Evaluation of bicuspid aortic valve-associated aortopathic changes using third-generation dual-source computed tomography

Abstract

BACKGROUND

Bicuspid aortic valve (BAV) affects 0.5%-2% of the population and is associated with progressive aortopathy leading to life-threatening complications. Aortic size index (ASI) adjusts aortic dimensions for body surface area (BSA), potentially improving risk stratification.

AIM

To characterise BAV-associated aortopathic changes using third-generation dual-source computed tomography (CT) and to derive a cohort-specific ASI threshold associated with CT-defined ascending aortic dilatation in an Indian population.

METHODS

This prospective cross-sectional study enrolled 100 BAV patients (age > 18 years) from July 2022 to November 2023. All patients underwent CT aortography using 192-slice third-generation dual-source CT scanner (Somatom Force, Siemens). Aortic measurements were obtained at multiple levels including annulus, sinus of Valsalva, ascending aorta, arch, and descending thoracic aorta. ASI was calculated as maximum ascending aortic diameter divided by BSA. Statistical analysis included receiver operating characteristic curve analysis and correlation coefficients.

RESULTS

Mean age was 48.95 ± 13.78 years, with 64% males. Ascending aortic dilatation was present in 87% of patients, with a mean diameter of 42.79 ± 8.69 mm. Mean ASI was 25.13 ± 5.71 mm/m² with 42% having ASI > 25 mm/m² (high-risk category). ASI strongly correlated with ascending aortic diameter (rho = 0.87, P < 0.001). At ASI cutoff ≥ 23.8 mm/m² sensitivity was 84% and specificity 90% for predicting aortic dilatation > 40 mm. The mean measurement difference between CT and echocardiography was 11.72 ± 8.27 mm.

CONCLUSION

In this single-centre cohort of Indian patients with BAVs, an ASI threshold of 23.8 mm/m² was associated with CT-defined ascending aortic dilatation and may serve as a cohort-specific reference for risk stratification. Because ASI is mathematically derived from ascending aortic diameter, this finding should be interpreted as a cohort-specific threshold rather than independent validation of ASI as a predictor. Third-generation dual-source CT yielded larger absolute aortic measurements than transthoracic echocardiography; this difference is likely partly methodological, reflecting non-equivalent measurement sites, measurement conventions, and inherent limitations of 2D echocardiography in BAV.

Key Words: Bicuspid aortic valve; Aortic size index; Dual-source computed tomography; Aortopathy; Ascending aortic dilatation; Coarctation; Dissection

Core Tip: Bicuspid aortic valve (BAV) is a common congenital cardiac anomaly, frequently associated with aortopathy. Using third-generation dual-source computed tomography (CT) in 100 patients, we observed ascending aortic dilatation in 87% and identified a cohort-specific aortic size index threshold of 23.8 mm/m² associated with CT-defined ascending aortic dilatation > 40 mm (sensitivity 84%, specificity 90%). Because aortic size index (ASI) is mathematically derived from the ascending aortic diameter, it is best interpreted as a cohort-specific threshold rather than as independent validation of ASI as a predictor. CT measurements were larger than transthoracic echocardiography (mean difference 11.72 mm), although this difference likely reflects differences in measurement sites, measurement conventions, and inherent limitations of 2D echocardiography in BAV rather than simple underestimation alone. These preliminary findings support further evaluation of indexed measurements alongside absolute diameter thresholds in populations with smaller body habitus, and will require validation in independent multicentre cohorts with longitudinal outcomes.

INTRODUCTION

Bicuspid aortic valve (BAV) represents one of the common congenital cardiac anomaly, affecting 0.77% of the general population according to the Copenhagen baby heart study, which screened 25556 newborns and established definitive prevalence data[1]. This condition demonstrates a male predominance with a 2.1:1 ratio and carries significant clinical implications, as patients face a lifetime morbidity burden of 86%, with 68.5% requiring aortic valve surgery during their lifetime[2]. The most concerning complication of BAV is progressive aortopathy, which manifests as ascending aortic dilatation in 33.2% of affected newborns, suggesting a developmental origin rather than purely hemodynamic causation[1].

The pathophysiology of BAV-associated aortopathy involves complex interactions between genetic predisposition and hemodynamic factors. Recent mechanistic studies have demonstrated that wall shear stress serves as the primary predictor of aortic expansion, with BAV-right-noncoronary morphotypes showing increased circumferential wall shear stress in the mid- and distal ascending aorta compared to BAV-right-left phenotypes[3]. This altered hemodynamic environment activates endothelial mechanotransduction pathways, leading to significantly increased expression of MMP-2 and MMP-9 with altered tissue inhibitors of metalloproteinases ratios, promoting progressive elastic fibre degradation[4]. The genetic component is substantial, with 7.3% of first-degree relatives harbouring BAV and 23.6% per-family prevalence, supporting current guidelines for systematic family screening[5].

Traditional risk stratification based on absolute aortic diameter may be inadequate for smaller patients, who can carry meaningful rupture risk without reaching conventional surgical thresholds. Studies from experienced aortic centres have validated that many at-risk aneurysms occur below the traditional 5.5 cm diameter threshold, highlighting the critical limitations of diameter-based assessment[6]. This recognition has driven the development of indexed measurements, with aortic size index (ASI) emerging as a superior predictor of complications. ASI ≤ 2.05 cm/m² carries a 4% average yearly complication risk, while ASI ≥ 4 cm/m² demonstrates an 18% yearly risk, providing more precise risk stratification than absolute measurements[7].

The imaging landscape for BAV assessment has undergone revolutionary changes with the advent of third-generation dual-source computed tomography (CT). Traditional echocardiography demonstrates significant limitations, with meta-analysis revealing a pooled sensitivity of 87.7% and specificity of 88.3% under ideal conditions, with accuracy significantly reduced in the presence of severe valvular calcification[8]. Modern photon-counting CT technology achieves up to 67% radiation dose reduction while maintaining superior image quality, with improved spatial resolution (0.15-0.176 mm pixel pitch) enabling precise anatomical assessment[9]. These advances have prompted the 2022 ACC/AHA Guidelines to recommend multimodality imaging for baseline assessment when echocardiography proves inadequate[10].

Despite these advances, significant gaps remain in BAV management. A concerning surveillance variability study revealed that 27.3% of BAV patients never see a cardiovascular specialist after diagnosis, and one-third do not receive appropriate follow-up imaging[11]. This study addresses these gaps by characterising BAV-associated aortopathic changes using third-generation dual-source CT and deriving a cohort-specific ASI threshold associated with CT-defined ascending aortic dilatation in an Indian BAV population.

MATERIALS AND METHODS

Study design and population

This prospective cross-sectional study was conducted at a federally funded, non-profit tertiary care teaching institution from July 2022 to November 2023. The study was reviewed and approved by the Institutional Ethics Committee of the Postgraduate Institute of Medical Education and Research (Approval No. IEC-INT/2022/MD-638). Written informed consent was obtained from all participants. The sample size of 100 patients was calculated to provide adequate statistical power for the primary analyses: 80% power to detect a correlation coefficient of r = 0.3 between ASI and aortic dimensions (two-tailed, α = 0.05), requiring a minimum of 85 patients, and sufficient events (aortic dilatation > 40 mm, observed in 61 patients) for stable receiver operating characteristic (ROC) curve analysis, exceeding the recommended minimum of 10 events per predictor variable. Inclusion criteria comprised patients aged 18 years and above with echocardiographically or CT-confirmed BAV referred for CT aortography. Exclusion criteria included pregnancy, contraindications to iodinated contrast agents including altered renal parameters or chronic kidney disease, history of anaphylaxis to contrast agents, and unwillingness to provide informed consent.

Patient recruitment and demographics

Patients were consecutively recruited and demographic data including age, gender, presenting symptoms, and past medical history were systematically recorded using a standardized proforma. Anthropometric measurements including height and weight were obtained at the time of CT acquisition using calibrated instruments. Body surface area (BSA) was calculated using the Dubois formula: BSA = 0.007184 × (height in cm)^0.725 × (weight in kg)^0.425. Clinical history documentation included assessment of hypertension, diabetes mellitus, coronary artery disease, and previous cardiac interventions. All data were entered into a secure database for subsequent analysis.

CT imaging protocol

All examinations were performed on a third-generation 192-slice dual-source CT scanner (Somatom Force, Siemens Healthineers, Erlangen, Germany) under supervision of experienced cardiac radiologist. Non-ionic iodinated contrast agent was administered through an 18-gauge intravenous cannula (in antecubital vein) at a flow rate of 4.5-5 mL/second using a dual-head power injector, followed by a 40 mL saline bolus. Prospective high-pitch electrocardiogram (ECG)-gated acquisition was done with images generated in the best diastolic phase. The scan coverage extended from thoracic inlet to just below the diaphragm. Bolus tracking technique was utilized with region of interest placed in the descending thoracic aorta, and image acquisition was triggered when attenuation reached 100 Hounsfield units. A second non-gated high pitch acquisition was performed of entire thoracoabdominal aorta.

Image analysis

Post-acquisition images were transferred to a server-based workstation (syngo.via-advanced multimodality workstation VB30) for comprehensive analysis. An experienced cardiac radiologist performed image evaluation and post-processing, including generation of thin multiplanar reconstructions (MPR), maximum intensity projections, and volume-rendered images. Measurements were obtained at standardized anatomical locations: Aortic annulus, sinus of Valsalva (cusp to commissure), ascending aorta at maximum diameter typically at the level of right pulmonary artery, mid-aortic arch, and descending thoracic aorta at proximal (2 cm distal to left subclavian artery origin) and hiatal levels. BAV morphology was classified according to the Sievers classification system, distinguishing between type 0 (no raphe) and type 1 (one raphe; Figure 1). Additional assessments included evaluation of aortic wall calcification, wall thickening exceeding 4 mm, valvular calcification, presence of dissection, and coarctation. ASI was calculated by dividing the maximum ascending aortic diameter by BSA.

Figure 1 True axial view obtained by double oblique technique demonstrating type 0 and type 1 bicuspid aortic valve. A: Type 0; B: Type 1. Images acquired during prospective electrocardiography-gated high-pitch acquisition in the best diastolic phase.

Echocardiographic correlation

Transthoracic echocardiography (TTE) was performed using standard protocols by experienced echocardiographers who were blinded to CT findings. The echocardiographic examination was conducted within a median interval of 7 days from CT acquisition. Parameters assessed included ascending aortic diameter measured at end-diastole in parasternal long-axis view using leading-edge to leading-edge technique, assessment of aortic regurgitation and stenosis severity, and evaluation of left ventricular function. These measurements were subsequently compared with CT-derived measurements to assess inter-modality agreement and identify potential discrepancies in aortic measurements.

Statistical analysis

Statistical analysis was performed using SPSS version 22 (IBM Corp., Armonk, NY, United States). Continuous variables were expressed as mean ± SD or median with interquartile range based on distribution normality assessed by Shapiro-Wilk test. Categorical variables were presented as n (%). Spearman’s correlation coefficient was used to evaluate relationships between ASI and aortic dimensions. ROC curve analysis determined optimal ASI cutoff for predicting aortic dilatation > 40 mm. Agreement between CT and echocardiography was assessed using Bland-Altman plots and weighted Kappa coefficients. Diagnostic performance parameters including sensitivity, specificity, positive and negative predictive values were calculated. Statistical significance was defined as P value < 0.05.

RESULTS

Patient characteristics

A total of 110 patients with BAV were initially screened, of whom 10 were excluded (3 aged below 18 years, 4 declined to provide informed consent, 3 had poor-quality CT acquisitions), resulting in 100 patients for final analysis (Table 1). The mean age was 48.95 ± 13.78 (range: 19-73) years, with peak prevalence in the fifth and sixth decades of life. Males comprised 64% of the study population, demonstrating a male predominance consistent with BAV epidemiology. The mean BSA was 1.72 ± 0.15 m[2], with mean height of 164.08 ± 9.68 cm and weight of 62.97 ± 10.51 kg. The 21% of patients had significant comorbidities, including hypertension (most common), type 2 diabetes mellitus, and coronary artery disease. Nine percent had undergone previous cardiac interventions, while the majority (79%) presented with isolated BAV without significant cardiovascular risk factors.

Table 1 Baseline patient characteristics and bicuspid aortic valve morphology (n = 100), mean ± SD/n (%).

Characteristic	Value
Demographics
Age (year)	48.95 ± 13.78
Male gender	64 (64.0)
Anthropometric measurements
Height (cm)	164.08 ± 9.68
Weight (kg)	62.97 ± 10.51
BMI (kg/m2)	23.45 ± 3.73
BSA (m2)	1.72 ± 0.15
Clinical characteristics
Significant comorbidities1	21 (21.0)
Previous cardiac intervention	9 (9.0)
BAV morphology (Sievers classification)
Type 0 (no raphe)	63 (63.0)
Type 1 (one raphe)	37 (37.0)
Type 2 (two raphes)	0 (0.0)

¹Includes hypertension, type 2 diabetes mellitus, and coronary artery disease.

Body surface areas were calculated using the Dubois formula: Body surface area (m²) = 0.007184 × height (cm)^0.725 × weight (kg)^0.425. BMI: Body mass index; BSA: Body surface area; BAV: Bicuspid aortic valve.

BAV morphology and classification

Morphological assessment using Sievers classification revealed type 0 (pure bicuspid valve without raphe) in 63 patients (63%) and type 1 (presence of one raphe) in 37 patients (37%). No type 2 morphology was observed in our cohort. Among patients with ascending aortic dilatation exceeding 40 mm, the prevalence was similar between type 0 (60.3%) and type 1 (62.2%) morphology (χ² = 0.038, P = 0.845), indicating no significant association between raphe presence and aortic dilatation in our cohort. The distribution pattern aligned with contemporary CT-based studies, confirming type 0 as the predominant morphology in the preoperative BAV population.

Aortic dimensions and ASI

Comprehensive aortic measurements revealed progressive dilatation predominantly affecting the ascending aorta (Table 2). The mean ascending aortic diameter was 42.79 ± 8.69 (range: 24-76) mm, with 87% of patients demonstrating dilatation above age- and sex-specific normal limits (> 37 mm for males, > 34 mm for females). Using the clinically significant threshold of > 40 mm widely adopted in guidelines as a trigger for intensified surveillance in BAV, 61% of patients demonstrated significant aortic dilatation (Table 2). This distinction is clinically important, as patients in the 37-40 mm range represent a surveillance zone requiring serial follow-up. Distribution analysis showed 13% with diameter < 35 mm, 22% between 35-39 mm, 32% between 40-44 mm, and 33% ≥ 45 mm. The mean aortic annulus diameter was 25.43 ± 5.02 mm, while sinus of Valsalva measured 34.55 ± 6.74 mm. The calculated mean ASI was 25.13 ± 5.71 mm/m² with distinct risk stratification categories: 12% had ASI < 20 mm/m² (low risk), 46% between 20-25 mm/m² (moderate risk), and 42% > 25 mm/m² (high risk). Correlation analysis demonstrated strong positive correlation between ASI and ascending aortic diameter (rho = 0.87, P < 0.001), moderate correlation with sinus of Valsalva (rho = 0.48, P < 0.001) and arch diameter (rho = 0.45, P < 0.001), and weak correlation with annulus diameter (rho = 0.22, P = 0.031).

Table 2 Aortic dimensions, aortic size index distribution, and aortopathic changes (n = 100), mean ± SD/n (%).

Parameter	Value
Aortic dimensions (mm)
Aortic annulus	25.43 ± 5.02
Sinus of Valsalva	34.55 ± 6.74
Ascending aorta	42.79 ± 8.69
Aortic arch	26.19 ± 4.77
Proximal descending thoracic aorta	22.81 ± 3.78
Descending thoracic aorta at hiatus	18.94 ± 2.80
Ascending aorta diameter categories (mm)
< 35	13 (13.0)
35-39	22 (22.0)
40-44	32 (32.0)
≥ 45	33 (33.0)
ASI (mm/m2)	25.13 ± 5.71
< 20 (low risk)	12 (12.0)
20-25 (moderate risk)	46 (46.0)
> 25 (high risk)	42 (42.0)
Aortopathic changes
Aortic dilatation > 40 mm	61 (61.0)
Aortic valve calcification	73 (73.0)
Aortic wall calcification	21 (21.0)
Aortic wall thickening > 4 mm	5 (5.0)
Aortic dissection (Stanford type A)	1 (1.0)
Aortic coarctation/narrowing	2 (2.0)

ASI: Aortic size index.

Aortopathic changes

Evaluation of aortopathy-related complications revealed high prevalence of structural changes. Aortic dilatation exceeding 40 mm was present in 61% of patients, representing the most common manifestation. Aortic valve calcification was observed in 73% of cases, indicating significant degenerative changes despite the relatively young mean age. Aortic wall calcification was documented in 21% of patients, while wall thickening exceeding 4 mm occurred in 5%, with identified causes including Takayasu arteritis (n = 1), atherosclerosis (n = 2), and idiopathic (n = 2). One patient (1%) presented with acute Stanford type A aortic dissection associated with severe ascending aortic dilatation of 76 mm. This patient presented with acute-onset chest pain and was referred emergently for CT aortography, which confirmed the dissection. The patient had a previously undiagnosed BAV (type 0) with no prior aortic imaging surveillance, illustrating the potential consequences of inadequate follow-up in BAV patients (Figure 2). Aortic coarctation or narrowing was identified in 2% of cases, including one post-ductal coarctation and one interrupted aortic arch, representing rare but important associations with BAV (Figure 3).

Figure 2 Bicuspid aortic valve with type A aortic dissection. A: True axial view demonstrating type 0 bicuspid aortic valve; B and C: Coronal (B) and volume rendered image (C) depicting dilated ascending aorta with type A aortic dissection (shown by blue arrows). Images acquired during prospective electrocardiography-gated high-pitch acquisition in the best diastolic phase.

Figure 3 Bicuspid aortic valve with post ductal coarctation of aorta. A: True axial view demonstrating type 0 bicuspid aortic valve; B and C: Sagittal maximum intensity projection (B) and volume rendered image (C) depicting dilated ascending aorta with post ductal coarctation (shown by arrows). Images acquired during prospective electrocardiography-gated high-pitch acquisition in the best diastolic phase.

Diagnostic performance analysis

ROC curve analysis established optimal diagnostic thresholds for predicting significant aortic dilatation > 40 mm (Table 3). The ROC analysis identified an ASI cutoff of ≥ 23.8 mm/m² associated with CT-defined ascending aortic dilatation > 40 mm in this cohort, with an area under the curve of 0.935 (95%CI: 0.891-0.979), sensitivity 84%, specificity 90%, positive predictive value 92.7%, and negative predictive value 77.8% (Figure 4).

Figure 4 Receiver operating characteristic curve analysis showing diagnostic performance of aortic size index in predicting aortic dilatation > 40 mm. ROC: Receiver operating characteristic.

Table 3 Receiver operating characteristic characteristics of aortic size index and comparison with imaging modalities.

Parameter	Value (95%CI)
ROC analysis for predicting aortic dilatation > 40 mm
ASI (cutoff ≥ 23.8 mm/m2)
AUROC	0.935 (0.891-0.979)
Sensitivity (%)	84 (72-92)
Specificity (%)	90 (76-97)
Positive predictive value (%)	92.7 (82-98)
Negative predictive value (%)	77.8 (63-89)
Diagnostic accuracy (%)	86 (78-92)
CT ascending aorta diameter (cutoff ≥ 40 mm)
AUROC	0.972 (0.942-1.000)
Sensitivity (%)	100 (94-100)
Specificity (%)	90 (76-97)
Diagnostic accuracy (%)	96 (90-99)
Echocardiography diameter (cutoff ≥ 34 mm)
AUROC	0.773 (0.633-0.912)
Sensitivity (%)	68 (48-84)
Specificity (%)	90 (67-99)
Diagnostic accuracy (%)	76.6 (62-88)
CT vs echocardiography agreement
Mean diameter difference (mm)	11.72 ± 8.27
Weighted Kappa coefficient	0.356 (fair agreement)

ROC: Receiver operating characteristic; ASI: Aortic size index; AUROC: Area under receiver operating characteristic curve; CT: Computed tomography.

CT vs echocardiography agreement

Quantitative comparison showed lower mean ascending aortic diameters on TTE than on CT (31.82 ± 9.37 mm vs 42.79 ± 8.69 mm respectively), with a mean intermodality difference of 11.72 ± 8.27 mm (P < 0.001). Bland-Altman analysis demonstrated wide limits of agreement (± 16.21 mm), with 93.6% of observations falling within these limits. The weighted Kappa coefficient of 0.356 indicated only fair agreement between modalities, with 70.2% of cases showing discordant classification when categorized by diameter ranges. This intermodality difference is likely partly attributable to non-equivalent measurement sites (CT measures the maximum ascending aortic diameter on double-oblique reformations whereas TTE measures a more proximal segment in parasternal long-axis view), differing measurement conventions (outer-wall-to-outer-wall on CT vs leading-edge-to-leading-edge on TTE), and inherent limitations of 2D echocardiography in assessing the asymmetric ascending aortic dilatation characteristic of BAV, rather than reflecting simple underestimation alone.

DISCUSSION

In this single-centre cohort, an ASI threshold of ≥ 23.8 mm/m² was associated with CT-defined ascending aortic dilatation > 40 mm in BAV patients (sensitivity 84%, specificity 90%). Because ASI is mathematically derived from the ascending aortic diameter, this finding is best interpreted as a cohort-specific threshold rather than as independent validation of ASI as a predictor. This cohort-specific cutoff falls within the risk-transition zone described by the Yale Aortic Institute’s validation study by Zafar et al[7], which established ASI thresholds showing that values ≤ 20.5 mm/m² carry low risk while those approaching 25 mm/m² indicate substantially elevated risk. Our threshold falls within this critical transition zone, supporting its clinical validity. The International Registry of Aortic Dissection similarly identified ASI > 2.3 cm/m² as predictive of increased complications, with the 2022 ACC/AHA Guidelines now recommending annual surveillance for patients exceeding this threshold[10]. Our slightly higher threshold of 23.8 mm/m² likely reflects population-specific characteristics and the use of advanced CT technology providing more precise measurements.

We acknowledge that since ASI is mathematically derived from the ascending aortic diameter (ASI = maximum ascending aortic diameter/BSA), evaluating its ability to predict aortic dilatation > 40 mm introduces a degree of mathematical circularity, as the predictor is derived from the outcome variable. However, our ROC analysis was designed not to establish ASI as an independent novel predictor, but rather to determine a population-specific ASI threshold that identifies clinically significant aortic dilatation in Indian BAV patients with smaller body habitus (mean BSA 1.72 m²). The clinical rationale for ASI rests on its original validation by Davies et al[12] against hard clinical endpoints (rupture P = 0.0014, dissection, and death) in 805 patients at the Yale Aortic Institute, where ASI proved superior to absolute diameter alone. Patients were stratified into three risk groups: < 2.75 cm/m² (low risk, approximately 4% per year), 2.75-4.24 cm/m² (moderate risk, approximately 8% per year), and ≥ 4.25 cm/m² (high risk, approximately 20% per year). The same absolute diameter confers vastly different risk depending on body size, which is precisely why indexed measurements are recommended by current guidelines[10].

The high prevalence of ascending aortic dilatation (87%) in our cohort, while greater than figures reported in most community-based series, partly reflects tertiary-referral selection bias and the use of sensitive CT-based measurement, and finds context in recent comprehensive studies. The Copenhagen Baby Heart Study documented aortopathy in 33.2% of BAV newborns[1], while long-term follow-up studies report progressive dilatation in 45%-60% of adults[2]. However, when using sensitive CT-based measurements and including mild dilatation (> 37 mm), prevalence rates approach our findings. Multi-centre European registry data report a similar pattern when ascending aortic dilatation is assessed systematically: The polish RE-BAV registry of 814 adult patients across 23 tertiary centres documented aortic dilatation in 63.6% of BAV patients, with extended aortic dilatation in 26.3%[13]. The discrepancy likely reflects our use of third-generation dual-source CT, which detects subtle changes missed by echocardiography, and our comprehensive measurement protocol at multiple aortic levels.

Our finding that 42% of patients had ASI > 25 mm/m² (high-risk category) is consistent with surgical outcome studies. Chemtob et al[14] demonstrated, in 491 patients undergoing prophylactic modified aortic root reimplantation at indexed thresholds (cross-sectional area-to-height ratio > 10 cm²/m, with BSA-indexed annulus sizing)—comparable to our high-risk category, there was 87% survival at 15 years with no operative deaths. The American Association of Thoracic Surgery consensus guidelines support intervention at cross-sectional area to height ratio ≥ 10 cm²/m, which correlates with ASI values in our high-risk range[15]. Together, these data suggest that ASI ≥ 23.8 mm/m² may warrant further evaluation as a marker for intensified surveillance and earlier multidisciplinary review, although its role as a stand-alone clinical decision point requires prospective validation against hard outcomes.

The mean measurement difference between CT and echocardiography was 11.72 ± 8.27 mm between modalities. This finding aligns with Jilaihawi et al’s multicentre TAVR (transcatheter aortic valve replacement) study[16], which reported a mean CT vs echocardiography difference of 8-12 mm, partly attributable to differences in measurement sites and conventions. Such intermodality differences have clinical relevance, as Acharya et al[6] demonstrated that 51.9% of patients with 4.5-5.0 cm diameters already have abnormal indexed aortic areas requiring intervention.

The large measurement difference between CT and echocardiography in our study can be attributed to three key factors. First, different measurement sites: Our CT measurement was obtained at the maximum diameter of the ascending aorta typically at the level of the right pulmonary artery using double-oblique multiplanar reconstruction, while the echocardiographic measurement was obtained in the parasternal long-axis view at a more proximal level approximately 2-3 cm above the aortic annulus. Since BAV-associated aortopathy predominantly affects the mid-to-distal ascending aorta, this spatial mismatch means that echocardiography inherently misses the segment of maximum dilatation. Second, different measurement techniques: CT used outer wall-to-outer wall on double-oblique reformations, while echocardiography used leading-edge-to-leading-edge technique. Park et al[17] in a study of 53 BAV patients (mean age 54 years, 74% men) demonstrated that outer wall-to-outer wall measurement on CT angiography had best agreement with the parasternal long-axis systolic leading-edge method on TTE (Lin concordance 0.93, bias 1.3 mm), while discrepancies increased substantially with non-optimal measurement combinations. Third, inherent limitations of 2D echocardiography in BAV: Asymmetric dilatation patterns make standard echocardiographic assessment challenging, as the maximum diameter may not lie within the standard imaging plane[18]. Beetz et al[19] demonstrated that CT and TTE ascending aorta measurements had wide limits of agreement, with 40% of measurement pairs falling outside clinically acceptable limits. Frazao et al[20] similarly reported significant measurement discrepancies between CT, magnetic resonance imaging (MRI), and echocardiography for thoracic aortic dimensions, reinforcing the need for modality-specific reference ranges.

The strong correlation between ASI and ascending aortic diameter (rho = 0.87) is expected, given that ASI is mathematically derived from the ascending aortic diameter; this correlation confirms the internal consistency of the indexed measurement rather than independently validating ASI as a predictor. Zafar et al[7] similarly found that both ASI and height-indexed measurements correlate strongly with outcomes (r > 0.8), supporting indexed measurements over absolute diameter. Our moderate correlations with sinus of Valsalva (rho = 0.48) and arch diameter (rho = 0.45) reflect the predominant ascending phenotype of BAV aortopathy, consistent with the International BAV Consensus classification identifying ascending phenotype predominance in 70% of cases[21].

The use of third-generation dual-source CT (Somatom Force) in our study provided several specific technical advantages for BAV aortopathy assessment. The temporal resolution of 66 ms with dual-source acquisition enables motion-free imaging of the aortic root and ascending aorta even at elevated heart rates. Apfaltrer et al[22] demonstrated that third-generation dual-source CT achieves a median effective dose of only 1.55 mSv compared to 12.29 mSv with conventional 64-slice CT, corresponding to an 87.4% radiation dose reduction (P < 0.001), while simultaneously achieving significantly higher contrast-to-noise and signal-to-noise ratios (P < 0.001) and better subjective image quality. The 192 mm × 0.6 mm detector configuration provides enhanced Z-axis coverage and spatial resolution, enabling comprehensive assessment of the entire thoracic aorta in a single acquisition with high-resolution MPR[9]. These technical advantages are particularly important in BAV aortopathy, where precise measurement of asymmetric dilatation using double-oblique MPR requires high spatial and temporal resolution.

Regarding the potential confounding effect of comorbidities on aortic dimensions, 21% of our patients had significant comorbidities including hypertension, type 2 diabetes mellitus, and coronary artery disease. However, when patients with hypertension were compared with normotensive BAV patients, mean ascending aortic diameter was not significantly different (44.2 mm vs 42.4 mm, P = 0.48), suggesting that BAV morphology itself was the primary driver of aortopathy in our cohort. This is consistent with Fleury et al[23] who in a prospective study of 310 patients (86 BAV and 224 TAV) from the PROGRESSA cohort demonstrated that BAV patients had a significantly higher aortic dilation rate than TAV patients (0.33 mm/year vs 0.21 mm/year, P = 0.04). Crucially, on multivariable analysis, the absence of hypertension and higher LDL levels were independent predictors of faster ascending aortic dilation in BAV patients, suggesting that BAV-associated aortopathy is predominantly driven by intrinsic aortic wall pathology rather than hemodynamic factors. Detaint et al[24] further reported that prescribed antihypertensive medications showed no independent association with lower aortic dilatation progression rate (all P > 0.13), supporting this conclusion.

Our findings are consistent with the recent guideline trend toward incorporating indexed measurements alongside absolute diameter for risk stratification. The 2020 ACC/AHA Valvular Heart Disease Guidelines and 2022 Aortic Disease Guidelines have incorporated indexed measurements, recognising their superiority for risk stratification[10,25]. In this cohort, the identified ASI threshold of 23.8 mm/m² categorised additional patients as high-risk who would not meet a traditional 40 mm absolute-diameter threshold; however, whether this categorisation translates into improved prevention of serious aortic events such as dissection or rupture can only be established by prospective external validation with longitudinal outcomes.

Study limitations include single-centre design and lack of long-term outcome data. All CT measurements were performed by a single experienced cardiac radiologist, and formal inter-observer and intra-observer reliability metrics (intraclass correlation coefficient) were not obtained. While ECG-gated CT with standardised double-oblique multiplanar reconstruction has demonstrated excellent reproducibility in published literature, with mean intraobserver differences of only -0.3 mm to 0.6 mm and inter-method correlations of 0.81-0.99[26], the absence of formal reproducibility analysis is acknowledged as a limitation. Our study cohort consists of BAV patients referred for CT aortography at a tertiary care centre, which introduces referral bias toward more advanced aortopathy. The high prevalence of ascending aortic dilatation (87%) should not be extrapolated to the broader BAV population, though Detaint et al[24] similarly reported 87% prevalence in their tertiary care BAV cohort of 353 patients. Certain advanced parameters such as 4D-flow-derived wall shear stress patterns and aortic distensibility indices were not assessed, as these require specialised MRI sequences not routinely available. Nevertheless, our findings are broadly consistent with prior international studies and may inform local practice while awaiting external validation is available. Future multicentre validation studies with longitudinal follow-up, formal inter-observer reproducibility assessment, and inclusion of 4D-flow parameters will further refine these thresholds and establish their prognostic value across diverse populations. Taken together, the single-centre tertiary referral design, the high prevalence of ascending aortic dilatation (87%), the absence of formal inter-observer and intra-observer reliability analysis, and the lack of longitudinal outcome data mean that the clinical implications of the identified ASI threshold should be interpreted with caution and considered hypothesis-generating until validated in independent multicentre cohorts with long-term follow-up.

CONCLUSION

In this single-centre cohort of Indian BAV patients, an ASI threshold of ≥ 23.8 mm/m² was associated with CT-defined ascending aortic dilatation > 40 mm. Because ASI is mathematically derived from ascending aortic diameter, this finding represents a cohort-specific threshold associated with CT-defined dilatation rather than independent validation of ASI as a predictor, and prospective external validation with longitudinal outcomes is required before clinical implementation. Third-generation dual-source CT yielded larger absolute aortic measurements than echocardiography; this difference is likely partly methodological, reflecting non-equivalent measurement sites, differing measurement conventions, and inherent limitations of 2D echocardiography in BAV, rather than simple underestimation alone.