Article Text

Aortopathy and bicuspid aortic valve: haemodynamic burden is main contributor to aortic dilatation
  1. Yong-Giun Kim1,
  2. Byung Joo Sun1,
  3. Gyung-Min Park1,
  4. Seungbong Han2,
  5. Dae-Hee Kim1,
  6. Jong-Min Song1,
  7. Duk-Hyun Kang1,
  8. Jae-Kwan Song1
  1. 1Cardiac Imaging Center, Asan Heart Institute, Asan Medical Center, University of Ulsan College of Medicine, Seoul, South Korea
  2. 2Department of Clinical Epidemiology and Biostatistics, Asan Heart Institute Asan Medical Center, University of Ulsan College of Medicine, Seoul, South Korea
  1. Correspondence to Professor Jae-Kwan Song, Cardiac Imaging Center, Asan Heart Institute, Asan Medical Center, University of Ulsan College of Medicine, 388-1 Poongnap-dong Songpa-gu, Seoul 138-736, South Korea; jksong{at}amc.seoul.kr

Abstract

Objective To evaluate factors associated with aortic enlargement in patients with a bicuspid aortic valve (BAV) and the impact of isolated aortic valve replacement (AVR).

Design and setting A retrospective analysis of clinical data in a tertiary referral hospital.

Patients and main outcome measures We performed a cross-sectional analysis of 595 patients with BAV to assess variables determining maximal ascending aortic dimension. To measure annual dilatation rates, baseline and follow-up echocardiograms were analysed in 70 patients with BAV (BAV-AVR group) and 48 with a tricuspid aortic valve (TVA-AVR group) who underwent isolated AVR, and compared with 65 patients with BAV who did not undergo AVR (BAV-NAVR group).

Results Aortic regurgitation (AR) severity was associated with aortic sinus diameter (p<0.001), whereas aortic stenosis severity with the tubular diameter (p<0.001). Multivariate analysis showed that age was an independent factor for both sinus and tubular diameter with AR severity being for aortic sinus diameter and moderate to severe aortic stenosis or AR being for aortic tubular diameter. Despite younger age and lower prevalence of moderate to severe valvular dysfunction at baseline, the annual dilatation rates at sinus and tubular part were significantly higher in the BAV-NAVR than in the BAV-AVR and TAV-AVR groups (p<0.05 each), which did not differ in the BAV-AVR and TAV-AVR groups (p=0.402 for sinus and p=0.394 for tubular part).

Conclusions Age-dependent aortic enlargement associated with significant valvular dysfunction and the protective effects of isolated AVR in patients with BAV indicate that valvular dysfunction is a major determinant to the development of aortopathy.

  • Aorta, Great Vessels and Trauma
  • Congenital Heart Disease
  • Valvular Disease

Statistics from Altmetric.com

Aortopathy in the bicuspid aortic valve (BAV) is a challenging clinical issue. Patients with BAV are at significantly higher risk for aortic dissection and have larger aortic dimensions than patients with a normal trileaflet aortic valve1–3 indicating that BAV is an inherent disorder. This hypothesis was strongly supported by an association between a novel mutation in the NOTCH1 gene and the development of BAV.4 ,5 Moreover, aortic dilatation in BAV was reported to be disproportional to coexisting valvular lesions or to be present in patients without significant valve dysfunction6–9 or after successful aortic valve replacement (AVR).10 All of these findings are thought to indicate that bicuspid aortopathy is a result of a genetic disorder or inherited fragility and has led to recommendations that the proximal aorta in these patients be treated more aggressively.11 Most previous studies, however, have been cross-sectional analyses of relatively small numbers of patients and contradictory results were reported regarding pathogenesis and optimal treatment of bicuspid aortopathy.12 ,13 Longitudinal follow-up studies have found that the course of this condition can range from quite benign14 ,15 to markedly unfavourable,16 with no quantitative measurements of aortic dilatation. In one study assessing aortic dimension after AVR using echocardiography, the numbers of subjects were too small (<20 per group) to determine whether progressive aortic dilatation in BAV is mainly due to the fragility of the aortic wall or due to haemodynamic factors.10 In the present work, we addressed these issues by undertaking a cross-sectional analysis of echocardiographic data in a large number of patients with BAV and by longitudinal follow-up directly comparing the annual dilatation rates in the ascending aorta after isolated AVR.

Methods

Subjects

We retrospectively reviewed the echocardiographic database and digital images of patients with BAV evaluated at our institution (Asan Medical Center) from 1997 through 2007. Bicuspid valve morphology was confirmed by independent review of each echocardiogram by two reviewers (DHK and JKS). We excluded all patients with suboptimal echocardiographic images, documented aortic dissection, heart diseases known to affect proximal aortic size (eg, Marfan syndrome, coarctation of the aorta, tetralogy of Fallot, ventricular septal defect and supra-aortic stenosis), acute aortic vegetations and history of prior open heart surgery. Our cross-sectional study included a total of 595 patients with BAV.

To quantitatively measure the annual dilatation rate of the ascending aorta during follow-up after isolated AVR, we selected patients who underwent follow-up echocardiographic examinations in our institution after uneventful AVR. These patients included 70 patients with BAV (BAV-AVR group) and 48 with a tricuspid aortic valve ((TAV) TAV-AVR group). None of these patients who underwent AVR underwent any surgical procedures for the ascending aorta, such as aortic graft replacement or the Bentall procedure. For comparison, we assessed 65 patients with BAV who did not undergo AVR (NAVR) (BAV-NAVR group). Mean echocardiographic follow-up was 41±25 months. This retrospective study was approved by our institutional review board.

Echocardiographic measurements and clinical variables

Demographic variables included age, gender, weight, height and body surface area, and clinical variables included systolic and diastolic blood pressure at initial echocardiographic examination. Echocardiography was performed using commercially available machines and standard techniques, but suprasternal views to measure arch aneurysm were not routinely performed. We measured left ventricular end systolic and diastolic dimensions, volumes and ejection fraction by Simpson's method. Comprehensive echo-Doppler examination was performed to assess the severity of valvular dysfunction.17 ,18 Based on current guidelines, patients with valvular dysfunction were categorised as having normal (no aortic valve dysfunction), mild, moderate and severe aortic stenosis (AS) and aortic regurgitation (AR) and patients with moderate to severe AS or AR were regarded as having ‘haemodynamically significant’ valvular dysfunction. Patients were also categorised based on the degree and type of valvular dysfunction into a non-significant group (patients with normally functioning BAV, mild AS or mild AR), an AS dominant group (patients with moderate to severe AS), an AR dominant group (patients with moderate to severe AR), and an ASR dominant group (patients with both moderate to severe AS and AR).

The maximal diameter of the ascending aorta was measured at the level of both aortic sinus and tubular portion (ascending aorta distal to the sinotubular junction) in the parasternal long-axis view, which was normalised by body surface area. The difference between aortic diameters before AVR and at last follow-up was calculated, and this difference was divided by duration of follow-up to yield the annual dilatation rate.

Statistical analysis

Data were presented as absolute numbers and percentages or as median with IQR for maximal aortic diameter and annual dilatation rate of aortic diameter and means±SDs for other parameters. For the cross-sectional analysis, simple and multiple linear regression analyses were used to find variables associated with maximal aortic diameter at the level of aortic sinus and tubular portion separately in patients with BAV. Age, gender, weight, height, body surface area, body mass index, systolic and diastolic blood pressure, haemodynamic significance of valvular dysfunction, severity of AS and AR, left ventricular end systolic and diastolic dimensions, and ejection fraction were entered into the model. One-way analysis of variance (ANOVA) with the Bonferroni correction for post hoc comparisons or the non-parametric Kruskal-Wallis test with Mann-Whitney U test for post hoc corrections were used for univariate comparisons of maximal aortic diameter of sinus and tubular portion in patients with BAV. To evaluate whether maximal aortic diameter shows any trend with AS or AR severity, Jonckheere-Terpstra test was done. To quantitatively assess annual aortic dilatation rates, data from the BAV-AVR, TAV-AVR and BAV-NAVR groups were compared, using the χ2 test for categorical variables and one-way ANOVA or the Kruskal-Wallis test with post hoc comparison for continuous variables. Paired t-tests or Wilcoxon signed rank test were used to compare ascending aortic diameters at baseline and follow-up in each of the three groups. All statistical analyses were performed using SPSS V.19.0 software (SPSS, Inc., Chicago, Illinois, USA), with a probability value less than 0.05 considered statistically significant.

Results

Factors associated with maximal aortic diameter in BAV

Table 1 shows the characteristics of the 595 patients enrolled in our cross-sectional study. Mean patient age was 53.4±14.3 years and more than 65% were male. Table 2 shows the widely varying degree of valvular dysfunction in patients with BAV. Fifty-seven patients (9.6%) had normal bicuspid valvular function, with neither AS nor AR. AR was present in 407 patients, ranging from mild in 256 to moderate in 69 to severe in 82. AS was present in 365 patients, including 64 with mild, 63 with moderate and 238 with severe AS. Of the BAV patients with classified valvular disease, 418 had haemodynamically significant valvular dysfunction, including 267 with predominant AS, 117 with predominant AR and 34 with predominant ASR. The remaining 117 patients with either normally functioning BAV or mild AS or mild AR were classified to have haemodynamically non-significant valvular dysfunction.

Table 1

Characteristics of the 595 patients in the cross-sectional study

Table 2

Valvular disease in bicuspid aortic valve patients

According to four different groups of valvular dysfunction, maximal aortic diameter showed significant difference (table 1). Maximal diameter at the level of aortic sinus was significantly larger in patients with predominant AR compared to those with non-significant valvular dysfunction (p=0.001), whereas those with predominant AS or ASR did not show any difference (figure 1). AS was strongly correlated with maximal aortic diameter at the tubular portion and there was no difference between patients with predominant AR and those with non-significant valvular dysfunction (figure 1). Figure 2 shows different patterns of aortic enlargement in patients with BAV according to AS severity. Maximal diameter at the level of aortic sinus was not affected by AS severity (p=0.088), but maximal aortic diameter at tubular portion showed increasing trend with increasing AS severity (p<0.001). Relationship between AR severity and aortic enlargement is demonstrated in figure 2. Although Jonckheere-Terpstra test showed significant difference in maximal aortic sinus diameter in four groups with different AR severity (p<0.001), the only significant difference was demonstrated between patients without AR and those with severe AR. At the tubular level, maximal aortic diameter was not affected by AR severity (p=0.152).

Figure 1

Association between maximal aortic diameters at the level of aortic sinus (upper) and tubular portion (lower) and classification of valvular dysfunction in bicuspid aortic valve patients. Aortic sinus diameter was significantly larger in patients with dominant aortic regurgitation (AR), whereas patients with moderate to severe aortic stenosis (AS) showed larger aortic tubular diameter.

Figure 2

Association between maximal aortic diameters and severity of aortic stenosis (AS) (upper panel) and aortic regurgitation (AR) (lower panel) in bicuspid aortic valve (BAV) patients. Increased severity of AS was strongly associated with increased aortic diameter at the tubular portion, but not with that at sinus portion. Only patients with severe AR showed significantly larger diameter at sinus portion.

Univariate variables associated with maximal aortic diameter at the level of aortic sinus included age, blood pressure, left ventricular (LV) dimension and volume, ejection fraction, LV mass index, AR severity, and classification of valvular dysfunction. Multivariate analysis revealed that age, LV mass index and AR severity were significant variables: aortic diameter=14.473+0.068×age+0.009×LV mass index+0.960×AR severity (r=0.373, p<0.001). In this model, we considered AR severity as a continuous-type variable (no AR=0, mild AR=1, moderate AR=2, severe AR=3).

Univariate variables associated with maximal aortic diameter at tubular portion were similar to those associated with maximal aortic diameter at the level of aortic sinus with male gender, AS severity, and classification and haemodynamic significance of valvular dysfunction being additional variables. Multivariate analysis revealed that age, male gender and haemodynamic significance of valvular dysfunction were independent variables: maximal diameter=17.535+0.128×age—2.809×male gender+1.664×haemodynamic significance (r=0.495, p<0.001).

Annual dilatation rate during follow-up

Clinical characteristics of the patients selected for calculation of annual aortic dilatation rate are summarised in table 3 shown in an online supplemental file. As expected, haemodynamically significant AS and AR were each more prevalent in the BAV-AVR and TAV-AVR groups than in the BAV-NAVR group, the latter of which included seven patients with normal bicuspid valvular function. The BAV-NAVR group was also younger than the other two groups. However, maximal aortic diameters at baseline were comparable.

All three groups showed progressive increases in maximal aortic dimensions during follow-up (figure 3). The annual dilatation rate at the level of both sinus and tubular part differed significantly among these groups (table 3). Increase in tubular diameter was observed in 41 patients (58.6%) of BAV-AVR group and 49 (75.4%) of BAV-NAVR group. Rapid aortic dilatation defined as higher than 75th percentile19 was observed in 17 patients (24.3%) of BAV-AVR group and 16 (24.6%) of BAV-NAVR group. Individual analysis showed that annular dilatation rate at the level of aortic sinus was significantly higher in the BAV-NAVR than in the BAV-AVR (p=0.008) and TAV-AVR (p=0.040) groups, but did not differ in the BAV-AVR and TAV-AVR groups (p=0.402). This trend was similarly observed in the annual dilatation rate of tubular aorta.

Figure 3

Changes in maximal aortic dimensions during follow-up and annual dilatation rate in each group. Numbers in the figure represent median and IQR.

Discussion

In the present study of a large numbers of patients with BAV, we found that aortic enlargement at the level of both aortic sinus and tubular portion is progressive and strongly associated with underlying valvular dysfunction. We also found that the annual dilatation rate was higher in BAV patients before AVR but that this progressive dilatation could be effectively prevented by isolated AVR. Together, these findings indicate that haemodynamically significant valvular dysfunction contributes considerably to development of the aortopathy in patients with BAV.

Inherent fragility versus haemodynamic burden

We found that haemodynamic abnormalities associated with valvular dysfunction are important in determining aortic enlargement. Interestingly, AR and AS showed different impact on aortic dilatation in different levels of the ascending aorta: the diameter of aortic sinus is affected by AR severity, whereas AS severity shows a linear relationship with the diameter of the tubular portion. These findings disagree with the previous observations that aortic dilatation in patients with BAV was out of proportion to coexistent valvular lesions and AR severity, rather than AS severity, was a main determinant of aortic dilatation.7 ,20 The discrepancy between the results of our study and the previous observations may be due primarily to differences in patient populations. The landmark study supporting the genetic aetiology of bicuspid aortopathy included only 118 subjects, a number that may have been too small to be representative of all haemodynamic abnormalities in BAV patients.7 For example, only 18 BAV patients (15%) had normal valvular function and 55 (47%) had normal bicuspid valve function or haemodynamically insignificant mild AS or AR. Fewer than 10 patients each had 13 of the 16 degrees of aortic valvular disease.7 Moreover, the combination of limited numbers of patients with haemodynamically significant valvular dysfunction and selection or referral bias in a tertiary hospital resulted in uneven distributions of valvular dysfunction, with a ratio of AR to AS patients of 1.75 (84/48). This may explain, at least in part, the failure to demonstrate any meaningful association between AS severity and aortic enlargement. In contrast, our study used echocardiographic data from 595 patients with BAV and thus we could include 177 BAV patients with normal or haemodynamically insignificant valvular dysfunction (29.7%) and larger numbers of patients with haemodynamically significant AS or AR. Thus, our ratio of AR to AS patients was 1.1 (407/365).

Different impact of AS and AR severity on aortic dilatation in different levels of the ascending aorta demonstrated in our study may represent marked heterogeneous features of bicuspid aortopathy. Distinct patterns of aortic dilatation was well described in the literature9 ,21 and potential association between AR severity and the diameter of aortic sinus alone was also documented by other investigators.9 Both our study and the previous report7 ,22 found that age was independently associated with aortic enlargement in BAV patients. This finding, along with the significant positive independent correlation between aortic diameters and degree of valvular dysfunction, suggests that a long-standing haemodynamic burden associated with progressive valvular dysfunction is one important component of bicuspid aortopathy.23

Annual dilatation rate after AVR

Our serial echocardiographic measurements of aortic dimension showed that successful isolated AVR had protective effects in BAV patients. In contrast to a previous study,10 we found that the annual dilatation rate after successful AVR did not differ between patients with BAV and TAV, further indicating that haemodynamic burden is the main cause of bicuspid aortopathy. Interestingly, despite the significantly younger age and lower prevalence of haemodynamically significant valvular dysfunction in our BAV-NAVR group, factors that might have protective effects against aortic enlargement in BAV patients, the annual dilatation rate was higher in our BAV-NAVR than in our BAV-AVR and TAV-AVR groups. This finding reinforces the importance of haemodynamic burden to aortic dilatation in BAV patients and the protective effects of isolated AVR. The original finding, that the proximal aorta shows progressive enlargement after isolated AVR was based on very limited patient numbers10 and has not been adequately confirmed. Due to the absence of well-designed, long-term, longitudinal, follow-up clinical studies, the incidence of aortic dissection in this particular clinical setting has not been determined. Moreover, several recent reports have concluded that aortic size is not the only factor associated with development of aortic dissection.24 ,25

Limitations

Many phenotypic patterns of BAV and bicuspid aortopathy have been described26–28 and a recent report suggested the possibility of different impact of BAV phenotypic patterns on dilatation rate of the ascending aorta in young patients with BAV.29 We did not evaluate the potential impact of these different BAV phenotypes on the pathogenesis of aortic enlargement, as such an evaluation would require much larger numbers of patients, and may therefore not be possible in a single centre study. In addition, echocardiography has inherent limitations in the evaluation of the entire thoracic aorta and BAV phenotypes. For example, BAV patients may show localised arch aneurysm, which could not be measured accurately by echocardiography. New cardiac imaging modalities, including CT and MRI, may be better for the accurate evaluation of the BAV phenotypes and the entire thoracic aorta. Although annular dilatation rate was not different in follow-up study, significant proportion of patients with BAV showed rapid annular dilatation after AVR during follow-up, but our study could not identify any associated factors. Further prospective investigations are necessary to overcome potential selection biases in our retrospective study and to confirm factors associated with aortic dilatation in patients with BAV.

Clinical implications

A critical review of bicuspid aortopathy based on objective data is necessary, as the prevailing inherent fragility theory undoubtedly affects the surgical approach to patients with this common clinical entity. Our findings suggest that a chronic haemodynamic burden is one of the important contributing factors of bicuspid aortopathy and that isolated AVR is an effective method to prevent pathological progression. Recommendations in current practice guidelines should therefore be re-evaluated through better clinical investigations and better data analysis.

References

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:

Footnotes

  • YGK and BJS contributed equally.

  • Contributors Each author substantially contributed to the research. In detail: YGK, BJS, GMP and JKS contributed to the conception and design of the study; YGK, SH, DHKim, JMS, DHKang and JKS contributed to data collection, analysis and interpretation of the results; YGK and JKS drafted and revised the manuscript. All authors read and approved the final version of the manuscript.

  • Funding This study was supported by a grant of the Korea Health Technology R&D Project, Ministry of Health & Welfare, Republic of Korea (A100591).

  • Competing interests None.

  • Ethics approval Institutional Review Board.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.