Surgical aortic valve replacement etiologies, hemodynamics, and outcomes in 1346 patients from the Malaysian heart centre

Background This study examined the characteristics and outcomes of surgical aortic valve replacement (SAVR) both isolated and in combination with other cardiac surgery in Malaysia from 2015 to 2021. Methods This was a retrospective study of 1346 patients analyzed on the basis of medical records, echocardiograms and surgical reports. The overall sample was both considered as a whole and divided into aortic stenosis (AS)/aortic regurgitation (AR)-predominant and similar-severity subgroups. Results The most common diagnosis was severe AS (34.6%), with the 3 most common etiologies being bicuspid valve degeneration (45.3%), trileaflet valve degeneration (36.3%) and rheumatic valve disease (12.2%). The second most common diagnosis was severe AR (25.5%), with the most common etiologies being root dilatation (21.0%), infective endocarditis (IE) (16.6%) and fused prolapse (12.2%). Rheumatic valve disease was the most common mixed disease. A total of 54.5% had AS-predominant pathology (3 most common etiologies: bicuspid valve degeneration valve, degenerative trileaflet valve and rheumatic valve disease), 36.9% had AR-predominant pathology (top etiologies: root dilatation, rheumatic valve disease and IE), and 8.6% had similar severity of AS and AR. Overall, 62.9% of patients had trileaflet valve morphology, 33.3% bicuspid, 0.6% unicuspid and 0.3% quadricuspid. For AS, the majority were high-gradient severe AS (49.9%), followed by normal-flow low-gradient (LG) severe AS (10.0%), paradoxical low-flow (LF)-LG severe AS (6.4%) and classical LF-LG severe AS (6.1%). The overall in-hospital and total 1-year mortality rates were 6.4% and 14.8%, respectively. Pure severe AS had the highest mortality. For AS-predominant pathology, the etiology with the highest mortality was trileaflet valve degeneration; for AR-predominant pathology, it was dissection. The overall survival probability at 5 years was 79.5% in all patients, 75.7% in the AS-predominant subgroup, 83.3% in the AR-predominant subgroup, and 87.3% in the similar-severity subgroup. Conclusions The 3 most common causes of AS- predominant patients undergoing SAVR is bicuspid valve degeneration, degenerative trileaflet valve and rheumatic and for AR-predominant is root dilatation, rheumatic and IE. Rheumatic valve disease is an important etiology in our SAVR patients especially in mixed aortic valve disease. Study registration IJNREC/562/2022.


Background
Surgical aortic valve replacement (SAVR) is an established treatment for aortic valve dysfunction, be it aortic stenosis (AS), aortic regurgitation (AR) or a mixture of the two (mixed) [1].There is a paucity of data on diagnosis, etiologies and outcomes in developing middleincome countries such as Malaysia.This is especially important for 2 reasons: (1) rheumatic valve disease (hereinafter, "rheumatic") accounts for a substantial proportion of our SAVR population [2]; and (2) our SAVR population is mainly of Malay, Chinese and Indian ethnic backgrounds, and the etiologies and outcomes in this study might therefore be dissimilar to those in previous studies performed in Western countries.In addition, bicuspid aortic valve is the most common congenital cardiac condition [3], and its prevalence in Malaysian SAVR recipients is not known.Therefore, we sought to ascertain the following in all our SAVR recipients from 2015 to 2021: (1) the diagnosis leading to SAVR, (2) the etiology and outcomes of SAVR (both overall and divided into AS-predominant, AR-predominant, and similar-severity subgroups), (3) the prevalence and outcomes of various SAVR etiologies and (4) the prevalence and outcomes of low-gradient severe AS in our AS patients.

Patient population
From the surgical registry, we analyzed all patients who underwent SAVR from 2015 to 2021 at the Institut Jantung Negara (IJN, National Heart Institute), which is Malaysia's premier cardiac center and has the highest volume of cardiac surgeries for a single center in Southeast Asia (SEA).The study analyzed all SAVR recipients, including those who received isolated SAVR, double (mitral and aortic valve) replacement (DVR), SAVR and/ or DVR with concomitant coronary artery bypass grafting (CABG), and SAVR with root replacement.We included all patients aged 12 years and above who did not have complex congenital heart disease.Both mechanical and biological SAVR are included as well (Fig. 1).This study was approved by the IJN (Institut Jantung Negara, Kuala Lumpur, Malaysia) ethics committee (IJNREC/562/2022).
Overall, 1503 patients underwent SAVR at IJN from 2015 to 2021.A total of 157 patients were excluded (63 patients excluded due to complex congenital heart disease and 94 excluded due to incomplete follow-up), and the other 1346 were analyzed.The included patients were considered overall and divided into AS-predominant, AR-predominant, and similar-severity subgroups.Mortality outcomes from both the diagnosis and diagnosis subgroups and for the different etiologies in the subgroups were analyzed.

Diagnosis, mechanisms and hemodynamics
Diagnosis in this study refers to the type of dysfunction (stenosis vs. regurgitation vs. mixed) and its severity (mild vs. moderate vs. severe).Etiology refers to the mechanisms or causes of the dysfunction (bicuspid valve degeneration, trileaflet valve degeneration, rheumatic, dissection, etc.).The analysis was performed for a) the patients overall and b) the AS-predominant, AR-predominant, and similar-severity subgroups.For SAVR involving any degree of AS, we analyzed the echocardiographic features, particularly the aortic valve area and aortic valve area index (AVA and AVAi, respectively), mean pressure gradient across the aortic valve (meanPG) and stroke volume index (SVI), to classify them into high-gradient severe AS (HG-AS, meanPG > 40 mmHg, AVA < 1.0 cm 2 ), classical low-flow low-gradient severe AS (CLF-LG-AS, meanPG < 40 mmHg, AVA < 1.0 cm 2 , EF < 50%, SVI < 35 ml/m 2 ), paradoxical low-flow low-gradient severe AS (PLF-LG-AS, meanPG < 40 mmHg, AVA < 1.0 cm 2 , EF > 50%, SVI < 35 ml/m 2 ), normal-flow low-gradient severe AS (NF-LG-AS, meanPG < 40 mmHg, AVA < 1.0 cm 2 , SVI > 35 ml/m 2 ) and reverse area-gradient mismatch AS.Some terms that need further clarification are "degeneration", which means calcification, thickening and/or retraction of leaflets causing stenosis and/or regurgitation, and "fused prolapse", which means that the fused leaflets of the bicuspid aortic valve prolapse and cause regurgitation.

Echocardiographic analysis
All preoperative transthoracic echocardiography (TTE) and intraoperative transesophageal echocardiography (TEE) recordings were analyzed by an expert echocardiographer specializing in valvular heart disease to determine the diagnosis, etiology and hemodynamics.Measurements of the aortic root (left ventricular outflow tract radius (LVOTr), sinus of Valsalva (SOV), sinotubular junction (STJ) and ascending aorta (AscAo)) and the number of leaflets (bicuspid vs. trileaflet vs. quadricuspid vs. unicuspid) were performed using TEE.The surgical reports where available was also analyzed to determine the number of leaflets but some of the reports does not have that information.Other pertinent echocardiographic parameters, such as biplane ejection fraction (EF), AVA, AVAi, peak velocity across the aortic valve (v max), meanPG, acceleration time (AT) and dimensionless velocity index (DVI), were analyzed using pre-and postoperative TTE.Preoperative TTE is echocardiography performed just before SAVR, and postoperative echocardiography is performed within 1 month after SAVR (if multiple scans were taken in that time, we considered only the first one after surgery).

Electronic medical records (EMRs)
We extracted each patient's age, gender and race from the EMRs.We analyzed in-hospital, 1-month, 6-month and 1-year mortality.Of these four variables, 1-month, 6-month, and 1-year mortality were defined to include only mortality after discharge from the hospital (i.e., inhospital mortality was not included).We also calculated the total 1-year mortality, which was defined to include all in-hospital and post-discharge mortality in the first year.Finally, we performed Kaplan-Meier analysis to  1).The most common diagnosis was severe AS (34.6%), followed by severe AR (25.5%), severe AS combined with moderate AR (14.3%), and moderate AR (8.8%).The mean age of patients with severe AS was 63.4 ± 10.3 years, and that of patients with severe AR was 46.7 ± 16.0 years.For the 3 largest etiology groups of severe AS, the mean age for SAVR was 67.9 ± 7.7 years for trileaflet valve degeneration, 59.5 ± 10.0 years for bicuspid valve degeneration and 49.5 ± 14.3 years for rheumatic (Table 2).The three most common causes of severe AS were bicuspid valve degeneration (45.3%, n = 211), followed by trileaflet valve degeneration (36.3%, n = 169) and rheumatic (12.2%, n = 57).There are many more varieties for severe AR.The most common etiology of severe AR was root dilatation (21.0%, n = 72) defined by any part of the aortic root with diameter > 40mm [4], followed by infective endocarditis (IE; 16.6%, n = 57), fused prolapse (12.2%, n = 42), rheumatic (8.7%, n = 30), prolapsed right coronary cusp (RCC; 7.6%, n = 26), unknown (5.2%, n = 18), dissection (5.0%, n = 17) and prolapsed left coronary cusp (LCC; 4.4%, n = 15).Rheumatic was the most common etiology for mixed diagnosis, commonest in moderate AS and moderate AR (48.5%, n = 33), severe AS and moderate AR (34.7%, n = 67), moderate AS and severe AR (35.3%, n = 12) and second most common for severe AS and severe AR (27.1%, n = 13).The most common diagnosis for mixed severe AS and severe AR was bicuspid valve degeneration (31.3%, n = 15).A complete list of the etiology for all the diagnoses is presented in Table 2.

Table 1 Demographics, procedure details and echocardiography
The majority of patients were male, and the most common ethnicity was Malay.The most common procedure was isolated AVR, followed by DVR and then AVR with CABG    For all our SAVR patients, the mean inpatient stay was 11.7 ± 12.8 days.Overall in-hospital mortality was 6.4%.After discharge, the 1-month mortality was 2.6%, the 6-month mortality was 4.7%, and the 1-year mortality was 7.0%.The overall 1-year mortality, including inhospital mortality, was 14.8%.In terms of mortality by Pure severe AS was the most common diagnosis, followed by pure severe AR.The most common etiology of pure severe AS was bicuspid valve degeneration, while that of pure AR was root dilatation.The most common diagnosis for AS predominant is bicuspid valve degeneration, and the most common diagnosis for AR predominant is root dilatation.Rheumatic is the most common diagnosis in the similar-severity group specific diagnosis, pure severe AS had the highest in-hospital mortality (8.2%) and total 1-year mortality (including in-hospital mortality) (17.6%).For the grouping, in-hospital mortality was highest for AS-predominant (7.6%), followed by AR-predominant (5.0%) and finally similar severity (4.3%).After discharge, 1-year mortality (excluding in-hospital mortality) also follows a similar trend, with AS-predominant at 8.2%, AR-predominant at 6.1% and similar severity at 4.0%.The overall 1-year mortality (including in-hospital mortality) was 17.2% for AS-predominant patients, 12.1% for AR-predominant patients and 10.0% for patients with similar severity.For isolated SAVR, in-hospital mortality is lower at 5.6% vs. 6.4% for overall SAVR patients and total 1-year mortality is also lower at 14.1% vs 14.8% for overall SAVR patients (Table 3).

Discussion
In this study, severe AS (34.6%) was the most common diagnosis, followed by severe AR (25.5%).In-hospital mortality is 6.4% and total 1-year mortality is 14.8% for overall SAVR patients and 5.6% (in-hospital) and 14.1% (total 1-year) for isolated SAVR patients.SAVR has been the mainstay of treatment for aortic valve dysfunction for more than 50 years since the first aortic valve replacement was performed in 1962 [1].SAVR has been successful, with an overall mortality rate of 2.5% for AS and less than 1% for asymptomatic severe AS in patients < 70 years old [5].As far as we are aware, there are no publications looking at etiology and outcome of SAVR in developing countries such as Malaysia.This is important, as the etiology may be different with more rheumatic heart disease and different racial and genetic make-up of our populations [6,7].In this study, patients with pure severe AS underwent SAVR at a later age (63.4 ± 10.3 years) than those with severe AR (46.7 ± 16.0 years).For severe AS, this is somewhat similar to the landmark Braunwald circulation paper in 1968, where the onset of symptoms was at 60 years old, whereas now in Western societies, it is Fig. 2 AS-predominant vs. AR-predominant vs. similar severity pathology.AS predominant was the most common pathology, followed by AR predominant and then similar severity Fig. 3 a Leaflet morphology.Leaflet morphology ascertained from TEE and/or surgical reports.The most common morphology was trileaflet (62.9%), followed by bicuspid (33.3%), unicuspid (0.6%) and quadricuspid (0.3%).In 2.9% of patients, the number of leaflets could not be determined.b AS classification, N = 974 Hemodynamics (gradient, flow and severity) for patients with any degree of AS.The most common category was HG-severe AS (49.9%), and the least common was reverse area-gradient mismatch (2.7%) Moderate AS and Moderate AR Severe AS, Moderate AR Moderate AS, Severe AR at 75 years old due to the lower prevalence of rheumatic heart disease [8].
For isolated severe AS, we had more patients with bicuspid valve degeneration (45.3%) compared to trileaflet valve degeneration (36.3%).A previous study in a Western population also showed slightly more bicuspid (49%) vs. trileaflet (46%) aortic valves [9].Rheumatic form a larger proportion compared to those from Western countries for both isolated severe AS (12.2%, 3rd most common) and even more so for AS predominant (19.5%, 3rd most common) [9,10].In fact, when we look at mixed aortic valve diseases, the most common etiology is rheumatic valve disease, especially when dealing with similar severity (moderate AS/AR, severe AS/AR).For AR, the etiology is much more diverse, with the most common cause being root dilatation (21%), followed by IE (16.6%) and fused prolapse (12.2%).Compared with contemporary data from the Mayo Clinic, cusp prolapse is the most common diagnosis, followed by degenerative and root dilatation.Their paper, however, did not separate fused prolapse (bicuspid) with single leaflet prolapse, whereas in this study, the frequency of each type of leaflet prolapse was ascertained [11].
As expected when we considered SAVR recipients overall as opposed to those with severe AS or ASpredominant pathology only, the majority had trileaflet valves (62.9%, n = 847), followed by bicuspid valves (33.3%, n = 448).This study had a higher percentage of bicuspid valves (33.3% vs. 23.0%)and a lower percentage of trileaflet valves (62.9% vs. 77.0%)than the 2020 study of SAVR in the Swedish population [12].Low gradient severe aortic stenosis may arise in patients with inherently small aortic annuli or low forward stroke volume due to poor systolic function, high arterial afterload, or small left ventricular chamber size [13].In this study, looking at severe AS (isolated or mixed), 31% (218/704) had low-gradient severe AS, which is less than the most recent study in Australia showing that half of severe AS cases have low-gradient hemodynamics [14].This is most likely due to less recognition and treatment for this confusing entity.The most recent entity recognized [15], reverse area-gradient mismatch, is the least common at 2.7%.
Overall in-hospital mortality was 6.4%.After discharge, the 6-month mortality was 4.7%, and the 1-year mortality (excluding in-hospital) was 7.0%.This appears The overall in-hospital and total 1-year mortality rates were 6.4% and 14.8%, respectively.Pure severe AS has the highest in-hospital and total 1-year mortality.
Bicuspid valve degeneration has had the best survival at 1 year for AS-predominant etiology, followed by rheumatic, and the worst is trileaflet valve degeneration.Dissection and IE have the worst outcome for AR predominant subgroup.Hemodynamically for AS, the poorest outcome occurs in LF-LG severe AS In-hospital" indicates those patients who died in-hospital before discharge Follow-up" indicates those patients who were alive at discharge but died at 1-month, 6-months and 1-year follow-up Total 1-year" indicates total number of patients who are dead at 1-year including in-hospital deaths higher than the Swedish study (30-day mortality = 2.3% and 1 year = 3.4%) [12].Possible reasons for this are later presentation, waiting too long before intervention, less access to and awareness of healthcare with less than 10 government hospitals that is able to provide cardiothoracic surgery services.Patients with pure severe AS had the highest inpatient and total 1-year mortality (8.2% and 17.6%, respectively).The same was also true with AS-predominant having higher mortality compared to AR-predominant and similar severity.Trileaflet valve degeneration was the etiology with the highest in-hospital and total mortality for AS-predominant patients (13.5% and 27.3%, respectively), most likely because this cohort of patients had the most advanced age at surgery.For both isolated severe AS and AS-predominant, the bicuspid has much better survival similar to study done in developed country [12].The total 1-year survival after SAVR for AS-predominant subgroup for rheumatic For the AS-predominant group, the probability of 5-year survival was 75.7%; for the AR-predominant group, it was 83.3%; and for the similar-severity group, it was 87.3% aortic stenosis (15.8%) is between trileaflet valve degeneration (27.3%) and bicuspid valve degeneration (8.5%).
In AR-predominant patients, the etiology with the highest in-hospital and total mortality was dissection (12.5% and 25.0%, respectively), followed by IE (10.9% and 20.4%, respectively).This is not surprising, as both of these conditions, even with modern treatment, are still very dangerous, as shown by a study in Finland after surgery for type A dissection (30-day mortality = 15.7%)[16] and another study by Nguyen et al. [17] showing high 5-year mortality after SAVR for IE.In terms of different severe aortic stenosis hemodynamics, the highest in-hospital and total 1-year mortality occurred in LF-LG severe AS (13.6%, 28.9%), and the lowest occurred in straightforward HG-severe AS (6.0%, 14.2%).LF-LG severe AS also had the poorest total 1-year mortality in an Australian study (30.5%), but in their study, NF-LG severe AS had the best total 1-year mortality (11.6%) [14].Reverse area-gradient mismatch had the best in-hospital and total 1-year mortality (3.8% and 12.5%, respectively) in this study.To the best of our knowledge, there are no studies on SAVR outcomes in this clinical entity.

Study limitations
This is a retrospective study and we cannot control for selection bias, confounding variables and generalizability.This study involved all patients who underwent SAVR regardless of concomitant mitral bypass grafts or root replacements; these concomitant surgeries can also affect a patient's outcome.Second, the EMR available in our center is still noncomprehensive, such that each patient's comorbidities could not be analyzed in detail; this shortcoming could also affect the study results.

Conclusions
The most common SAVR diagnosis was severe AS, followed by severe AR.Bicuspid valve degeneration, trileaflet valve degeneration and rheumatic were the 3 most common etiologies for both severe AS and AS-predominant patients.Root dilatation was the most common cause of severe AR.In-hospital mortality is 6.4% and total 1-year mortality is 14.8% for overall SAVR patients and 5.6% (in-hospital) and 14.1% (total 1-year) for isolated SAVR patients.Isolated severe AS, especially LF-LG severe AS, had the highest in-hospital and overall 1-year mortality.

Fig. 4 a
Fig.4a Kaplan-Meier curves for patients overall For all SAVR patients, the probability of survival at 5 years was 79.5%.b Kaplan-Meier curves for the AS-predominant, AR-predominant, and similar-severity groups.For the AS-predominant group, the probability of 5-year survival was 75.7%; for the AR-predominant group, it was 83.3%; and for the similar-severity group, it was 87.3%

Table 2
Diagnosis and etiology for SAVR patients

Table 3
Clinical outcomes