002* |
| Shortest diameter (mm) | |||
| Preoperative | 36.0±3.98 | 34.3±2.82 | 0.089 |
| Follow-up | 36.8±3.46 | 34.2±2.89 | 0.039* |
Data given as mean±SD. *P<0.05. 3-D, 3-dimensional; BAV, bicuspid aortic valve; TAV, tricuspid aortic valve.
| BAV (n=14) (follow-up, 2.5±1.9 years) |
TAV (n=14) (follow-up, 2.2±1.8 years) |
P value | |
|---|---|---|---|
| Longest diameter (mm) | |||
| Preoperative | 32.7±2.13 | 33.2±2.71 | 0.758 |
| Follow-up | 33.8±2.26 | 33.3±2.74 | 0.520 |
| Shortest diameter (mm) | |||
| Preoperative | 31.4±3.04 | 32.1±1.75 | 0.382 |
| Follow-up | 32.3±2.30 | 32.1±1.77 | 0.786 |
Data given as mean±SD. Abbreviations as in Table 2.
| BAV (n=14) (follow-up, 2.5±1.9 years) |
TAV (n=14) (follow-up, 2.2±1.8 years) |
P value | |
|---|---|---|---|
| Longest diameter (mm/year) | |||
| Ascending aorta | 1.02±1.03 | −0.075±0.78 | <0.001* |
| Proximal transverse arch | 0.63±0.65 | 0.035±0.062 | <0.001* |
| Shortest diameter (mm/year) | |||
| Ascending aorta | 0.41±0.62 | −0.016±0.59 | 0.003 |
| Proximal transverse arch | 0.22±0.45 | 0.019±0.41 | 0.087 |
Data given as mean±SD. *P<0.05. Abbreviations as in Table 2.
In the TAV group, the progression rate of the longest diameter was not different from that calculated using conventional axial CT (Figure 3). In contrast, in the BAV group, the progression rate of the longest diameter in the proximal aorta was significantly greater than that calculated in the conventional way (ascending aorta, 1.02±1.03 vs. 0.36±0.67 mm/year, P=0.002; proximal aortic arch, 0.63±0.65 vs. 0.14±0.15 mm/year, P=0.002; Table 5).
Diameter measurement using (A) Mimics 3-D reconstruction and (B) conventional axial CT. (A) Longest diameter of the maximum cross-section perpendicular to the 3-D centerline is measured to calculate the progression rate. (B) The diameter to be measured is shorter because the aortic cross-section on axial CT is not perpendicular to the central axis of the aorta. Abbreviations as in Figure 1.
| Mimics longest diameter (mm/year) |
CT diameter (mm/year) |
P value | |
|---|---|---|---|
| BAV | |||
| Ascending aorta | 1.02±1.03 | 0.36±0.67 | 0.002* |
| Proximal transverse arch | 0.63±0.65 | 0.14±0.15 | 0.002* |
| TAV | |||
| Ascending aorta | −0.075±0.78 | −0.16±0.43 | 0.701 |
| Proximal transverse arch | 0.035±0.062 | 0.018±0.093 | 0.242 |
Data given as mean±SD. *P<0.05. CT, computed tomography. Other abbreviations as in Table 2.
This study has shown that ascending aortic dilatation associated with stenotic BAV progresses even after AVR at a maximum rate of 1.02±1.03 mm/year in the right-anterolateral direction, and that the expansion frequently progresses to the proximal transverse arch. In the TAV group, in contrast, ascending aortic dilatation rarely progresses after AVR. This suggests that AVR could not prevent progressive dilatation of the ascending aorta with stenotic BAV, whereas for aortic dilatation associated with stenotic TAV, post-stenotic dilatation could be treated with isolated AVR. Moreover, in BAV patients, the progression rate of the longest diameter calculated using Mimics was significantly greater than that of the conventionally measured diameter. Hence, conventional CT is not always sufficient to monitor the progression of BAV aortopathy.
The question of whether the ascending aortic dilatation associated with stenotic BAV progresses even after AVR is controversial. Yasuda et al reported that the ascending aortic diameter in BAV patients increased at an annual rate of 0.18±0.08 mm/m2 after AVR.9 They showed that AVR might not be the best way to prevent progressive enlargement of the BAV aorta. In contrast, Girdauskas et al reported that patients with stenotic BAV and concomitant mild-moderate ascending aortic dilatation (40–50 mm) are at a similarly low risk of adverse proximal aortic events such as type A aortic dissection and aortic aneurysm progression after isolated AVR, as patients with stenotic TAV.11 They reported that freedom from proximal aortic surgery at 15 years after AVR was 94±3% in the BAV group vs. 89±5% in the TAV group (P=0.2).11
It is difficult to measure the longest diameter of the aorta and calculate the accurate progression rate of aortic dilatation in the conventional way, because the aortic cross-sections on axial CT are not perpendicular to the central axis of the aorta. In the present study, aortic diameter was measured on perpendicular cross-sections tangential to the central axis of the aorta, using Mimics 3-D reconstruction. This technique enabled calculation of the progression rate of the longest diameter of the stenotic BAV aorta, which represents the degree of aortic enlargement toward the convexity of the aorta, where WSS should be greater.
Asymmetric Dilatation of Ascending AortaAscending aortic dilatation associated with stenotic BAV frequently has an asymmetric configuration.6 The 3-D configuration of the aorta is difficult to evaluate conventionally, because of the 3-D complexity of the aortic configuration. Given that the aortic cross-sections on conventional axial CT are not perpendicular to the central axis of the aorta, axial CT measurements do not always accurately represent aorta status. In addition, results include errors due to analog data manipulations. We have previously described how to evaluate aortic configuration using Mimics 3-D reconstruction, which allows automated calculation of the 3-D centerline of the aorta.12 Aorta size was determined by calculating the cross-sectional area of the perpendicular section tangential to the centerline, and symmetry was evaluated by calculating the ellipticity of the maximum perpendicular cross-section. The previous study showed that ascending aortic dilatation with stenotic BAV is more asymmetric and more frequently progresses to the proximal transverse arch compared with that in TAV.12 Doyle et al also evaluated the symmetry of abdominal aortic aneurysms using the 3-D reconstruction tool. They evaluated abdominal aorta symmetry by measuring the perpendicular distance from the proximal and distal points of the centerline to a defined point on the centerline.13
Etiology of Asymmetric Aortopathy Associated With BAVThe asymmetric nature of aortopathy has been confirmed in some molecular biological studies in which asymmetric spatiotemporal extracellular matrix (ECM) protein expression and vascular smooth muscle cell (VSMC) apoptosis were noted in aortas with BAV.14,15 Of the ECM proteins, fibronectin and tenascin were increased at the convexity of the ascending aorta, where WSS should be greater. Type I and II collagens were also more obviously decreased at that location.14,15 Cotrufo et al speculated that asymmetric expression originates in valve-related blood flow abnormalities because the blood flow downstream from BAV is eccentric.14
Four-dimensional flow magnetic resonance imaging is an innovative tool that allows quantitative assessment of blood flow and aortic WSS. The most prevalent flow abnormality associated with BAV is clockwise-nested helical flow accompanied by peripheral skewing of the jet towards the right-anterolateral aspect.7 It has been suggested that the flow abnormality causes asymmetric WSS at the convexity of the proximal aorta.7,8 The flow abnormalities are most remarkable in the ascending aorta, with a large proportion normalizing in the proximal descending aorta.16 Variations in the spatiotemporal distribution of blood flow resulting from differences in valve morphology might lead to asymmetric aortopathy.
Study LimitationsThe present pilot study had several limitations. First, the small number of patients and the large SD were partly due to the strict patient exclusion criteria. For example, some patients with suboptimal CT were excluded because the quality of the reconstructed 3-D models depends on the underlying CT data. The suboptimal CT data included non-contrast and >5-mm-sliced data, with which 3-D models faithful to the real aorta could not be reconstructed. We believe that the strict inclusion criteria make the present findings more meaningful, because reliable reconstruction allows minimization of measurement errors. Next, the study sample was limited to patients with AS who presented with mild ascending aortic dilatation, and only progression rate of mildly dilated proximal aorta associated with stenotic BAV or stenotic TAV after AVR was analyzed.
This is the first description of the progression rate of the aortic diameter after AVR calculated using 3-D reconstruction software. Ascending aortic dilatation associated with stenotic BAV progresses even after AVR at a maximum rate of 1.02±1.03 mm/year in the right-anterolateral direction. The expansion toward the convexity of the aorta frequently progresses to the proximal aortic arch. Given that the progressive enlargement of the proximal aorta toward the right-anterior quadrant may be difficult to evaluate on conventional CT, we should remember that asymmetric dilated BAV aorta sometimes dilates more rapidly than expected.
This study was supported by Grants-in-Aid for Scientific Research in Japan. (No 24592044: “Theoretical analysis of intra-arterial flow using mathematical biology in perfusion from peripheral arteries.”)
The authors declare no conflict of interest.
