Transcatheter Aortic Valve Replacement in Patients With Geographic and Socioeconomic Vulnerabilities

Abstract

Background

Transcatheter aortic valve replacement (TAVR) is an established treatment for patients with severe aortic stenosis (AS). It remains unclear whether disparities exist in rural or socially vulnerable populations undergoing TAVR. This study assessed whether outcomes differ based on geographic location or vulnerability of patients undergoing TAVR.

Methods

Patients undergoing TAVR at a single institution from August 2012 to June 2023 were studied (n = 1565). Zip codes determined Social Vulnerability Index (SVI) and measured distance to our facility. Outcomes defined by the Valve Academic Research Consortium 3 (VARC-3) included stroke, transient ischemic attack (TIA)/delirium, pacemaker implantation, new atrial fibrillation/atrial flutter, and myocardial infarction (MI). Average time between preoperative coronary angiogram (CATH)/computed tomography angiography (CTA) and TAVR was calculated. Kaplan-Meier curves estimated survival probability.

Results

The average time between CATH and TAVR in patients living furthest away was ∼9 days more than patients living closest to the implant site. The average number of days between CATH and TAVR for low and high SVI were 71 and 78 days, respectively. The average number of days between CTA and TAVR for low and high SVI were 40 and 39 days, respectively.

Discussion

Further distances traveled were associated with longer wait times between preoperative workup and TAVR. Patients with longer waits between CATH and TAVR had no differences in medium-term survival probability but had decreased long-term survival probability compared to patients with rapid pre-procedural evaluation. Although geographic and socioeconomic vulnerability can disadvantage patients, our study demonstrates that patients undergoing TAVR can have timely care and similar outcomes.

Keywords

cardiac minimally invasive surgery socioeconomic thoracic surgery

Key Takeaways

• Geographical remoteness affects timely access to pre-operative testing and can cause delays in receiving TAVR but does not negatively impact short-term survival.

• Minimizing evaluation time before TAVR is critical to improving long-term outcomes.

• In a rural state, TAVR is safe and effective for patients living in socially vulnerable areas.

Introduction

Aortic stenosis (AS) is the most common form of valvular heart disease in the developed world, with nearly 7% of the population over 65 years of age suffering from this degenerative disease. While asymptomatic in its early stages, aortic valve replacement is the only effective treatment when severe. The first transcatheter aortic valve replacement in humans was in 2002¹ and the initial indication for TAVR was in high-risk surgical patients. The indications for transcatheter aortic valve replacement (TAVR) have expanded since its initial approval in 2012 and now include patients at all levels of surgical risk, including low risk. More patients who live remotely from implanting centers and tertiary hospitals are also undergoing TAVR. Currently, TAVR is the most common treatment for patients with severe aortic stenosis, overtaking surgical aortic valve replacement.² TAVR requires extensive preoperative evaluation and testing however, including computed tomographic angiography, cardiac catheterization, and consultation with a Structural Heart Team consisting of interventional cardiologists and cardiac surgeons experienced in treating patients with aortic stenosis. This testing and expertise is typically available at larger, tertiary care facilities rather than smaller hospitals.

Whether patients who live remotely or are socially vulnerable have access to these facilities and procedures is unclear. These patients may be at increased risk for negative outcomes after TAVR as geographic location and distance from health care negatively affects outcomes of patients with cardiovascular disease.^3-6 Additionally, socially vulnerable patients have increased rates of cardiovascular disease and cardiovascular mortality,^3,7,8 and underutilization of TAVR in Native Americans has been reported.⁹ Whether social vulnerability affects outcomes after TAVR is unclear, however. In this study, we investigate the impact of social vulnerability and remoteness from the tertiary care center on patients undergoing TAVR. We calculated social vulnerability on a county-by-county basis which incorporates sixteen social factors including unemployment, racial and ethnic minority status, disability, socioeconomic status, household characteristics, housing type, and transportation mode.¹⁰ We hypothesize that TAVR is a safe and effective procedure, regardless of the level of social vulnerability and remoteness from the TAVR institution.

Methods

North Dakota is one of the least densely populated states in the United States. Sanford Health Fargo is a large, tertiary care medical center located in a metropolitan area with a geographically large referral base in the upper Midwest. To determine the impact of geographic remoteness and social vulnerability on patients with severe aortic stenosis, all patients who underwent TAVR at Sanford Health Fargo from August 2012 to June 2023 were retrospectively reviewed (n = 1565). Google My Maps™ was used to create a map of the locations of each patients’ home address, with most of the patients being primarily from North Dakota and Minnesota (Figure 1). Valve Academic Research Consortium 3 (VARC 3) outcomes were captured including all-cause mortality, stroke, transient ischemic attack (TIA)/Delirium, myocardial infarction, new pacemaker implantation, new atrial fibrillation, new myocardial infarction, and transfusion status.¹¹

Geographic distance was determined using patient home zip codes and Google Maps™ to measure the distance from the implanting center. Distances were categorized into quintiles (Ranges: x < 14.9 miles, x = 15-49.9 miles, x = 50-99.9 miles, x = 100-149.9 miles, x = 150+ miles) and outcomes within each quintile quantified, including Kaplan-Meier survival estimate one year after TAVR. Additionally, the duration of the preoperative evaluation within each quintile was measured using the time from cardiac catheterization to valve implant and the time from computed tomographic angiography to valve implant.

Social vulnerability was quantified using the Social Vulnerability Index (SVI) tool as defined by the U.S. Census Bureau. An SVI score for each patient was obtained using home zip code and address to determine the U.S. Census SVI score for the address. Patients were then grouped into quartiles depending on the level of social vulnerability of the county in which they live. Quartiles were predetermined by using the CDC/ATSDR levels of severity, (Low severity = 0-0.2500, Low-Medium severity = 0.2501-0.5000, Medium-High severity = 0.5001-0.7500, and High severity = 0.7501-1.000).¹⁰ Each vulnerability category was evaluated to determine the percentage of each of the total outcomes within each category. Kaplan-Meier survival estimates one year after TAVR were calculated within each quartile of vulnerability.

Geographic location and/or vulnerability may also affect patient evaluation and workup. To determine if rural patients or socially disadvantaged patients suffered disparities in the time to procedure, cardiac catheterization (CATH) to TAVR and computed tomography angiography (CTA) to TAVR times were calculated and outcomes assessed, along with Kaplan-Meier survival estimates at one year. Average wait times between CATH and TAVR and CTA and TAVR were calculated for patients in each social vulnerability and distance traveled category and were subsequently compared to the other categories (Tables 1 and 2).

Table 1.

Effect of Social Vulnerability on Duration of Workup.

Average Time (in days) Between Preoperative Workup and TAVR per SVI Category
CATH and TAVR	Mean	Standard deviation	25% Q1	50% median	75% Q3
Low	70.99	112.42	24.0	41.0	69.5
Low-medium	71.67	111.67	24.0	39.0	64.0
Medium-high*	79.27	107.45	25.0	41.0	65.0
High	77.64	107.69	27.0	43.0	74.5
CTA and TAVR	Mean	Standard deviation	25% Q1	50% median	75% Q3
Low	39.75	87.17	10.0	20.0	36.0
Low-medium	33.82	66.77	9.0	18.0	34.0
Medium-high**	19.2	22.77	7.0	14.0	28.0
High	38.39	72.70	8.0	20.0	36.0

P < .0001; *P = 0.0127; **P = 0.0056.

Table 2.

Effect of Distance on Duration of Workup.

Average Time (in days) Between Pre-operative Work & TAVR per Distance Traveled (in Miles) Category
CATH and TAVR	Mean	Standard Deviation	0% Min	50% Median	100% Max
x < 14.9	68.69	112.53	1.0	37.0	846.0
15-<49.9	66.59	120.62	1.0	34.0	1128.0
50-<99.9	73.46	122.37	1.0	38.0	1434.0
100-<149.9	70.60	103.68	4.0	41.0	1028.0
x > 150	77.57	96.85	2.0	46.0	802.0
CTA and TAVR	Mean	Standard deviation	0% Min	50% median	100% Max
x < 14.9	39.39	88.43	1.0	19.0	1022.0
15-<49.9	26.83	34.13	1.0	16.0	297.0
50-<99.9	36.93	97.10	1.0	17.0	1427.0
100-<149.9	33.88	55.61	2.0	18.0	475.0
x > 150	41.51	62.83	1.0	23.0	504.0

P < .0001.

Mean and median values were computed for continuous variables and frequency distribution calculated for categorical values. Descriptive statistical analysis was used for demographic and clinical variables. Wilcoxon signed-rank test was used for non-normally distributed variables and Student’s t-test used for normally distributed continuous variables. Chi-square or Fisher’s exact test was used for categorical variables. Odds ratios and 95% confidence intervals were estimated using multivariate logistic regression. Statistics were performed using SAS software (SAS Institute, Cary, N.C. version 9.4). All statistical tests were two tailed with P < 0.05 considered significant.

Results

All patients undergoing TAVR since the program initiation to June 30, 2023, were retrospectively reviewed (n = 1565). Patients were predominantly male (58%) and white (98%). The average Society of Thoracic Surgeons risk score was 4.18%. Preoperative clinical characteristics and comorbidities are detailed in Table 3. Patients had similar wait times between CTA and TAVR regardless of their vulnerability status. Wait times between CATH and TAVR were ∼7 days longer in patients from the highest 2 vulnerability categories compared to patients in the lowest 2 vulnerability categories.

Table 3.

Patient Demographics.

Demographics (n = 1565)
Age
Mean (SD)	79 (8.6)
Gender
Male	903 (57.7%)
Female	662 (42.3%)
Race
White	1535 (98.0%)
Other	26 (1.7%)
Declined to state	4 (0.3%)
STS risk score (SD)	4.18% (0.03%)
Comorbidities	Had comorbidity	Did not have comorbidity
Current tobacco use (missing 425)	102 (8.9%)	1038 (91.1%)
Tobacco use history	918 (58.7%)	647 (41.3%)
Smoking history	901 (57.6%)	664 (42.4%)
Smokeless tobacco history (missing 5)	124 (7.9%)	1436 (92.1%)
Diabetes	548 (35.0%)	1017 (65.0%)
Chronic kidney disease	435 (27.8%)	1130 (72.2%)
Dialysis	47 (3.0%)	1518 (97.0%)
Hypertension	1344 (85.9%)	221 (14.1%)
Hyperlipidemia	1231 (78.7%)	334 (21.3%)
Hypercholesterolemia	326 (20.8%)	1239 (79.2%)
Chronic obstructive pulmonary disease	270 (17.3%)	1295 (82.7%)
Coronary artery disease	1063 (67.9%)	502 (32.1%)
Congestive heart failure	639 (40.8%)	926 (59.2%)
Cerebrovascular accident	228 (14.6%)	1337 (85.4%)
Peripheral vascular disease	201 (12.8%)	1364 (87.2%)

Differences in preoperative evaluation times did exist in patients who lived remotely. The time from CATH to implant and CTA to implant was significantly different depending on the patient’s home location (Table 2). Patients who lived furthest away from the implant site had a prolonged time for pre-operative evaluation. The average time between CATH and TAVR in patients who lived the furthest away was ∼9 days more than the patients who lived the closest to the implant site (69 days in closest vs 78 days in furthest away; P < .0001). Patients had similar wait times between CTA and TAVR regardless of the distance traveled (P < .0001).

There were no differences in clinical outcomes after TAVR in patients living remotely from the implantation site or in patients living in counties with a higher social vulnerability index (Table 4). Patients living closest to the implant center had nearly twice as many myocardial infarctions compared to those who lived furthest from the implant site (80 vs 36; P < .0002).

Table 4.

Clinical Outcomes: Effect of Social Vulnerability and Distance.

	Social Vulnerability Index
Outcome	Low (n = 732)	Low-Medium (n = 642)	Medium-High (n = 15)	High (n = 176)
MI (.14)	132 (18.0%)	96 (15.0%)	0 (0%)	27 (15.3%)
Stroke (.33)	37 (5.1%)	33 (5.1%)	2 (13.3%)	13 (7.4%)
New Afib/flutter (.78)	98 (13.4%)	77 (12.0%)	1 (6.7%)	22 (12.5%)
TIA/Delirium (.28)	29 (4.0%)	17 2.6%)	0 (0%)	3 (1.7%)
New pacemaker (.43)	82 (11.2%)	68 (10.6%)	1 (6.7%)	26 (14.8%)
Average days between CATH and TAVR	71	72	79	78
Average days between CTA and TAVR	40	34	19	38
	Distance traveled (in miles)
Outcome	< 14.9 (n = 348)	15 - < 49.9 (n = 141)	50 - < 99.9 (n = 448)	100 - < 149.9 (n = 298)	> 150 (n = 330)
MI (<.00)	80 (22.9%)	28 (19.9%)	71 (15.8%)	40 (13.4%)	36 (10.9%)
Stroke (.35)	19 (5.5%)	12 (8.5%)	20 (4.5%)	19 (6.4%)	15 (4.5%)
New Afib/flutter (.81)	48 (13.8%)	20 (14.2%)	50 (11.2%)	38 (12.8%)	42 (12.7%)
TIA/Delirium (.72)	14 (4.0%)	4 (2.8%)	12 (2.6%)	11 (3.7%)	8 (2.4%)
New pacemaker (.99)	39 (11.2%)	16 (11.3%)	51 (11.4%)	36 (12.1%)	35 (10.6%)
Average days between CATH and TAVR	69	67	74	71	78
Average days between CTA and TAVR	39	27	37	34	42

Ranges for number of days between preoperative workup and TAVR were examined to see if delays affected survival outcomes. (Ranges: x = 1-<24.5 days, x = 24.5-<41 days, x = 41-<67 days, x = 67-1434 days). Long term survival probability at 10 years was lowest for patients who waited the longest times (upper 2 categories) between CATH and TAVR compared to patients who waited the shortest times between CATH and TAVR (lower 2 categories) (P = 0.0060) (Figure 2). Long-term survival probability (10 years) across all social vulnerability categories was approximately 40% for all groups (P = 0.7219) (Figure 3). Survival probability between distance traveled categories at 10 years was highest for patients who lived the furthest away from implant site (P = 0.0010) (Figure 4). There were no differences in medium term survival (5 years) for patients in different SVI groups or distance.

Discussion

Geographic Remoteness

On average, each patient had a 67-78 day waiting period between cardiac catheterization and their TAVR procedure, but patients who lived most remotely had significantly longer preoperative evaluation times. Patients who lived furthest compared to those who lived closest had an average 9 day longer wait between CATH and TAVR. Time between CTA and TAVR was similar between all distance categories, with each patient having an average wait of 27-42 days, not significantly different. The similarity between time to cardiac catheterization and TAVR even for remote patients is likely due to performance of the catheterization at the remote hospital if catheterization laboratory facilities were available. Coordination and communication with local providers is obviously critical to reduce evaluation duration for remote patients.

Short-term outcomes were consistent across distance categories, except for myocardial infarctions. Within the first year post-TAVR myocardial infarction were nearly twice as common for patients who lived closest to the implant center. While unclear, it is plausible that patients who lived closest to the implantation site had greater health care accessibility and easier access to cardiac specialists, potentially leading to timelier diagnoses compared to those with fewer specialized resources.

Delays and differences in time of workup to implant did negatively impact survival. The Kaplan Meier survival estimates for patients in the upper and lower ‘days between CATH and TAVR’ categories demonstrate that those with the longest wait times had significantly lower survival compared with patients with the most rapid pre-procedural evaluation (Figure 2). This finding emphasizes the importance of timely pre-procedural evaluation. Measures to reduce the time of workup such as obtaining cardiac catheterization or CTA close to home (when possible) become especially important. One means to accomplish this is to have established referral patterns between remote sites and the implant center, ensuring timely testing and close communication. In our wide-spread referral area, telemedicine as well as simple electronic communication are widely used and help keep workup times low, ultimately benefiting patients.

Kaplan-Meier survival estimates up to 5 years after TAVR were not different for patients living remotely (Figure 4). There was a statistically significant difference in the probability of survival 10 years post-implantation, with survival being higher in patients who lived the furthest from the implantation site (Figure 4). The reason for this long-term difference is unclear but is unlikely to be due to the TAVR event.

Social Vulnerability

Average time between preoperative workup and TAVR implantation was also examined in each patient based on their social vulnerability. We found that patient vulnerability status did not affect pre-procedural evaluation times. This is in contradistinction to geographic distance, which did affect workup time. Patients who had the highest vulnerability had a 7 day longer wait time between CATH and TAVR compared to patients with the lowest vulnerability. When examining time between CTA and TAVR, patients with the highest vulnerability had a shorter wait time than patients who had the lowest vulnerability by 1 day. Potentially this may be due to the relatively rural setting of our referral area and the fact that rural remoteness may not positively correlate with vulnerability. For these patients, remoteness was far more important to preoperative evaluation than vulnerability. This may be unique to rural settings, where geography limits access to tertiary care. In large urban areas, where all patients are geographically proximate to care but social vulnerability may vary, SVI may be a more important determinate of health due to the many differing factors affecting vulnerability, not just distance from health care services alone. Social vulnerability may be underestimated in our study however as those individuals living with aortic stenosis who do not have access to care are socially vulnerable but without diagnosis, never are considered for TAVR. When diagnosed though post-TAVR outcomes were similar across all categories of SVI.

There were no significant differences in survival at both 5 and 10 years in all vulnerability categories, with survival probability being approximately 40% between all the vulnerability groups at 10 years post-TAVR. These data do not negate the known negative impact of vulnerability on wellness or survival, but rather emphasize that patients with severe aortic stenosis need to undergo TAVR in an expeditious manner and that time, rather than vulnerability matters most. In our rural region, distance and time spent evaluating the patient before TAVR have the most impact on outcomes and survival.

Patient wait times were impacted between preoperative workup and TAVR across social vulnerability categories. Wait times between CATH and TAVR were ∼7 days longer in patients from the highest 2 vulnerability categories compared to patients in the lowest 2 vulnerability categories. Wait times were similar between CTA and TAVR regardless of vulnerability status. Wait times between CATH and TAVR were longer in patients traveling 150+ miles compared to those traveling <15 miles, but similar between CTA and TAVR across all distance categories. Long term survival probability at 10 years was lowest for patients who waited the longest times (upper 2 categories) between CATH and TAVR compared to patients who waited the shortest times (lower 2 categories). Clinical outcomes were similar across all distance and vulnerability categories except myocardial infarction, which occurred ∼2x more in patients who lived closest to the implantation site compared to those who lived furthest away. There was no difference in survival probability at 10 years across all vulnerability groups, with survival probability being ∼40%. Across distance categories there was no difference in survival probability at 5 years, but survival is highest in patients who live furthest away at 10 years. In summary, patients who have a lower SVI and who live remotely can undergo TAVR safely, although challenges remain for timely pre-procedural evaluation.

Limitations

Our study was retrospective and single institution with recognized limitations, but was sizable, real-world, and inclusive of all patients over an 11-year period. We defined vulnerability by patient home location using validated census data, but individual patient vulnerability was not considered, possibly diminishing the impact of vulnerability because of our definition.¹⁰ Patient vulnerability and distance traveled may also have been impacted by using the specific home address we have for the patient on file. This could be due to some patients having addresses in different areas across the United States rather than in the upper Midwest like most of the patients in this study, which could impact social vulnerability and distance. The specific cause of delays in workup is unclear; patient delays between preoperative workup and TAVR could be due to different, individual circumstances such as patients missing a scheduled visit or losing contact entirely and may not be due to remoteness. Nonetheless, distance from implant center significantly increased evaluation times.

Ethics Approval

This retrospective chart review study involving human participants was in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. The Human Investigation Committee (IRB) of Sanford Health Fargo approved this study.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Jacob M. Tupa

References

Cribier

Eltchaninoff

Bash

, et al. Percutaneous transcatheter implantation of an aortic valve prosthesis for calcific aortic stenosis: First human case description. Circulation. 2002;106(24):3006-3008. doi:10.1161/01.cir.0000047200.36165.b8

Kundi

Faridi

Wang

, et al. Geographic patterns of growth for transcatheter aortic valve replacement in the United States. Circulation. 2019;140(11):969-971. doi:10.1161/CIRCULATIONAHA.119.040788

Douthit

Kiv

Dwolatzky

Biswas

. Exposing some important barriers to health care access in the rural USA. Public Health. 2015;129(6):611-620. doi:10.1016/j.puhe.2015.04.001

Brundisini

Giacomini

DeJean

Vanstone

Winsor

Smith

. Chronic disease patients' experiences with accessing health care in rural and remote areas: A systematic review and qualitative meta-synthesis. Ont Health Technol Assess Ser. 2013;13(15):1-33. Published 2013 Sep 1.

Valet

Perry

Hartert

. Rural health disparities in asthma care and outcomes. J Allergy Clin Immunol. 2009;123(6):1220-1225. doi:10.1016/j.jaci.2008.12.1131

Marquis-Gravel

Stebbins

Kosinski

, et al. Geographic access to transcatheter aortic valve replacement centers in the United States: insights from the society of thoracic surgeons/American college of cardiology transcatheter valve therapy registry. JAMA Cardiol. 2020;5(9):1006-1010. doi:10.1001/jamacardio.2020.1725

Jain

Khan

, et al. Association between social vulnerability index and cardiovascular disease: a behavioral risk factor surveillance system study. J Am Heart Assoc. 2022;11(15):e024414. doi:10.1161/JAHA.121.024414

Khan

Javed

Lone

, et al. Social vulnerability and premature cardiovascular mortality among US counties, 2014 to 2018. Circulation. 2021;144(16):1272-1279. doi:10.1161/CIRCULATIONAHA.121.054516

Roberts

Gauthier

Teigen

, et al. Underutilization of transcatheter aortic valve replacement in northern plains American Indians with severe aortic stenosis. J Racial Ethn Health Disparities. 2023;1:1. doi:10.1007/s40615-023-01604-7

10.

Centers for Disease Control and Prevention/Agency for Toxic Substances and Disease Registry/Geospatial Research, Analysis, and Services Program . CDC/ATSDR Social Vulnerability Index 2020. Available at: https://www.atsdr.cdc.gov/placeandhealth/svi/data_documentation_download.html

11.

Généreux

Piazza

. Valve academic research Consortium 3: Updated endpoint definitions for aortic valve clinical research. J Am Coll Cardiol. 2021;77(21):2717–2746. doi:10.1016/j.jacc.2021.02.038