# The aortopathy of bicuspid aortic valve

Hardikar, AA ORCID: 0000-0003-3558-6454 2021 , 'The aortopathy of bicuspid aortic valve', PhD thesis, University of Tasmania.

 PDF (Whole thesis) Hardikar_whole_...pdf | Document not available for request/download Full text restricted until 1 July 2023.

## Abstract

Bicuspid aortic valve (BAV) is the most common congenital cardiac anomaly with a prevalence of 0.5% to 2.0 %. Thoracic aortic dilatation (over 40 mm diameter) happens in 40% of cases with BAV in their lifetime. The most important risk in this subset is development of acute aortic syndromes, encompassing aortic dissection, rupture and sudden death. The guidelines around imaging, diagnosis, surgical intervention and surveillance have evolved over the last 20 years, but we are still far from perfect in predicting which BAV aortopathy would lead to acute aortic syndrome and when. Given the worldwide volume of cases, this poses a significant burden on the healthcare systems across the world. Because the Tasmanian Cardiothoracic service is in the rare position of complete capture of presentation and follow-up, I started this work by establishing a Tasmanian registry of BAV and aortic pathology (number 0013456) and affiliated this to the BAV-CON consortium of registries. I then studied BAV aortopathy in Tasmanian patients with special reference to current guidelines, mechanobiology, longitudinal studies of aortic growth, economic implications and finally an assessment of the results with emphasis on quality of life and healthcare delivery across socio-geological boundaries.
The work in this thesis is divided into six sections. The first section (Chapters 1-2) deals with the background issues with BAV aortopathy. From 1998 to 2018, 12 different international guidelines have focused on aortopathy related to BAV, despite almost no new insights into this disease. Influenced by these guidelines, surgery for BAV aortopathy in the USA rose by more than 3-fold between 1998 and 2009, stabilizing after publication of a more conservative guideline approach in 2014. Despite the fact that 40% of thoracic aortic dissections happen in aorta with diameters less than 5 cm; aortic size alone has been the main determinant of the timing of surgical intervention. Actually, the physics of thoracic aortic enlargement based on law of Laplace would suggest that blood pressure, aortic diameter and aortic thickness are all determinants of aortic wall stress and that the likelihood of rupturing a stressed aorta is also related to the nature of the constituent aortic tissue. I reviewed the literature published on this topic and summarised it in our publication. I proposed that a composite index in predicting aortic complications would work better than aortic size alone.
The second section (Chapters 3 to 6) aims to elaborate on the mechanobiological aspect of thoracic aortic dilatation in BAV. There are 2 main theories to explain the aortopathy associated with BAV. The genetic theory proposes that a genetic abnormality is the cause of weakened aortic wall and hence causes the aortopathy. The hemodynamic theory proposes that mechanical factors are the main drivers affecting aortic dilatation. Applying the law of Laplace, it is clear that besides aortic diameter, increasing blood pressure and thin or weaker aortic wall will also have a role to play in prediction of acute aortic events.
Chapter 3 deals with the role of aortic thickness in predicting aortic wall stress. The study aimed at risk-stratifying aortic dilatation using aortic wall thickness (AWT) and comparing methods of AWT assessment. Data was collected for 72 consecutive aortic surgeries and AWT was measured on imaging studies as well as intra-operatively in 4 quadrants using calipers. Aortic wall stress was calculated using standard formulae. I found congruence between thickness measured by CT or actually on table. I found that the convex or right sided aortic wall was preferentially thinned out as compared to the concave left sided wall. This would explain the preponderance of aortic dissection originating on the right side.
Chapter 4 deals with patterns of aortopathy and the effects of aortic angulation or geometry on these patterns. Various methods of aortopathy classification have been described and I proposed a way which has implications on the surgical management. I postulated that aortic geometry including angles and asymmetry might impact pattern of dilatation and rate of aortic growth. I analysed 221 patients with dilated aortas, including 102 with BAV. The aortic valve angle was significantly different between BAV and tricuspid aortic valves (TAV) (45$$^0$$ vs 41$$^0$$ ; p<0.001) I found that male sex and hypertension predicted diffuse pattern of aortopathy while aortic stenosis and statin therapy correlated with localised aortopathy. I found that an angle between aortic root and ascending aorta of more than 21$$^0$$ is a significant correlate of aortic dilatation.
The next chapter discusses the effect of aortic valve morphology on the aortic wall displacement. I recruited 150 patients, 50 BAV and 100 TAV undergoing transesophageal echocardiography prior to aortic valve surgery. The longitudinal and radial displacements of the major and minor aortic curvatures were assessed using velocity vector imaging. I proposed a global radial displacement index, which was greater in BAV patients than TAV patients after adjustment for age, sex, pulse pressure and body surface area. I also studied aortic compliance, distensibility and stiffness in the same cohort. This greater displacement would explain why BAV aortas are comparatively larger than TAV aortas.
Chapter 6 studies the correlation of aortic histopathology with patterns of aortic dilatation and compares and contrasts the differences between BAV and TAV. The aortic media is the strongest layer contributing to the tensile strength and compliance, consisting of concentrically arranged elastin sheets, collagen fibres, smooth muscle cells, glycoprotein microfibrils and fibrillin scaffold for elastin. I looked all the aortic pathology reports from our Tasmanian BAV and aortic pathology registry and segregated the BAV and TAV subgroups. I graded cystic medial necrosis, elastin fragmentation, myxoid degeneration, calcification, fibrosis, haemorrhage, areas of necrosis. Special notes were made about changes of chronic inflammation, degree of atherosclerosis or hyaline degeneration. I found a minority of cases with histiocyte or giant cell infiltration as well. I found the BAV subgroup to have increased changes of inflammation. The degree of elastic fragmentation, cystic medial necrosis as well as fibrosis was more with TAV.
The third section is devoted to the longitudinal study of aortic growth. The guidelines focus on aortic dimensions at a static point in time. Aortic dilatation is a dynamic process and hence I place more importance on study of aortic growth. Aortic growth rates are an important tool of dynamic assessment of the aortic pathology, and have been incorporated into the guidelines of aortic repair since 2014. I started in chapter 7 by reviewing 29 papers (with a total of 4528 patients) on aortic growth, using Forest plots to calculate mean growth rates across different populations. The mean growth rates at aortic root, ascending aorta, arch and descending thoracic aorta were 0.54, 0.68, 0.97 and 2.47 mm/yr. Bicuspid aortic valve type, younger age, genetic syndromes, aortic valve pathology, smoking, diastolic blood pressure, renal failure and vascular disease were reported accelerators of growth. Previous aortic surgery, anticoagulant usage and tricuspid aortic valve type were protective factors. I concluded that assessment and documentation of aortic growth rate is vital in individual patient risk profile and the knowledge of factors accelerating aortic growth would help identify the subset of thoracic aortas which have a higher risk of acute aortic events.
Chapter 8 details the rate of aortic growth in various segments of thoracic aorta in our Tasmanian registry patients. In this longitudinal study of cohort of 221 patients with aorta > 4 cm and with at least 2 CT aortogram studies in our registry, 102 were BAV patients and the scanning interval varied from 11 months to 3.5 years. Our group had 145 operated and 76 unoperated patients with a total 747 patient-years follow up. I found that the average rate of growth is low (<1mm/year), especially in postoperative patients. A cut off point of 21 degrees for the angle between aortic root and ascending aorta accurately predicted the top quintile of aortic growth. The rate of aortic growth was directly proportional to the initial aortic diameter implicating a significant role for aortic size in prediction of the growth rate. The growth rate was also higher with increasing aortic stenosis or regurgitation and whether the patient had been operated or not. I looked at the comparative rates amongst various studies to better understand heterogeneity. I looked at different factors impacting the rate of aortic growth at 4 different levels of thoracic aorta to present a summary view.
Chapter 9 assesses the role of aortic valve leaflet fusion pattern in predicting aortic dilatation. Sievers classified BAV based on number of cusps and fusion patterns. Fusion of right and left cusps (Sievers type 1A) and right and non-coronary cusps (Sievers type 1B) are the commonest subtypes of BAV. I analysed 102 consecutive cases of aortic dilatation associated with BAV and found that 74 were type 1A and 25 were type 1B. I compared the demographics, morphometrics, clinical features and aortic growth rates as well. There have been a few other studies which have evaluated this topic, and half of them do not show any relation of leaflet fusion pattern with type of aortic dilatation. I found that the 1A subtype had more males and the patients were taller and bigger with preponderant aortic stenosis. However, I did not find any significant differences between the groups while comparing aortic diameters, or angles or growth rates at different levels of thoracic aorta.
The guidelines should be built on data, and reliability is essential. I hence sought to examine the interobserver and intra observer agreement in reporting CT aortogram results in section 4. Chapter 10 deals with assessment of 20 randomly selected CT aortograms by 6 experienced clinicians and the agreement was evaluated.
We live in an era of increasing burden on a healthcare system with limited resources, and decisions involving a disease with 1-2% population prevalence are likely to be impactful. Section 5 (Chapter 11) elaborates on the economic implications of early aortic surgery in BAV aortopathy. In the absence of randomised trials, the optimal management of a patient with BAV and aorta dilated above 5 cm remains unclear. Using a Markov model, I hence sought to compare the strategies of immediate surgical management and watchful waiting using a decision-analytic model, using a 65 year old male with a 5 cm sized thoracic aorta as the “base case” scenario, taking the perspective of the healthcare provider. I found that the net monetary benefit of aortic surgery over watchful waiting is better only if the rate of acute aortic events is over 35% per year. Incremental cost-effectiveness plotting showed that watchful waiting was more effective and less costly than immediate surgery.
In the last section I looked at the results and analysed the problems with healthcare delivery for patients with BAV aortopathy in Tasmania. Chapter 12 looks at the results achieved in Tasmanian population. Of the 553 patients in the study, 208 had BAV, and of these, 131 underwent only aortic valve replacement while 77 patients had aortic surgery - 40 were Bentall’s or valve-sparing root replacement (VSRR) procedures, while 37 were root sparing procedures. The TAV subset had 221 isolated aortic valve replacements, 70 root sparing aortic surgeries and 54 Bentall’s or VSRR procedures. Immediate postoperative complications including death, stroke, renal failure, infections, sternal dehiscence and bleeding were noted and all the patients were followed for a mean period of 5.4 years. These patients were then contacted either in person at follow up clinics or telephonically and a standard questionnaire using the EuroQOL quality of life protocol was asked. I also kept a record of the medication profiles of these patients.