Transcatheter aortic valve implantation for management of patients with severe aortic stenosis

of aortic stenosis There are 3 main etiologies of aortic stenosis (AS): congenital aortic valve disease (unicuspid or bicuspid), rheumatic heart disease, and degeneration of aortic valve. Among elderly, the most common cause is severe calcification and degeneration of aortic valve leaflets, resulting in reduction of aortic valve area. AS will induce trans-aortic valve gradient, causing several pathophysiological consequences, including (1) left ventricular hypertrophy and systolic dysfunction, (2) diastolic dysfunction, (3) myocardial ischemia due to reduced coronary perfusion, (4) low periperal perfusion, (5) post-stenotic aortic dilation. 1.2. Clincal and sub-clinical manifestations of AS patients Clinical symptoms include fatigue, decrease exercise tolearnce, dypsnea on exertion, angina, syncope, and end-stage heart failure symptoms. Echocardiogram is the basic investigation for the diagnosis, evaluation, and prognosis of AS. Echocardiographic assessment with provide indications of aortic valve replacement (AVR). Low-dose dobutamine stress echocardiogram is performed to differentiate true severe AS from pseudosevere AS in low flow-low gradient patients. 1.3. Management of AS 1.3.1. Medical management Medical management of heart failure may reduce symptoms, but do not prolong life in symptomatic patients. 1.3.2. Balloon aortic valvuloplasty Percutaneous balloon aortic valvuloplasty relieves stenosis by fracturing calcific deposits within the valve leaflets, resulting in valve opening and lower trans-aortic gradient. However, the restenosis rate is high. Therefore, balloon aortic valvuloplast has limited role clinical settings of congenital severe AS, and as a bridge to AVR.. 1.3.3. Surgical aortic valve replacement (SAVR) SAVR has been the mainstay of treatment for symptomatic AS patients, as it offers substantial improvements in symptoms and life expectancy. However, there are several SAVR-related complications, 3 such as severe bleeding, infection, biosthetic valve dysfunction. In clinical practice, at least 33% of patients cannot undergo surgery. For these patients, a less invasive treatment may be a worthwhile alternative. 1.4. Transcatheter aortic valve implantation (TAVI) Since the first-in-human transcatheter aortic valve implantation (TAVI) performed by Alain Cribier in 2002, this innovative procedure has gained widespread recognition as the treatment of choice for severe aortic stenosis. Over the same period, TAVI has evolved from a challenging intervention to a standardised, simple, and streamlined procedure. The standardized TAVI protocol includes the following steps: (1) Heart team discussion to stratify surgical risk, (2) echocardiogram, (3) ECG-gated MSCT of the aortic valve to choose the suitable size of the device, (4) TAVI procedure in the cathlab or hybrid operation room (OR). Potential complications of TAVI include: death, stroke, myocardial infarction, bleeding, device embolization, acute kidney injury, permanent pacemaker implantation, paravalvular leak, Compared with SAVR, the higher incidence of pacemaker implantations is the only remaining concern of TAVI. Clinical trials have shown the efficacy of TAVI in many clinical settings: - In PARTNER 1B, TAVI is superior to medical treatment in inoperable patients - In PARTNER 1A, TAVI is similar to SAVR in high surgical risk populations - In PARTNER 2 and SURTAVI trials, TAVI has comparable outcomes with SAVR in intermediate-risk patients. - In PARTNER 3, NOTIOn, and EVOLUT trials, TAVI is similar of better than SAVR in low-risk patients With recent results from these studies, the indications for TAVI have changed dramatically, shifting quickly from compassionate cases, to inoperable and high-risk patients, and more recently towards intermediate-risk and low-risk populations. 4 Table 0.1: AHA/ACC 2017 guidelines for indication of TAVI Recommendations Class TAVI is recommended for symptomatic patients with severe AS (Stage D) and a prohibitive risk for surgical AVR who have a predicted post-TAVR survival greater than 12 months IA Surgical AVR or TAVI is recommended for symptomatic patients with severe AS (Stage D) and high risk for surgical AVR, depending on patient-specific procedural risks, values, and preferences IA TAVI is a reasonable alternative to surgical AVR for symptomatic patients with severe AS (Stage D) and an intermediate surgical risk, depending on patient-specific procedural risks, values, and preferences IIa-B CHAPTER 2 METHODS 2.1. Location and time of the research This study was implemented at 5 following cardiology centers across Vietnam, from July 2013 to July 2019: - Vietnam National Heart Institute, Bach Mai Hospital; - Heart center, Hanoi Medical University Hospital; - Cardiology Department, Ho Chi Minh city Medical-Pharmacy University hospital; - Heart center, Vinmec International Hospital Times City; - Dong Do Heart Hospital. 2.2. Research participants TAVI procedure was done on 48 patients with aortic stenosis at cardiology hospitals/departments. Eligibility criteria included: - Patients with indication of TAVI, according to the 2012 ESC guidelines for management of valvular heart diseases, with updated recommendations from 2014 ACC guidelines: o Symptomatic severe AS, NYHA II or higher o Asymptomatic AS with low EF (EF<50%) o Low gradient AS, with clinical symptoms - Patients not suitable to SAVR, due to one of the following reasons: 5 o High surgical risk, according to the STS score o Inoperable patients due to other comorbidities, determined after consultation with cardio-thoracic surgeon. o Patients refuse to have surgery - Patients agree to participate in the study 2.3. Research methods 2.3.1. Study design Clinical intervention without a control group 2.3.2. Sample size and sampling methods Using the sample size formula for one-sided one sample proportion test, with hypothesized mortality rate during the first one year of 10% (as reported by the TAVR-Asia study), a minimum sample size of 47 patients was required. All patients with aortic stenosis meeting the eligibility criteria were recruited and treated with TAVI procedure at all cardiology centers participating in the study, until the minimum sample size was reached. 2.3.3. Study procedures These study procedures were done at all heart centers, following the of ESC guidelines for management of valvular heart diseases. • Enrollment: clinical diagnosis, classification of surgical risks, explanation of benefits and risks of the procedure and the study (Low surgical risk if STS<4%; intermediate risk if STS from 4-8%; high risk if STS>8%). • Evaluation before TAVI: clinical examination, history taking, ECG, cardiac ultrasound, Doppler ultrasound of peripheral arteries, MSCT of the aortic valve (following guidelines of the American Society of Cardiovascular Computed Tomography), coronary angiography • TAVI procedure following guidelines of ESC, using one of 3 biosynthetic valves: CoreValve (Medtronic), Hydra (Vascular Innovations), Evolut R (Medtronic) • The procedure was considered successful if: (1) the valve was placed at the right place, with trans-aortic pressure less than 20mmHg; (2) not having one of these severe complications: death, stroke, myocardial infarction, acute aortic dilation, aortic dissection • Close follow-up of patients after the procedure at the ICU: detection and management of complications: death, myocardial infarction, 6 stroke, bleeding, arterial complications, arrhythmias, paravalvular leak • Evaluation of patients through clinical examination and ultrasound immediately after the procedure, 30 days after the procedure and 1 year after the procedure (if happening during the study period). Follow-up indexes included: mortality, NYHA classification, CCS, cardiac ultrasound indexes (aortic valve area, aortic pressure, LVEF) Figure 2.1. Study protocol 2.3.4. Data analysis Data was entered with the Epi Data software and was analyzed with Stata 14.0. Descriptive data was reported as means and standard deviations, proportions. Hypothesis tests were done at p<0,05. Kaplan- Meier test was used in the survival analysis. 7 2.5. Ethics The study was approved by the Institutional review board at Hanoi Medical University and participating heart centers before implementation. Personal identification data of patients was coded, kept confidential, and only used for study purposes. CHAPTER 3 RESULTS Between July 2013 and July 2019, 48 patients underwent TAVI were enrolled in the study. 3.1. Patients characteristics 3.1.1. Clinical manifestations Table 3.1. Baseline characteristics of the patients Parameters (n=48) Mean or Percentage Male sex (%) 31 (64,6%) Age (year) 75 ± 7 (56-87) NYHA class III-IV 40 (83,3%) Cardiogenic shock 2 (4,2%) Invasive mechanicl ventilation 5 (10,4%) Medical history Hypertension 28 (58,3%) Diabetes 17 (35,4%) Chronic pulmonary disease 17 (35,4%) Coronary heart disease 14 (29,2%) STS score 5,8 ± 3,7% STS < 4% 18 (37,5%) STS 4-8% 21 (43,8%) STS > 8% 9 (18,7%) Inoperable due to comorbidities 3 (6,3%) 8 3.1.2. Echocardiographic findings Table 0.2: Echocardiographic findings of patients Parameters (n=48) Mean or Percentage Degenerative AS 48 (100,0%) Aortic valve area (cm2) 0,60 ± 0,19 Maximum trans-aortic gradient (mmHg) 91,7 ± 26,3 Mean trans-aortic gradient (mmHg) 57,0 ± 17,8 Moderate-severe aortic regurgitation (≥ 2+) 15 (31,3%) Moderate-severe mitral regurgitation (≥ 2+) 18 (37,5%) Ejection fraction (%) 54,7 ± 17,8 EF ≥ 50% 32 (66,7%) EF 30 - 49% 13 (27,1%) EF < 30% 3 (6,2%) Pulmonary artery systolic pressure (mmHg) 47,0 ± 9,8 3.1.3. MSCT findings Figure 0.1: Prevalence of bicuspid aortic valve There were 23 patients with bicuspid aortic valve (47,9%). 23 (47,9%) 25 (52,1%) Bicuspid Tricuspid 9 Table 0.3: MSCT findings of the patients Parameters (n=48) Mean value Min Max Minimum annulus diameter (mm) 21,4 ± 2,5 15,7 28,0 Maximum annulus diameter (mm) 27,1 ± 2,6 21,7 32,8 Average annulus diameter (mm) 24,2 ± 2,2 20,2 28,6 Annulus perimeter (mm) 76,6 ± 6,8 63,3 90,8 Aortic valve area (cm2) 0,46 ± 0,08 0,32 0,64 Distance between annulus and left coronary artery ostium (mm) 19,8 ± 3,3 10,9 24,3 Distance between annulus and right coronary artery ostium (mm) 15,4 ± 3,7 12,3 23,8 Ascending aorta diameter (mm) 35,1 ± 5,4 20,3 49,0 Aortic angulation (degree) 49,7 ± 8,6 35,0 76,3 3.2. Outcomes of TAVI 3.2.1. Procedural characteristics Figure 0.2: Procedural success rate The procedural success, defined as post-procedural trans-aortic gradient less than 20mmHg, with no life-threatening complication, was achieved in 47/48 cases. The success rate was 97,9%. There was one failure due to left ventricular perforation, resulting in death before device implantation. 47 1 Success Failure 10 Table 0.1: Procedural characteristics Parameters n=48 Percentage Location Cathlab 40 83,3 Hybrid OR 8 16,7 Anaesthesisa General anaesthesisa 46 95,8 Local anaesthesia 2 4,2 Access sites Trans-femoral 48 100,0 Trans-carotid 0 0,0 Direct aortic access 0 0,0 Vascular access Percutaneous Seldinger technique 43 89,6 Surgical cut-down 5 10,4 Intra-procedural TEE 11 22,9 Balloon aortic valve pre-dilatation 39 81,3 Balloon aortic valve post-dilatation 12 25,5 Conversion to surgery 1 2,1 All TAVI cases were performed using trans-femoral approach. Seldinger technique was the main way for vascular access, there were only 5 cases with surgical cut-down (10,4%). There were 11 cases with intra-procedural trans-esophageal echocardiography, accounting for 22,9%. 11 Figure 0.3: Biosynthetic valves Most of TAVI cases were performed using Evolut R (51,1%). Figure 0.4: Size of biosynthetic valves The most common sizes were 29mm (24 cases, 50,1%), and 26mm (38,3%). There were only one 30mm case and two 31mm cases. 24 (51,1%) 14 (29,8%) 9 (19,1%) Evolut R CoreValve Hydra 2 18 24 1 2 23mm 26mm 29mm 30mm 31mm 12 Table 0.5: Invasive trans-aortic gradient Trans-aortic gradient Pre-TAVI Post-TAVI p CoreValve (n=14) 67,8 ± 6,6 7,8 ± 2,0 <0,01 Hydra (n=9) 79,9 ± 10,5 12,0 ± 1,7 <0,01 Evolut R (n=24) 69,3 ± 4,7 10,8 ± 2,0 <0,01 Overall (n=47) 68,4 ± 23,0 10,1 ± 8,2 <0,01 There was a significant reduction of trans-aortic gradient, regardless of type of biosynthetic valve used. 3.2.2. Complications of TAVI 3.2.2.1. Death Figure 0.5: Peri-procedural death There were 4 peri-procedural deaths (the mortality rate was 8,3%), including one intra-procedural death due to left ventricular perforation, and three post-procedural deaths due to severe infection. 4 (8%) 44 (92%) Peri-procedural death Alive 13 Figure 0.6: Mean STS score among death and alive groups Patients who died during hospitalization had a significantly higher STS score (11.5%, compared with 5.5% in the alive group, p<0,05). 3.2.2.2. Other complications Figure 0.7: Complications of TAVI 11.5 5.5 0 3 6 9 12 Mean STS score Death Alive 1 3 1 1 2 3 10 3 6 9 5 1 2 4 0 2 4 6 8 10 12 14 Figure 0.6: Prevalence of para-valvular leak Moderate PVL No PVL / Mild PVL Bicuspid valves 4 (17,4%) 19 (82,6%) Tricuspid valves 0 (0,0%) 24 (100,0%) Bicuspid valves related to a significantly higher incidence of para-valvular leaks (PVL). All 4 cases with moderate PVL were in the bicuspid group. 3.2.3. Long-term follow-up Besides 4 hospitalization deaths, 44 other patients were followed up, with mean duration of 26,4 months (5-68 months). 3.2.3.1. Survival rate Figure 0.8: Kaplan-Meier survival estimation 15 Table 0.7: Mortality rate during following-up Time Before Death After Cumulative survival rate (%) Cumulative mortality rate (%) 6th month 48 4 43 91,7 8,3 12th month 43 0 38 91,7 8,3 18th month 38 0 25 91,7 8,3 24th month 25 2 19 84,3 15,7 30th month 19 2 16 75,2 24,8 36th month 16 1 13 70,2 29,8 42th month 13 1 7 63,2 36,8 48th month 7 1 4 50,5 49,5 54th month 4 0 4 50,5 49,5 60th month 4 1 3 37,9 62,1 The estimated survival rate after 12 months was 91,7% (95% CI: 79,3-96,8%). Survival rate was 84,3% after 24 month, and 70,2% after 36 months. Figure 0.8: Determinants of mortality during following-up Factors Hazard ratio 95% CI p Age > 80 0,5 0,1-2,5 0,42 NYHA class III-IV 5,1 1,5-18,0 0,01 EF < 50% 1,9 0,6 – 6,3 0,29 Bicuspid aortic valves 1,3 0,4 – 4,6 0,64 STS score > 8% 5,7 1,7 – 19,3 0,005 Permanent pacemaker after TAVI 1,2 0,1-9,5 0,89 NYHA class III-IV and STS score > 8% were the only two determinants of mortality after TAVI. 16 3.2.3.2. Clinical following-up Figure 0.9: NYHA class before and after TAVI TAVI improved functional class of AS patients. After one year, there was no patients with NYHA class III or IV. 3.2.3.3. Echocardiographic findings Figure 0.10: Trans-aortic gradient and AVA 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Baseline At discharge After 1 month After 12 months NYHA 4 NYHA 3 NYHA 2 NYHA 1 57.0 11.4 11.4 0.60 1.45 1.52 0 0.4 0.8 1.2 1.6 0 10 20 30 40 50 60 Baseline After 1 month After 12 months A o rt ic v al v e ar ea ( cm 2 ) T ra n s- ao rt ic g ra d ie n t (m m H g ) Trans-aortic gradient Aortic valve area 17 Figure 0.11: Patients’ EF before and after TAVI CHAPTER 4 DISCUSSION 4.1. Patient selection for TAVI procedure 4.1.1. Clinical characteristics In our study, 83.3% of participants were 70 years old or older. The mean age was 75,3. This was higher than the mean ages of patients undergoing SAVR. Older patients tended to have other comorbidities, which increased surgical risks. All of our patients had at least one metabolic disorder, such as hypertension, type II diabetes, coronary heart disease or chronic kidney disease. This finding was similar to other studies on AS. Since the mechanism of degenerative AS was increased inflammation and proliferation, with risk factors similar to those of atherosclerosis, aortic stenosis was usually accompanied by other cardiac and metabolic disorders. 4.1.2. Anatomic characteristics of aortic valve All 48 patients had aortic stenosis due to wearing and deterioration of the valve over time. It was expected as this was the most common cause of aortic stenosis among elderly people. 35.8 64.2 54.7 50.6 67 61.9 63.9 68.3 66.9 0 10 20 30 40 50 60 70 80 EF < 50% EF ≥ 50% Overall L e ft v e n tr ic u la r E F ( % ) Baseline After 1 month After 12 months 18 The prevalence of bicuspid aortic valve was 47,9%. Bicuspid valves had certain anatomic characteristics that were barriers to the TAVI procedure, such as asymmetry, severe calcification, dilated ascending aorta, etc. This has been considered a contraindication of TAVI. However, recent studies showed that TAVI could be done safely to patients with bicuspid valves, with complication and mortality rates not significantly different from those with tricuspid valves. Our study confirmed this finding. Hemodynamic outcomes did not depend on the anatomy of the aortic valves, and the mortality rates of those with bicuspid valves and tricuspid valves were similar. Table 4.1. Trans-aortic gradients before and after TAVI Type of valve Before TAVI (mmHg) After TAVI (mmHg) p Bicuspid 69,3 ± 24,9 9,6 ± 9,8 <0,01 Tricuspid 67,4 ± 21,6 10,5 ± 6,4 <0,01 Overall 68,4 ± 23,0 10,1 ± 8,2 <0,01 4.1.2. Surgical risk stratification The mean surgical mortility rate in our study, according to the STS score, was 5,8 ± 3,7%. The most common category was medium risk (STS 4-8%), with 43,8% falling into this category. This finding reflected the reality in clinical practice Vietnam, where many patients refused to have a surgery despite not having a high surgical risk. STS score before the procedure was associated with the prognosis of patients. Low-risk patients (STS<4%) has the lowest all-cause mortality rate, meanwhile high-risk patients (STS>8%) has the highest mortality rate. 19 Figure 4.1. Mortality rate according to STS score Moreover, two patients had hematologic diseases, one patient had cirrhosis at Child Pugh C stage. Even though these patients did not have high surgical risk, the cardiac surgeons recommended against surgery due to coagulation dysfunction. In reality, there were patients with even low risk according to the STS or EuroSCORE II scales, but surgical management was not appropriate. This finding underscored the importance of a new classification of surgical risk that can help physicians selecting the suitable candidates for TAVI. 4.2. Subclinical evaluation before TAVI Transthoracic echocardiogram (TTE) has been the routine evaluation for patients with AS. However, echocardiogram could only provide the 2D imaging of the heart (unless 3D trans-esophageal echocardiogram was performed). Meanwhile, aortic valve is a 3D structure, with an elliptical shape ring, with long and short axes. Therefore, TTE might yield inaccurate measurements. MSCT scan of the aortic valve was required before TAVI, so that appropriate equipment could be determined. A study among patients with AS undergoing SAVR showed that the mean diameter of the valve on MSCT was 24,0 ± 2,1mm, which was close to the diameter 20 measured during surgery (23,8 ± 0,2mm). In our study, the mean diameter of aortic valve was 24,2 ± 2,2mm, similar to other findings. 4.2. Characteristics of the TAVI procedure in Vietnam 4.1.2. Success rate of the procedure The procedure was successfully performed on 47 out of 48 patients (97,9%). Hemodynamic assessments post-procedure showed significant improvements in trans-aortic gradient. Our findings were consistent with those of other studies in the world. For example, in the ADVANCE trial, 96% of patients achieved a trans-aortic gradient under 20 mmHg. In the Asian-TAVR study, the success rate was 97,5%. We concluded that TAVI was a procedure with high success rate. 4.1.2. Characteristics of the procedure There were some changes in the protocol of TAVI at the heart centers in Vietnam between the period 2013-2016 and the period 2017- 2019. First, hybrid operation rooms have been used recently, with less procedures done in the cathlab. Second, following global trends, TAVI procedure has been simplified, with increasing use of the Seldinger method for obtaining vascular access, decreasing use of transesophageal echocardiography. Predilatation of the aortic valve before TAVI has also become less common. Performing transfemoral TAVI under local anaesthesia, using a fully percutaneous approach, and eliminating transoesophageal echocardiographic guidance (maintaining back-up transthoracic echocardiography) and balloon predilation are strategies to reduce the invasiveness and costs of the procedure. 21 Figure 4.2. Changes in TAVI procedures over time 4.1.2. Sizes of biosynthesis valves We used 26mm and 29mm valves for most of our patients. In the Asian-TAVR study, which analyzed data from TAVI cases in Asia from 2010 to 2014, the majority patients received 23mm and 26mm valves. Therefore, our patients received larger valves. Our study was conducted in a later period than the Asian-TAVR study, from 2013 to 2019, when there was a global trend of using larger artificial valves to optimize effective oriface area and to avoid para-valvular leaks. 4.4. Complications of TAVI Among our 48 patients, one case failed due to left ventricular perforation when the equipment was inserted through the ventricular wall, leading to patient’s death. Left ventricular perforation was a rare but severe complication of TAVI. When perforation happened, the only way to save patients was to convert to an open heart surgery. It was critical to have good collaborations with the cardiothoracic surgical 100 91.3 30.4 86.9 68 100 16 76 0 10 20 30 40 50 60 70 80 90 100 TAVI in cathlab Seldinger for vascular access Intra-procedural TEE Balloon aortic predilatation P ro p o rt io n s (% ) 2013-2016 (23 patients) 2017-2019 (25 patients) p 0,05 p < 0,05 p = 0,16 22 team, so that emergency managements of severe complications could be done successfully and the safety of the procedure could be guaranteed. Apart from one patient who died during the procedure, 3 more patients died after the surgery. In-patient mortality rate was 8,3%. All of these cases happened during the initial period of the study. From 2013 to 2016, 4 out of 23 patients died (17,4%), while no patient died during the period from 2017-2019 (0,0%). Similar to findings in other studies, the mortality rate in our study decreased over time. In a study among 32.400 TAVI cases in Germany from 2008 to 2013, in-patient mortality rate decreased from 9,1% in 2008 to 5,6% in 2013. 4.5. Longitudinal follow-up of patients 4.5.1. Survival rate One-year survival rate was 91,7%, similar to other trials such as PARTNER 2, Asian-TAVR and ADVANCE. These studies showed that higher age, chronic kidney and lung diseases, peripheral vascular diseases and higher STS score were prognostic factors of mortality after TAVI. However, in our study, except for an STS score of more than 8% and functional symptoms at NYHA III-IV, other clinical characteristics and history of patients did not predict post-procedure mortality. 4.5.2. Clinical follow-up TAVI significantly improved clinical symptoms of patients. The proportion of patients with NYHA III-IV decreased form 83,3% before the procedure to 18,0% after the procedure. At the one-year follow-up visit, no patients had NYHA III-IV functional class. These findings were comparable to other studies in the world. 4.5.3. Echocardiographic follow-up TTE showed decreased trans-aortic gradients and increased aortic valve areas. There were no differences in the assessments at 30 days and one year post-procedure. Other studies in the world also found similar results. Patients with EF <50% had a mean change in EF of 28,1% (from 35,8% to 63,9%), while patients with EF ≥ 50% had a lower increase of 4,1% (from 64,2% to 68,3%). Therefore, patients with lower EF had significan