Procedural Pattern, Clinical Outcomes, and Cost Analysis of Transcatheter Aortic Valve Replacement in Korea

Abstract

Background

Little information exists about national-based data on patients undergoing transcatheter aortic valve replacement (TAVR) in Asia and the associated costs.

Objectives

This study aims to assess the baseline characteristics, procedural patterns, clinical outcomes, and cost expenses of patients with severe aortic stenosis who underwent TAVR in Korea.

Methods

We used the patient-level data from the National Health Insurance Review and Assessment Service registry, spanning from May 2015 to June 2021. The primary clinical outcome was all-cause mortality. Rehospitalization rates and procedural costs were also analyzed.

Results

During the study period, 3,463 patients (mean age: 80.5 ± 5.9 years; 54.0% female) underwent TAVR at 39 hospitals in Korea. The annual number of TAVR procedures and the number of performing centers increased steadily. The proportion of low-risk patients (Society of Thoracic Surgeons Predicted Risk of Mortality ≤4%) was 41.7%. The transfemoral approach was conducted in 99.0% of cases, and the procedural success rate was 99.7%. All-cause mortality rates at 30 days, 2 years, and 5 years were 2.7% (95% CI: 2.1-3.2), 12.9% (95% CI: 11.8-14.0), and 19.5% (95% CI: 18.0-21.1), respectively. Rehospitalization related to cardiovascular causes occurred in 31.7% (95% CI: 29.3-34.1) at 2 years and 40.5% (95% CI: 34.1-46.9) at 5 years. The estimated mean index cost per patient was $32,173 ± $7,552, with 83.2% attributed to material expenses.

Conclusions

The number of TAVR recipients and performing centers in Korea has substantially increased over time. The procedural and survival outcomes were favorable. The overall medical costs for TAVR procedures are still substantial, mainly driven by device costs.

Central Illustration

For elderly patients with severe aortic stenosis (AS), transcatheter aortic valve replacement (TAVR) has emerged as a minimally invasive alternative to traditional surgical aortic valve replacement (SAVR). Since its first successful implementation in 2002,^1,2 TAVR has dramatically evolved over the last 2 decades. Based on cumulative evidence from multiple large-sized randomized clinical trials,3, 4, 5, 6 TAVR has become the established treatment for severe, symptomatic AS across all surgical risk categories. Nowadays, TAVR has been positioned as a standard therapeutic approach in several clinical guidelines.^7,8

Among Asian countries, the adoption of TAVR in Asia was slower than in the United States and Europe, mainly due to the lack of full reimbursement for TAVR and local regulations for new device approval.⁹ In March 2010, the first TAVR procedure was performed in Korea, following formal approval by the Korean Food and Drug Administration for the treatment of severe symptomatic AS.¹⁰ Given the rapid increase in life expectancy in Korea, there is a projected increase in the proportion of elderly individuals with degenerative severe AS in the near future,¹¹ leading to a significant increase in the anticipated number of TAVR candidates over time. However, until recently, there has been a paucity of research on the patient characteristics, procedural patterns, clinical outcomes, and cost implications of TAVR in Korea.^9,10

Therefore, by analyzing the entire patient-level dataset obtained from the national TAVR registry in Korea, which was a mandatory requirement for reimbursement, we sought to gain insights into the real-world implementation of TAVR and its impact on clinical outcomes. In addition, we assessed the financial burden of TAVR on the health care system by examining the direct costs associated with TAVR procedures, including hospitalization, medical supplies, and follow-up care.

Methods

Data source

The current study used individual patient-level data acquired from the National Health Insurance Review and Assessment (HIRA) to identify all consecutive patients who underwent TAVR in South Korea from May 1, 2015, through June 30, 2021 (HIRA research data M20230410004). All TAVR procedures performed during this period were included in the analysis to represent TAVR procedures performed with the latest devices in high-risk, intermediate-risk, and low-risk patients. Medical insurance coverage mandates hospitals to submit all relevant information for patients who receive commercially approved TAVR devices in the dedicated HIRA system. Standardized data elements and regular quality checks are implemented by the HIRA national registry; all information on demographic, procedural, and clinical outcomes variables should be prospectively registered (Supplemental Table 1). The HIRA system strictly controlled the dataset for these registries, and most variables were verified against the original electronic medical records. Baseline demographics, medical history, patient characteristics, procedural details, index procedural outcomes, and clinical outcomes before discharge were obtained from the HIRA-TAVR registry. All participating sites are regularly monitored for independent audits of the HIRA. Institutional review boards at the Asan Medical Center, University of College of Medicine, approved this study and granted a waiver of informed consent.

For further cost and long-term mortality analysis, we have cross-linked the HIRA data with the National Health Insurance (NHI) database in Korea (Figure 1). The NHI is the only mandatory social insurance service that provides health coverage to all citizens in Korea. All health care providers are required to enroll in the NHI system and are reimbursed on a fee-for-service basis. The dataset contains information on more than 50 million patients, representing 98% of the total population covered by the universal coverage system.12, 13, 14 The HIRA-TAVR registry included the unique resident registration number of each patient, which was used to merge the database of medical expense records and the NHI database. Therefore, we used the HIRA medical expense record to assess all cost-related data and cardiovascular rehospitalization. For cost analysis, we used the classification system provided by the HIRA system for categorizing each expense item. Cardiovascular rehospitalization was defined as events admitted with primary diagnoses corresponding to International Classification of Diseases, 10^th version codes I00-I99, which represent diseases of the circulatory system.

Figure 1.

Data Integration Schematic

This schematic represents the integration of three datasets: 1) Health Insurance Review and Assessment (HIRA): Transcatheter Aortic Valve Replacement (TAVR) registry, detailing procedures linked by the resident registration number (RRN); 2) HIRA: Medical Expense Record, categorizing the costs associated with each medical service; and 3) National Health Insurance Services (NHISS): All-cause mortality, indicating instances of loss of eligibility and their dates. The central integrated data record (red box) combines data from these sources using the RRN as a key identifier. For security purposes, RRNs are encrypted within the integrated dataset to ensure confidentiality and protection of personal information while facilitating comprehensive data analysis. STS = Society of Thoracic Surgeons. Costs are presented in US Dollars (USD).

The status of mortality was assessed by examining the insurance qualification status of patients through the NHI database. In the event of a patient's death being registered with the national system, it results in an immediate loss of insurance eligibility. For this study, the status of mortality was assessed as of April 1, 2023. This registration serves as a definitive indicator for determining a patient's survival status. This dataset included data from up to 2 years preceding the data extraction point.

Outcomes

The primary outcome was all-cause mortality. Secondary outcomes included a 30-day composite complication outcome (ie, stroke, moderate or severe paravalvular leak, major vascular access-site complications, Valve Academic Research Consortium major or life-threatening or disabling bleeding, and acute kidney injury) and outcomes for each component of the composite outcome (Supplemental Material).

Statistical analysis

A descriptive analysis was performed to assess the baseline characteristics. Categorical variables are presented as frequencies and percentages. Continuous variables are summarized as means with standard deviations or medians with interquartile ranges. All-cause mortality was estimated using the Kaplan-Meier method, with event or censoring times calculated from the date of enrollment. Event rates were calculated based on Kaplan-Meier estimates in time-to-first-event analyses and were compared using the log-rank test. For the cost analysis, descriptive statistics were used, and the mean values and standard deviations of the collected data were calculated. The statistical analyses were conducted using SAS Enterprise Guide version 9.4.2 and R server version 3.5.1.

Results

Patient characteristics and procedural outcomes

From 2015 through 2021, a total of 3,463 TAVR procedures were performed at 39 hospitals in Korea. The number of cases performed had substantially increased over time (Figure 2). Concomitantly, the number of institutions performing TAVR in South Korea had also increased from 10 centers in 2015 to 38 centers in 2021 (Figure 3); of those, the majority (74.4%) were classified as small-volume centers with fewer than 100 cases overall (Supplemental Figure 1).

Figure 2.

Number of TAVR Procedures in Korea

Annual number of TAVR procedures performed in Korea from 2015 to 2020. Abbreviations as in Figure 1.

Figure 3.

Number of TAVR Centers in Korea

Number of nationally certified centers for TAVR in Korea from 2015 to 2021. Abbreviations as in Figure 1.

Baseline demographics, comorbidities, and anatomical characteristics are described in Table 1. The mean age of the patients was 80.5 ± 5.9 years and 54.0% (n = 1,868 of 3,463) were female. According to the Society of Thoracic Surgeons risk assessment score, 41.7% (n = 1,442 of 3,463) of patients were in the low-risk category, 31.3% (n = 1,083 of 3,463) were in the intermediate-risk category, and 27.1% (n = 937 of 3,463) were in the high-risk category. Approximately 60% (n = 2,127 of 3,463) of patients presented with NYHA functional class III or IV dyspnea. Half of the patients had concomitant coronary artery disease. Degenerative AS was a predominant cause of AS and a bicuspid valve was present in 8.7% (n = 300 of 3,463) of patients.

Table 1.

Baseline Characteristics of the Patients (N = 3,463)

Age, y	80.5 ± 5.9
Female	1,868 (54.0)
Diabetes	1,166 (33.7)
Requiring oral hypoglycemic agents	1,033 (29.8)
Requiring insulin	133 (3.8)
Hypertension	2,638 (76.2)
Liver disease	88 (2.5)
Lung disease	651 (18.8)
Neurologic disease	571 (16.5)
TIA	48 (1.4)
CVA with full recovery	310 (9.0)
CVA with residual deficit	135 (3.9)
Others	77 (2.2)
Peripheral artery disease	331 (9.6)
Previous cardiac surgery	208 (6.0)
Isolated CABG	116 (3.6)
CABG with aortic-valve surgery	8 (0.2)
Isolated aortic-valve surgery	57 (1.8)
Other cardiac operation	26 (0.8)
Previous percutaneous valvuloplasty	21 (0.6)
Aortic valve alone	14 (0.4)
Aortic valve and mitral valve	1 (0.0)
Mitral valve alone	5 (0.1)
Previous PCI	950 (27.4)
Previous TAVR	15 (0.4)
Risk group according to STS score
Low risk (STS < 4)	1,442 (41.7)
Intermediate risk (4 ≤ STS < 8)	1,083 (31.3)
High risk (8 ≤ STS)	937 (27.1)
NYHA functional classification
I	147 (4.2)
II	1,188 (34.3)
III	1,537 (44.4)
IV	590 (17.0)
Coronary artery disease
Any	1,728 (49.9)
One-vessel disease	770 (22.2)
Two-vessel disease	506 (14.6)
Three-vessel disease	388 (11.2)
Left main disease	174 (5.0)
Renal replace treatment	192 (5.5)
Serum creatinine level, mg/dL	1.3 (1.4)
LVEF, %
Mean value	58.03 ± 12.0
Category
≥50%	2,794 (80.7)
40% to 50%	307 (8.9)
<40%	361 (10.4)
AV mean pressure gradient, mm Hg	52.61 ± 18.69
AV peak velocity, m/s	4.76 ± 3.98
Effective orifice area, cm2	0.71 ± 0.22
Index effective orifice area, cm2/m2	0.45 ± 0.14
Annulus diameter, mm	22.55 ±16.94
Etiology of aortic stenosis
Degenerative	3,267 (94.3)
Rheumatic	35 (1.0)
Others	161 (4.6)
Aortic valve morphology
Unicuspid	5 (0.1)
Bicuspid	300 (8.7)
Tricuspid	3,151 (91.0)
Quadricuspid	6 (0.2)
Aortic regurgitation
None or mild	2,809 (81.1)
Moderate	552 (15.9)
Severe	101 (2.9)
Mitral regurgitation
None or mild	3,078 (88.9)
Moderate	347 (10.0)
Severe	37 (1.1)

Values are mean ± SD or n (%).

AV = aortic valve; CABG = coronary artery bypass grafting; CVA = cerebrovascular accident; LVEF = left ventricular ejection fraction; PCI = percutaneous coronary intervention; STS = Society of Thoracic Surgeons; TAVR = transcatheter aortic valve replacement; TIA = transient ischemic attack.

Baseline procedural characteristics and outcomes are shown in Table 2. Nearly all (99.0%; n = 3,428 of 3,463) TAVR procedures were performed via the transfemoral approach. The overall procedural success rate was 99.7% (n = 3,452 of 3,463), and serious procedural complications were rare. A permanent pacemaker was implanted in 7.0% (n = 243 of 3,463) of patients. By echocardiography, the effective orifice area has significantly increased from 0.71 ± 0.22 to 1.76 ± 0.51 cm². After the TAVR procedure, most patients experienced symptom relief, with 98.8% (n = 3,384 of 3,463) of patients reporting only mild dyspnea (NYHA functional class ≤2). Moderate or severe paravalvular leakage was present in 5.1% (n = 179 of 3,463).

Table 2.

Procedural Characteristics and Outcomes (N = 3,463)

Total length of stay, days	9 (6-15)
Length of stay after TAVI procedure, days	5 (3-8)
Access site approach
Transfemoral	3,428 (99.0)
Transapical	20 (0.6)
Transsubclavian	7 (0.2)
Transaortic	7 (0.2)
Post-balloon	930 (26.9)
Procedural success	3,452 (99.7)
Procedural complications	37 (1.07)
Emergent surgery	18 (0.5)
Valve migration to aorta	3 (0.08)
Valve migration to LV	2 (0.06)
LV perforation	6 (0.17)
Annular perforation	1 (0.03)
Aortic dissection	2 (0.06)
Coronary obstruction	1 (0.03)
Others	4 (0.12)
Urgent PCI	21 (0.6)
Insertion of ECMO	44 (1.3)
Permanent pacemaker implantation, %	243 (7.0)
Post-TAVR Vmax, m/s	2.35 ± 0.49
Post-TAVR EOA, cm2	1.76 ± 0.51
Paravalvular leak, %
None or trace	1,908 (55.1)
Mild	1,375 (39.7)
Moderate	178 (5.1)
Severe	1 (0.0)
Post-TAVR LVEF, %	59.77 ± 10.69
Post-TAVR NYHA functional classification
I	2,802 (80.9)
II	582 (16.8)
III	45 (1.3)
IV	32 (0.9)

Values are mean ± SD, median (Q1-Q3), or n (%).

ECMO = extracorporeal membrane oxygenation; EOA = effective orifice area; LV = left ventricle; LVEF = left ventricular ejection fraction; PCI = percutaneous coronary intervention; Vmax = maximum velocity; other abbreviation as in Table 1.

In-hospital outcomes and long-term mortality

Clinical outcomes during the index hospitalization are summarized in Table 3. Overall, 101 (2.9%) patients died during hospitalization, more than half of which (55.2%) were attributed to a cardiovascular cause. The occurrence of new conduction disorders was detected in 23.5% of patients. The observed rate of post-procedural cerebral infarction was 2.1%. Bleeding events occurred in 9.4% of the patients, of which more than half (56.3%) were minor bleeding.

Table 3.

Clinical Outcomes at the Index Hospitalization (N = 3,463)

All-cause mortality	101 (2.9)
Cardiovascular death	57 (1.6)
Noncardiovascular death	44 (1.3)
Cardiac tamponade	62 (1.8)
Coronary obstruction	8 (0.2)
Myocardial infarction	6 (0.2)
Conduction disorder	812 (23.5)
Stroke
Ischemic	60 (1.7)
Hemorrhagic	6 (0.2)
Undetermined	6 (0.2)
Bleeding events
Life-threatening or disabling	42 (1.2)
Major bleeding	101 (2.9)
Minor bleeding	184 (5.3)
Aortic dissection	10 (0.3)
Access site complications	217 (6.27)
Access site dissection	37 (1.1)
Access site rupture	26 (0.8)
Access site hematoma	154 (4.4)
Lower extremity ischemia	19 (0.5)

Values are n (%).

The all-cause mortality rates at 30 days, 2 years, and 5 years were 2.7% (95% CI: 2.1-3.2), 12.9% (95% CI: 11.8-14.0), and 19.5% (95% CI: 18.0-21.1), respectively (Figure 4). The cumulative incidence of hospital readmission due to cardiovascular causes within the first 2 years was 31.7% (95% CI: 29.3-34.1), which increased to 40.5% (95% CI: 34.1-46.9) over the 5-year period (Figure 5).

Figure 4.

Five-Year Event Curves of All-Cause Mortality

Kaplan-Meier cumulative event curves for 1 month, 2-year, and 5-year all-cause mortality.

Figure 5.

Two-Year and 5-Year Event Curves of Re-Admission Rate for Cardiac Cause

Kaplan-Meier cumulative event curves for 2-year and 5-year re-admission rate for cardiac causes.

Cost analysis

The schematic diagram of the cost analysis using combined HIRA and NHI sharing service data is shown in Figure 1. The mean ± SD per-patient cost was $32,173 ± $7,552 (exchange rate: 1 USD = 1,300 KRW). Analysis based on Korea's itemized billing system revealed diverse criteria. Specifically, the breakdown of costs under physician fees, medication, and material expenses was $4,899, $547, and $26,770, respectively. Material costs constituted the largest proportion, accounting for 83.2% of the total expenses. When excluding TAVR-device materials, the highest cost was attributed to treatment ($1,984; 6.16%), followed by hospitalization expenses, including intensive care unit costs ($1,590; 4.94%) (Table 4).

Table 4.

Total Index Cost

Total Cost ($1 = ₩1,300)	n = 3,462	Mean ± SD		$32,173.52	±7,552.44
Classification Typea			Categoriesa
Physician fee	4,899.06	± 4,158.67	Consultation	352.70	± 241.94
Medication	547.43	± 610.75	Admission	1,590.36	± 1308.04
Material costs	26,770.33	± 4,624.62	Medication	100.35	± 108.23
			Injection	469.55	± 1148.47
			Anesthesia	246.22	± 142.62
			Treatment	1,984.70	± 3,001.25
			Examination	1,258.75	± 930.97
			Radiation	956.29	± 1,029.09
			Etcb	25,265.61	± 4,306.43

Values are mean ± SD.

^aThe medical expense statements were categorized according to the national foundation data items. For the “classification type,” distinctions were made ^based on the treatment sector codes, whereas for “categories,” details were noted based on individual item-specific codes. This follows the code classification criteria of the national health insurance system of South Korea.

^bFor “categories,” non-insured items are tallied separately. As a result, THV is listed under "Etc" as a non-insured item. Costs are presented in US dollars.

A subsequent evaluation of medical costs up to 2 years post-TAVR was conducted (Table 5). This analysis involved 1,727 patients who had completed at least 2 years of follow-up at the time of assessment. The average total cost during this period was approximately $3,612. Physician fees accounted for most expenses, totaling $2,954, which represents 81.7% of the total cost. Some patients required re-admission, and the average cost associated with hospitalization (excluding costs of patients who did not require re-admission) was $2,348.

Table 5.

Follow-Up Cost

Total Cost ($1 = ₩1,300)	n = 1,727	Mean ± SD	3,612.12		± 10,376,89
Classification Typea			Categories
Physician fee	2,954.79	± 8769.22	Consultation	301.75	± 617.84
Medication	469.77	± 1513.66	Admission	2,348.36	± 6,122.31
Material costsb	1,334.29	± 2950.18	Medication	161.92	± 359.96
		± 55.83	Injection	610.36	± 2,294.45
		± 0.53	Anesthesia	165.52	± 277.18
		± 43.77	Treatment	1,941.56	± 4,413.97
		± 152.02	Examination	732.56	± 138.31
			Radiation	197.70	± 819.68
			Etcb	482.45	± 2,699.18

Values are mean ± SD.

^aThe medical expense statements were categorized according to the national foundation data items. For the “classification type,” distinctions were made based on the treatment sector codes, whereas for “categories,” details were noted based on individual item-specific codes. This follows the code classification criteria of the national health insurance system of South Korea.

^bThe average of the detailed variables is obtained only for the costs incurred. Costs are presented in US dollars.

Discussion

This study is the first nationwide investigation that consolidated data from multiple public institutions on all TAVR procedures in Korea. Despite the limited reimbursement system in which patients are responsible for 80% of total TAVR costs, the number of patients undergoing TAVR and the number of TAVR centers have substantially increased over a 5-year period. The procedural outcomes and short- and long-term mortality after TAVR were favorable. The overall medical costs for TAVR procedures are still substantial, which was mainly driven by device costs.

We found that the number of institutions performing TAVR in Korea has increased by more than 3-fold. In 2019, there were 34 TAVR centers in Korea as opposed to 646 centers in the United States and 170 centers in Japan,¹⁵ which translates to 6.5 centers per 100,000 people in Korea compared to 20.1 centers per 100,000 people in the United States and 13.6 centers per 100,000 people in Japan.¹⁶ When confining the analysis to elderly patients 65 years of age and older, Korea had 4.4 centers per 100,000 people, whereas the United States had 12.3 centers per 100,000 people and Japan had 4.8 centers per 100,000 people. By 2030, the number of Koreans 65 years of age and older is expected to increase by approximately 1.7 times compared to 2019.¹¹ Thus, it is expected that the number of TAVR candidates or recipients may substantially increase over time.

In the current study, the mean age of TAVR recipients was 80.5 years. Given the predominance of TAVR in the elderly population, ensuring the safety of TAVR procedures is crucial. Our national registry-based analysis showed an in-hospital mortality rate of 2.9%, with cardiovascular deaths accounting for 1.6% of the cases. In addition, the procedural success rate was approximately 99%. Overall rates of procedure-related complications, stroke, and bleeding events were also comparable to those reported in other large-sized registries or national studies.⁹ Given the rapid increase in the elderly population in Asian countries, a significant increase in the number of TAVR procedures is expected; this is especially true with the availability of more affordable devices, leading to broader use of TAVR. Further racial- and ethnic-based comparative studies will provide a better understanding of the observed racial differences in the clinical and anatomic characteristics and outcomes of TAVR recipients between Asian and Western populations, which may facilitate risk stratification and the development of novel race-based approaches for TAVR procedures.

Furthermore, we conducted a long-term analysis of all-cause mortality incorporating the latest national statistical mortality data. Our study reported 2-year and 5-year mortality rates of 12.9% and 19.5%, respectively. Considering the average life expectancy in Korea as of 2020 (83.5 years) and the average age of TAVR patients (80.48 years), the observed survival rates may indirectly imply a significant improvement in the care of TAVR recipients. Notably, TAVR procedures were directly associated with a significant improvement in patients’ symptoms. Before the TAVR procedure, 61.4% of patients experienced significant limitations in daily activities due to dyspnea (NYHA functional class III or IV); however, following TAVR, only 2.1% of patients reported a similar level of dyspnea. Dyspnea in the elderly is known to correlate with physical and mental constraints, which are factors contributing to increased mortality.¹⁷ Our large-scale study showed a significant impact of TAVR on mortality and symptom improvement in Korean patients.

Lastly, it is important to consider the medical costs associated with the TAVR procedure. Whereas TAVR offers advantages such as shorter hospital stays and reduced risks compared to SAVR, the high cost of TAVR presents a significant challenge in medical care. Our research on the Korean health care system revealed that the average total cost of TAVR-related hospitalization for all patients was approximately $32,173, with procedural materials accounting for 83.2% of the cost ($26,770). In Korea, the cost of the TAVR system, regardless of transcatheter heart valve type, ranges from $23,808 to $25,061 (subject to exchange rate fluctuations). Given that the overall material cost averages $26,770, it is evident that the TAVR system itself constitutes most of the material expenses. A cost analysis from the Placement of Aortic Transcatheter Valves (PARTNER) II trial reported an average cost of $61,400 for hospitalization, with procedural costs accounting for 62.7% of the total expense. Follow-up cost analysis showed that the index cost per patient was approximately $54,256, and follow-up costs amounted to $46,284, representing 85% of the initial expenses. In contrast, our study found that the follow-up cost in Korea was only $3,612, which is 8.9% of the initial cost. This significant difference is largely due to Korea's national health care policies, which result in lower medical expenses.¹⁸ Although the overall TAVR costs in Korea are relatively lower compared to those in the United States, significant financial constraints are still associated with the widespread use of TAVR. Considering the projected increase in life expectancy in Korea, it may be crucial to develop cost-effective policies to address the financial implications of TAVR in the future.

Study limitations

First, the use of nationwide data resulted in the unavailability of patients' personal information. We relied on data submitted by health care institutions to administrative agencies, which limited our access to comprehensive patient details. To enhance data reliability, we implemented a cross-referencing approach by matching the TAVR registry with patients' medical records maintained by the administrative agency. Second, the absence of a control group for SAVR has imposed limitations on the development of a comprehensive cost- and treatment-effect model for TAVR, including the quality-adjusted life year and the incremental cost-effectiveness ratio analysis. Third, as strokes related to TAVR were only classified into bleeding or ischemic causes in the database, we could not provide comprehensive information on stroke severity (eg, disabling or nondisabling). Fourth, when calculating re-admission rates, we used data extracted from International Classification of Diseases diagnosis codes, leading to the inclusion of re-admissions specifically related to cardiovascular causes. Consequently, the detailed reasons for re-admission after TAVR, including non-TAVR-related admissions, could not be ascertained, potentially leading to an overestimation of rehospitalization rates. Fifth, interpreting the 5-year survival data would be challenging without a comparator group. The observed 5-year mortality (approximately 19%) in our study is relatively low compared to other populations, such as the 46% for intermediate-risk patients in a previous trial.⁴ This might be due to racial differences and selection bias in our study population. Finally, our cost analysis relied on patients' payment statements, which encompassed all medical expenses incurred during their treatment. Although we made efforts to exclude costs associated with materials and procedures such as percutaneous coronary intervention (PCI) during the hospitalization period, there remains a possibility of including unrelated costs in the analysis. Additionally, the absence of censoring in the follow-up cost analysis may result in less accurate cost estimates over the entire follow-up period.

Conclusions

The number of TAVR recipients and performing centers in Korea substantially increased from 2015 to 2021. The procedural outcomes and short- and long-term survival after TAVR were favorable. The overall medical costs for TAVR procedures are still substantial, which is mainly driven by device costs (Central Illustration). Based on these findings, further research into cost-related policies for TAVR procedures is crucial to address the economic challenges and improve accessibility and outcomes for patients in Korea.

Central Illustration.

Use of TAVR in Korea Substantially Increased Between 2015 and 2020

Although transcatheter aortic valve replacement (TAVR) use led to favorable outcomes, its high costs, primarily due to device expenses, introduce challenges in the Korean health care system.

Funding Support and Author Disclosures

This research was supported by a grant from the Patient-Centered Clinical Research Coordinating Center (PACEN) through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health and Welfare, Republic of Korea (grant number: HC21C0180). The funding organization had no role in the design or conduct of this research. Dr S.J. Park has received grants from Edwards Lifesciences, Medtronic, and Abbott. Dr D.W. Park has received grants and consulting, and speaker fees from Edwards LifeSciences, Medtronic, Abbott, and Daiichi-Sankyo during the conduct of the study. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.