https://orcid.org/0009-0002-3765-0923
Figures
Abstract
Purpose
Care fragmentation (CF), defined as readmission to a non-index facility, has historically been linked with adverse clinical and financial outcomes for a variety of cardiac procedures. However, its impact on patients undergoing transcatheter aortic valve replacement (TAVR) remains understudied.
Methods
This retrospective study investigated the prevalence and impact of CF on outcomes of patients undergoing TAVR. All adults (≥18 years) undergoing an isolated TAVR procedure, who were readmitted within 90 days, were tabulated using the Nationwide Readmissions Database from 2016 to 2021. A 90-day readmission window was chosen in accordance with prior literature on readmission. Patients treated at a non-index facility were categorized into the CF cohort. Multivariable logistic models were developed to characterize the association of care fragmentation status with perioperative complications and resource utilization; candidate variables were selected using elastic net regularization to minimize variable collinearity.
Results
Of an estimated 55,891 patients who were readmitted within 90 days, 37.4% were treated at a non-index facility. While TAVR utilization more than doubled, both the readmission and care fragmentation rates fell during the study period. Following adjustment, CF status was associated with respiratory (AOR [Adjusted Odds Ratio] 1.20, 95% CI [Confidence Interval] 1.11–1.29, p < 0.001), gastrointestinal (AOR 1.21, 95% CI 1.05–1.38, p = 0.008), and infectious complications (AOR 1.17, 95% CI 1.10–1.25, p < 0.001) at readmission. Furthermore, CF was linked with greater risk of non-home discharge (AOR: 1.13 95% CI [1.03–1.25], P = 0.01).
Citation: Chaturvedi A, Mallick S, Le NK, Tabibian K, Balian J, Yalzadeh D, et al. (2026) Association of care fragmentation with financial and clinical outcomes of transcatheter aortic valve replacement in the US: A retrospective cohort study. PLoS One 21(7): e0353361. https://doi.org/10.1371/journal.pone.0353361
Editor: Kuo-Cherh Huang, Taipei Medical University, TAIWAN
Received: March 7, 2025; Accepted: June 23, 2026; Published: July 21, 2026
Copyright: © 2026 Chaturvedi et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The database and its content can be obtained via the link below after completion of a data use agreement with the United States Agency for Healthcare Research and Quality: https://hcup-us.ahrq.gov/tech_assist/centdist.jsp.
Funding: The author(s) received no specific funding for this work.
Competing interests: I have read the journal’s policy and the authors of this manuscript have the following competing interests: author Peyman Benharash received proctoring fees from AtriCure, and this study does not analyze product or services from AtriCure. This does not alter our adherence to PLOS ONE policies on sharing data and materials.
Introduction
Transcatheter aortic valve replacement (TAVR) has become widely adopted as an effective treatment strategy for severe symptomatic aortic stenosis, with its indications continuing to expand [1]. In fact, several recent clinical trials have demonstrated comparable short- and long-term outcomes following surgical aortic valve replacement (SAVR) and TAVR across strata of surgical risk [2,3]. These factors, alongside the expansion of TAVR eligibility and increased operator experience, have contributed to a nearly ten-fold increase in procedural volume between 2010 and 2023 [1,4]. Despite such advances including expedited discharge, TAVR readmission rates remain high, with more than 10% of patients being rehospitalized within 30 days [5,6]. Additionally, Kolte and colleagues found that patients readmitted within 30 days following TAVR experienced higher rates of acute kidney injury, major bleeding, and cardiac arrest, compared to others [7].
Care fragmentation (CF), defined as readmission to a non-index facility, may further contribute to the poor clinical and financial outcomes among rehospitalized patients [8]. Prior investigation has cited a multitude of factors including inadequate medical history, variations in center quality, and limited data sharing as potential mechanisms underlying worse outcomes in CF [9]. Existing literature has highlighted the increased healthcare expenditures and inferior clinical outcomes associated with CF across multiple procedural contexts [10]. However, implications of care fragmentation in cardiac populations, remain insufficiently characterized [10,11]. In a study of approximately 9,000 TAVR patients between 2011–2015, Wang and colleagues found that over 5,000 patients experienced care fragmentation and reported CF to be linked with increased mortality and all-cause readmissions at one year [12–17]. However, while technological advancements and increased data sharing were associated with partially mitigating CF, their study relied on decade-old data and was not nationally representative. Given the significant rise in TAVR volume across the US, a contemporary national analysis of the incidence and impact of CF following TAVR procedures is warranted.
In the present work, we examined the association of CF with readmission outcomes after hospitalization in a contemporary cohort of TAVR patients. We hypothesized CF to be associated with greater risk of complications at readmission along with similar resource utilization.
Methods
This was a retrospective cohort study of the 2016–2021 Nationwide Readmissions Database (NRD). Aggregated from 28 state inpatient databases, the NRD is a nationally representative all-payer database and provides accurate estimates for ~60% of all hospitalizations in the United States. Each record in the NRD is given a unique identifier, allowing for subsequent readmissions to be linked within each state and calendar year.
All adult (≥18 years) hospitalizations for isolated (TAVR) were tabulated in the 2016−2021 Nationwide Readmissions Database (NRD) using previously reported International Classification of Diseases Tenth Revision procedure codes (ICD-10) [1]. While only patients requiring nonelective rehospitalization within 90 days of index discharge were considered, those readmitted to a non-index facility comprised the CF cohort (others non-CF). Our cohort included only patients undergoing first-time TAVR procedures, excluding those who received valve-in-valve procedures due to their distinct risk profile and prior surgical history. Records with concurrent cardiac procedures or missing key data as well as discharges after September 30th of each year, were not considered (Fig 1).
All estimates are based on survey-weighted methodology. Key data include age, sex, costs, in-hospital mortality, elective status, and income. CF, care fragmentation.
Patient demographics and hospital characteristics including age, sex, income quartile, primary payer, and hospital teaching status were defined in accordance with the Healthcare Cost and Utilization Project dictionary data [18]. Hospitalization costs were estimated by applying HCUP center-specific cost-to-charge ratios to total charges and inflating all values to 2021 dollars using the Personal Health Care Price Index. [19]. Facility-wide cost-to-charge ratios were applied uniformly across payer categories, preventing adjustment for payer-specific reimbursement differences and potentially introducing nondifferential measurement error in estimated hospitalization expenditures. Hospitals were ranked according to the annual institutional TAVR caseload with centers in the top quartile classified as High-Volume Centers (HVH, Others: Low Volume-Centers (LVH)). Non-home discharge was defined as discharge to an acute care hospital, intermediate care facility, or skilled nursing facility.
The previously validated Comorbid Operative Risk Score (CORE score) was employed to numerically capture the burden of chronic conditions [20]. The CORE score employs machine learning algorithms to generate a mortality prediction score based on comorbidities [20]. The presence of specific conditions such as coronary artery disease, chronic pulmonary disease, and coagulopathy was ascertained using previously validated ICD-10 codes [21]. Other comorbidities including obesity, defined as a body mass index greater than 30, and smoking, identified by documentation of nicotine dependence or tobacco use, were similarly captured using validated ICD-10 classifications [21]. Cardiac (myocardial infarction or cardiac arrest), neurological (stroke or transient ischemic attack), infectious (sepsis or deep wound infection), gastrointestinal (bile leak, gastrointestinal bleeding, intestinal perforation, hemoperitoneum, or abdominal fluid collection), thromboembolic (deep vein thrombosis or pulmonary embolism), and respiratory (pneumothorax, acute respiratory distress syndrome, acute respiratory failure, prolonged ventilation >96 hours, pneumonia, or pleural effusion) complications, were similarly identified using previously published ICD-10 codes [21]. Diagnosis related group codes were used to tabulate the primary reason for readmission.
The primary outcome of interest was the development of any in-hospital complication (listed above) at readmission. Secondary endpoints included hospitalization costs, postoperative length of stay, and non-home discharge upon readmission.
An exploratory, post hoc subgroup analysis was performed to assess whether the relationship between care fragmentation and outcomes varies based on institutional procedural experience, given the established volume-outcome relationship in TAVR. Patients within the CF-cohort were stratified into four subgroups based on operative volume of their index hospitalization and readmission. Patients receiving TAVR and subsequently readmitted to high-volume centers comprised the high-high cohort (HH) while those only treated at low volume centers comprised the low-low cohort (LL). Patients initially receiving TAVR at a HVH and then were readmitted to a LVH were grouped as high-low (HL) while those with an index admission at a LVH with subsequent readmission to a HVH comprised the low-high cohort (LH).
Continuous variables are summarized as medians with interquartile range (IQR) or means with standard deviations (SD), while categorical variables are reported as proportions (%). The Adjusted Wald, chi-square, and Mann-Whitney U tests were used to compare patient and hospital characteristics where applicable. Cuzick’s nonparametric test was applied to determine the significance of temporal trends (nptrend) [22]. Multivariable logistic regression and generalized linear models were fit to evaluate independent associations between CF and study endpoints. Covariates were selected using elastic net regularization, which minimizes collinearity through a penalized least squares methodology [23]. Model covariates included both patient-level and hospital-level factors. Model performance was evaluated using receiver operating characteristics and calibrations belts, as appropriate. Interaction terms were incorporated into risk-adjusted models to evaluate the effect of institutional TAVR volumes at index admission and readmission in the CF only sub analysis. To contextualize the potential for type II error, we calculated the minimum detectable effect; under a two-sided α = 0.05, 95% power, and 1:3 group allocation, we found 17,400 patients would be required, which is well below our realized cohort. Regression outputs are reported as adjusted odds ratio (AOR) or beta coefficients (β) for logistic and linear models with 95% confidence intervals (CI), respectively. An α of 0.05 was set for statistical significance. All statistical analysis was performed on Stata 16.1 software (StataCorp, LLC, College Station, TX). This study was deemed exempt from full review by the Institutional Review Board at University California, Los Angeles.
Results
Of an estimated 300,842 TAVR hospitalizations performed during the study period, 55,891 (17.6%) entailed a nonelective readmission within 90 days of index discharge, 37.4% of which (n = 20,903) were at a non-index facility (CF). The total number of TAVR procedures more than doubled during the study period (32,216 procedures in 2016 vs 69,398 in 2021, nptrend<0.001; Fig 2). Additionally, rates of both 90-day readmissions (20.7 in 2016 vs 15.7% in 2021, P < 0.01), and care fragmentation rates (41.4 in 2016 vs 35.3% in 2021, P = 0.01) decreased significantly (Fig 2).
The rate care fragmentation decreased significantly over the study period, nptrend < 0.001.
Subjects in the CF cohort were slightly older (80.0 ± 8.5 vs 79.0 ± 8.4 years, P < 0.01) and more likely to be female (46.5 vs 45.2%, P < 0.01; Table 1), compared to others. The CF group also presented with a higher median CORE score (44 [30–59] vs 43 units [28–58], P < 0.01) and greater frequency of coronary artery disease (64.4 vs 63.3%, P < 0.01), coagulopathy (14.2 vs 12.7%, P < 0.01), and smoking (40.4 vs 38.2%, P < 0.01; Table 1). The groups were comparable in regards to insurance coverage and income distribution (Table 1).
After risk adjustment, several factors were associated with greater odds of CF, including an increasing CORE score (AOR: 1.12 per unit 95% CI [1.08–1.15], P < 0.01) and age (AOR: 1.01 per year 95% CI [1.01–1.01], P < 0.01). Additionally, higher center volume (AOR: 0.99 per case 95% CI [0.98–0.99], P < 0.01) at index admission was associated with reduced odds of care fragmentation. However, income and primary payer status did not appear to alter the risk of CF (Table 2).
On unadjusted analysis, CF patients experienced greater rates of neurological (7.4 vs 6.0%, P < 0.01), respiratory (17.9 vs 16.0%, P < 0.01), gastrointestinal (4.1 vs 3.5%, P < 0.001), and infectious (22.6 vs 19.4%, P < 0.001) complications upon readmission, compared to their non-CF counterparts (Table 3). However, CF patients exhibited reduced rates of cardiac (10.8 vs 12.3%, P < 0.01) and thromboembolic (2.3 vs 2.7%, P = 0.05) complications. Additionally, the CF cohort had longer postoperative hospital stays at index admission (3.8 ± 4.9 vs 3.7 days ± 4.9, P = 0.004) as well as greater rates of nonhome discharge (25.5 vs 21.5%, P < 0.01), compared to the non-CF cohort. Hospitalization costs were similar between the groups as shown in Table 2.
After risk-adjustment, care fragmentation remained independently associated with increased odds of respiratory (AOR: 1.20 95% CI [1.11–1.29], P < 0.01), gastrointestinal (AOR: 1.21 95% CI [1.05–1.38], P = 0.01), and infectious (AOR: 1.17 95% CI [1.10–1.25], P < 0.01) complications during readmission (Table 4).
Additionally, CF was linked to higher odds of nonhome discharge (AOR: 1.13 95% CI [1.03–1.25], P = 0.01). However, CF appeared not to be significantly associated with LOS (β: −0.09 days 95% CI [−0.21 – + 0.04], P = 0.16) or aggregate inpatient facility costs captured in the NRD (β: -$327 95% CI [−1240 – + 35], P = 0.48; Fig 3).
An additional analysis was performed on elderly (≥80 years) TAVR recipients to assess the impact of care fragmentation within a more homogenous cohort. Following risk adjustment with non-CF as reference, care fragmentation was independently associated with greater odds of respiratory (AOR: 1.24, 95% CI [1.10–1.42], P < 0.01), infectious (AOR: 1.19, 95% CI [1.09–1.30], P < 0.01), and gastrointestinal complications (AOR: 1.16, 95% CI [1.06–1.39], P < 0.01) during readmission. Moreover, CF was linked to higher rates of nonhome discharge, but comparable hospitalization costs and LOS compared to their non-CF counterparts (Table 5).
A subgroup analysis was performed to examine the association of hospital volume on index admission and subsequent non-index hospitalization for the CF cohort. Relative to the HH cohort, LL was associated with increased odds of respiratory (AOR: 1.31, 95% CI [1.01–1.71], P = 0.04), gastrointestinal (AOR: 1.67, 95% CI [1.23–2.70], P < 0.01), and infectious (AOR: 1.23, 95% CI [1.09–1.52], P = 0.01) complications upon readmission (Table 6). The HL and LH cohorts noted no differences in developing perioperative complications compared to their HH counterparts. Additionally, the LL cohort experienced prolonged LOS (β: + 2.75 days, 95% CI [+0.42 – + 5.07], P = 0.02) while both the LL (β: + $7,787, 95% CI [+4,298 – + 11,275], P < 0.01) and HL (β: + $5,895, 95% CI [+2,945 – + 8,846], P < 0.01) cohorts were associated with greater hospital expenditure compared to their HH counterparts. Non-home discharge rates remained similar across all cohorts. However, formal interaction testing between index and readmission hospital volumes revealed no significant interaction effect on either financial or clinical outcomes.
Discussion
Care fragmentation has been associated with inferior clinical outcomes across several medical conditions but remains understudied in TAVR [10,11]. In this national analysis, TAVR utilization more than doubled between 2016 and 2021, while both care fragmentation and readmission rates declined. Older age and higher comorbidity burden were associated with increased odds of fragmented care, whereas treatment at high-volume centers was associated with lower fragmentation rates. Fragmented care was associated with increased odds of respiratory, gastrointestinal, and infectious complications during readmission. Because interaction testing between index and readmission hospital volume was not significant, volume-stratified findings should be interpreted cautiously and considered hypothesis-generating.
Congruent with previous literature, we found the 90-day TAVR readmission rates to significantly decline between 2016 and 2021 [24]. This observation, in part, may be explained through advances in clinician experience, better patient selection, and use of lower profile TAVR platforms, obviating the need to surgical vascular access and general anesthesia [1,25,26]. An additional mechanism for the decline in TAVR readmissions may be the utilization of TAVR in lower risk patients as described by Waksman and colleagues [3,27]. Despite these advances, patients undergoing TAVR continue to experience higher rates of care fragmentation and non-elective readmissions relative to those receiving other transcatheter procedures, such as mitral valve replacement [28]. Van Belle et al. reported TAVR patients to be more likely to be readmitted due to a higher prevalence of undiagnosed transthyretin cardiac amyloidosis, persistent left ventricular hypertrophy, and supraventricular arrhythmias [29]. Emerging strategies such as post-discharge remote monitoring may further improve early detection of complications following TAVR [30].
Fragmented care was associated with increased odds of gastrointestinal, infectious, and respiratory complications during readmission. This association persisted in a subgroup analysis of TAVR recipients over the age of 80, with those receiving fragmented care being associated with greater odds of perioperative complications [31]. Despite the modest magnitude of the observed effect, the consistency of this association across a range of complications may reflect the cumulative impact of care discontinuity. Ineffective care coordination among providers is a key contributor to care fragmentation [8]. Examined primarily within the transition from inpatient to outpatient settings, inadequate inter-provider communication and limited clinical data sharing have been estimated to account for a significant portion of adverse events during the intermediate post-discharge period [9]. TAVR patients are especially susceptible to experiencing fragmented care due to their complex health profiles and advanced age [23]. One potential avenue for reducing readmission and care fragmentation rates involves developing a TAVR care coordinator program. A 2022 pilot study of three German hospitals found the implementation of a TAVR care coordinator to be associated with a reduction in 30-day all cause readmissions, improved patient satisfaction, and stronger care continuity between providers [32]. The addition of a care coordinator may help to shorten the treatment pathway while facilitating better decision-making and streamlined care at non-index hospitals [33]. The use of health information technology interoperability has been further associated with a significant reduction in readmissions and care fragmentation. Li and colleagues found implementing health information technology interoperability led to a reduction in all cause readmissions, throughput time, and interhospital transfers for patients with acute myocardial infarction in New York State [34]. Increasing the utilization of health information interoperability for TAVR patients and incentivizing a widespread exchange of clinical data could improve the efficiency of healthcare delivery while reducing readmissions and hospital resource utilization.
Patients receiving care at high-volume centers demonstrated lower rates of care fragmentation. This effect was also observed even when examining CF patients alone, with those receiving care at low-volume hospitals demonstrating higher rates of complications compared to their counterparts who received the entirety of their care at high volume centers. Despite these directional differences, formal interaction testing between index and readmission hospital volume was not significant, possibly suggesting such associations may reflect residual variation as opposed to true effect modification by hospital volume. Higher volume centers are often speculated to have improved outcomes due to robust discharge planning and care coordination [35]. These trends have been previously observed in TAVR, with Vemulapalli and colleagues noting an inverse relationship between hospital volume and 30-day all-cause readmissions as well as in-hospital mortality for patients undergoing aortic valve replacement [36]. Ando et al. reported patients treated at non-metropolitan, non-teaching centers experienced more than a twofold increase in the likelihood of nonelective readmission compared to those at metropolitan, teaching institutions [37]. The link between metropolitan teaching hospitals, often synonymous with high-volume centers, and reduced odds of readmission may be in part explained by increased hospital staffing coupled with enhanced postoperative patient monitoring and management. Given the demonstrated benefits of treatment at tertiary hospitals, some have called for the centralization of care to high-volume centers in a variety of fields including TAVR [38]. While regionalization of treatment to high-volume TAVR centers may streamline care and reduce operative variability, its impact on increased travel distance and potential adverse consequences towards continuity of care cannot be ignored [39].
There are several important limitations to the present study. The NRD is a retrospective, administrative database subject to variations in coding practices between healthcare providers and institutions. Given the retrospective nature, causal relationships between care fragmentation and outcomes of interest cannot be established. The NRD captures readmissions to participating hospitals and within each calendar year, creating the potential for missed rehospitalizations. The NRD lacks certain granular data such as laboratory values, traditional operative risk assessments such as the Society of Thoracic Surgeons Predicted Risk of Mortality, as well as imaging studies, and aortic valve area and ventricular function. Accordingly, these findings should not be interpreted as evidence of overall episode-of-care cost neutrality, as professional fees and post-discharge expenditures are not captured. Detailed expenditure information, including specific drivers of cost such as ICU stays, radiology fees, and procedural expenses remain unavailable in the NRD, thus limiting comprehensive economic interpretation. Moreover, patient satisfaction in their care at the index hospital is not documented. The NRD does not have data regarding the number of hospitals in each region and does not permit geographic analysis. The NRD only captures inpatient admissions without outpatient encounters or the ability to link patients across multiple providers and health systems. In addition, while care fragmentation was treated as a binary outcome, it may be affected by unmeasured socioeconomic and health system-levels factors not fully captured in the NRD, including variables such as health literacy and community-level resources potentially biasing the observed association. Care fragmentation was operationalized as a binary indicator of readmission to a non-index facility, which may not fully capture the spectrum of multi-institutional care. Under nondifferential misclassification of a binary exposure, observed effect estimates are attenuated toward the null by a factor approximating (sensitivity + specificity − 1). If even a modest proportion of fragmented readmissions were misclassified as non-fragmented due to undetected inter-facility transfers or same-system readmissions, true associations could be meaningfully larger than those reported. Accordingly, observed outcome differences between fragmented and non-fragmented care likely represent conservative estimates of the true effect. Furthermore, the binary operationalization precludes assessment of potential dose-response relationships between the degree of multi-institutional care and readmission outcome severity, as the NRD does not encode the number of distinct facilities involved in a patient’s care pathway or the sequencing of inter-facility transitions. Hospital-level clustering was not explicitly modeled and may have resulted in underestimated variance and residual confounding from between-center heterogeneity. Furthermore, in-hospital mortality during readmission was not treated as a competing risk, such that early death may have limited complication accrual and led to underestimation of complication rates among the most critically ill patients. The subgroup analysis stratifying patients by index and readmission hospital volume may have been underpowered, limiting our ability to detect a true interaction between these variables. Furthermore, defining care fragmentation as a binary metric may oversimplify the complexity of multi-institutional care patterns, particularly in the NRD’s inability to capture outpatient encounters, which could result in the misclassification of fragmented care. Such nondifferential misclassification would be expected to bias effect estimates toward the null and may underestimate the true association between care fragmentation and outcomes. Moreover, a comprehensive sensitivity analysis of unmeasured socioeconomics was not feasible within the NRD due to it lacking individual-level data on income, education, and neighborhood deprivation. The NRD also does not include validated frailty assessments, prosthesis type, procedural access site, or operative complexity metrics. Discharge disposition at the index admission, which may influence readmission site selection, was not available as a model covariate. The absence of clinically relevant variables such as prosthesis type, procedural access site, operative complexity, and discharge disposition may contribute to possible residual confounding, which may not be fully resolved no elastic net covariate selection. Omission of these variables may contribute to residual confounding despite elastic net-guided covariate selection.“ Hospital volume strata was defined using sample-based quartiles, with differences in subgroup estimates possibly reflecting the influence of data-cutoffs. The NRD does not provide sufficient granularity to characterize care fragmentation beyond a binary classification of index versus non-index readmission, precluding evaluation of transfer chains, outpatient encounters, or transition timing between facilities. Consequently, future studies using longitudinal claims data or integrated health information networks are needed to determine whether increasing degrees of fragmentation are associated with progressively worse outcomes. Hospital-level clustering was not formally addressed through multilevel modeling, which may result in underestimation of standard errors and contribute to residual confounding from between-center heterogeneity. Additionally, in-hospital mortality during readmission was not modeled as a competing risk; early death may preclude complication accrual, potentially underestimating true complication rates in the most critically ill patients. Likewise, the NRD does not capture outpatient encounter data, limiting the analysis of multi-institutional care patterns. Finally, the omission of valve-in-valve TAVR procedures may bias and limit the generalizability of these finds to the broader TAVR population. Nonetheless, we used robust statistical methods and the largest all-payer dataset in the United States aro reduce the risk of bias and enhanced the generalizability of our findings.
In summary, care fragmentation appears to be linked with inferior clinical outcomes in patients undergoing TAVR. Notably within the care fragmented cohort, treatment at low-volume TAVR centers was associated with greater resource utilization and complication rates. Given the increasing utilization of TAVR nationally, further exploration of efficient information sharing, care coordination, and health equity policy is warranted to reduce the burden of care fragmentation.
References
- 1. Gandjian M, Verma A, Tran Z, Sanaiha Y, Downey P, Shemin RJ, et al. Influence of center surgical aortic valve volume on outcomes of transcatheter aortic valve replacement. JTCVS Open. 2022;11:62–71. pmid:36172405
- 2. Smith CR, Leon MB, Mack MJ, Miller DC, Moses JW, Svensson LG, et al. Transcatheter versus surgical aortic-valve replacement in high-risk patients. N Engl J Med. 2011;364(23):2187–98. pmid:21639811
- 3. Mack MJ, Leon MB, Thourani VH, Pibarot P, Hahn RT, Genereux P, et al. Transcatheter aortic-valve replacement in low-risk patients at five years. N Engl J Med. 2023;389(21):1949–60. pmid:37874020
- 4. Vedantam K, Torres CA, Elmariah S, Villablanca P, Goessl M, Zajarias A, et al. Racial disparities among patients undergoing balloon-expandable transcatheter aortic valve replacement. J Soc Cardiovasc Angiogr Interv. 2025;4(2):102495. pmid:40109708
- 5. Zahid S, Din MTU, Khan MZ, Rai D, Ullah W, Sanchez-Nadales A, et al. Trends, predictors, and outcomes of 30-day readmission with heart failure after transcatheter aortic valve replacement: Insights from the US nationwide readmission database. J Am Heart Assoc. 2022;11(16):e024890. pmid:35929464
- 6. Hirji SA, Zogg CK, Vaduganathan M. Quantifying the impact of care fragmentation on outcomes after transcatheter aortic valve implantation. The American Journal of Cardiology. 2020;128:113–9.
- 7. Kolte D, Khera S, Sardar MR, Gheewala N, Gupta T, Chatterjee S, et al. Thirty-day readmissions after transcatheter aortic valve replacement in the United States: Insights from the nationwide readmissions database. Circ Cardiovasc Interv. 2017;10(1):e004472. pmid:28034845
- 8. Kim S, Hadaya J, Joachim K, Ali K, Mallick S, Cho NY, et al. Care fragmentation is associated with increased mortality after ileostomy creation. Surgery. 2024;175(4):1000–6. pmid:38161087
- 9. Sakowitz S, Madrigal J, Williamson C, Ebrahimian S, Richardson S, Ascandar N, et al. Care fragmentation after hospitalization for acute myocardial infarction. Am J Cardiol. 2023;187:131–7. pmid:36459736
- 10. Ali K, Coaston T, Ng AP, Micalo L, Sakowitz S, Vadlakonda A, et al. Care fragmentation following bariatric operations: A national analysis. Am Surg. 2025;91(5):756–63. pmid:39760557
- 11. Rashid Z, Munir MM, Woldesenbet S, Tsilimigras DI, Khalil M, Khan MMM, et al. Care fragmentation in hepatopancreatic surgery and postoperative outcomes. Surgery. 2024;175(6):1562–9. pmid:38565495
- 12. Wang A, Li Z, Rymer JA, Kosinski AS, Yerokun B, Cox ML, et al. Relation of postdischarge care fragmentation and outcomes in transcatheter aortic valve implantation from the STS/ACC TVT registry. Am J Cardiol. 2019;124(6):912–9. pmid:31375245
- 13. Fallahtafti P, Soleimani H, Ebrahimi P, Ghaseminejad-Raeini A, Karimi E, Shirinezhad A, et al. Comparative analysis of PCI strategies in aortic stenosis patients undergoing TAVI: A systematic review and network meta-analysis. Clin Cardiol. 2024;47(8):e24324. pmid:39054901
- 14. Fallahtafti P, Bahramrafiee R, Sattarzadeh Badkoubeh R. Revisiting echocardiographic ranges of left ventricular end-diastolic volume index: An analysis of the discrepancies between the 2006 and the 2015 recommendation for chamber quantification guidelines. Clinical Cardiology. 2024;47:e70003.
- 15. Azarboo A, Fallahtafti P, Jalali S, Shirinezhad A, Assempoor R, Ghaseminejad-Raeini A. Screening accuracy of Single-Point Insulin Sensitivity Estimator (SPISE) for metabolic syndrome: A systematic review and meta-analysis. BMC Endocr Disord. 2025;25(1):142. pmid:40481439
- 16. Tavakoli K, Hosseini Mohammadi NS, Fallahtafti P, Khamene SS, Taheri M, Ebrahimi P, et al. Sex-related differences in survival and safety outcomes after transcatheter aortic valve replacement: A meta-analysis of reconstructed time-to-event data. Eur Heart J Qual Care Clin Outcomes. 2025;11(5):642–53. pmid:40261266
- 17. Kazemian S, Fallahtafti P, Sharifi M, Mohammadi NSH, Soleimani H, Moghadam AS, et al. Trends in transcatheter versus surgical aortic valve replacement outcomes in patients with low-surgical risk: A systematic review and meta-analysis of randomized trials. J Am Heart Assoc. 2024;13(21):e036179. pmid:39424419
- 18. NRD Description of Data Elements, https://hcup-us.ahrq.gov/db/nation/nrd/nrddde.jsp. Accessed 2025 April 5.
- 19. Cost-to-charge ratio files. https://hcup-us.ahrq.gov/db/ccr/costtocharge.jsp. Accessed 2025 April 5.
- 20. Chervu NL, Balian J, Verma A. Development of a surgery-specific comorbidity score for use in administrative data. Ann Surg. 2024.
- 21. Sanaiha Y, Hadaya J, Cale M, Cho NY, Madrigal J, Sareh S, et al. Impact of interhospital transfer on clinical outcomes and resource use after cardiac operations: Insights from a national cohort. Surgery. 2020;168(5):876–81. pmid:32641276
- 22. Cuzick J. A Wilcoxon-type test for trend. Stat Med. 1985;4(1):87–90. pmid:3992076
- 23. Zou H, Hastie T. Regularization and variable selection via the elastic net. Journal of the Royal Statistical Society Series B: Statistical Methodology. 2005;67:301–20.
- 24. Tripathi B, Nerusu LA, Sawant AC, Atti L, Sharma P, Pershad A. Transcatheter aortic valve implantation readmissions in the current era (from the national readmission database). Am J Cardiol. 2020;130:115–22. pmid:32665132
- 25. Biancari F, Rosato S, D’Errigo P, Ranucci M, Onorati F, Barbanti M, et al. Immediate and intermediate outcome after transapical versus transfemoral transcatheter aortic valve replacement. Am J Cardiol. 2016;117(2):245–51. pmid:26639038
- 26. Ruparelia’’Neil, Prendergast’’Bernard David. Technical aspects of transcatheter aortic valve implantation (TAVI), https://www.escardio.org/Journals/E-Journal-of-Cardiology-Practice/Volume-14/technical-aspects-of-transcatheter-aortic-valve-implantation-tavi. Accessed 2025 April 17.
- 27. Waksman R, Corso PJ, Torguson R, Gordon P, Ehsan A, Wilson SR, et al. TAVR in low-risk patients: 1-year results from the LRT trial. JACC Cardiovasc Interv. 2019;12(10):901–7. pmid:30860059
- 28. Lima FV, Kolte D, Rofeberg V, Molino J, Zhang Z, Elmariah S, et al. Thirty-day readmissions after transcatheter versus surgical mitral valve repair in high-risk patients with mitral regurgitation: Analysis of the 2014-2015 Nationwide readmissions databases. Catheter Cardiovasc Interv. 2020;96(3):664–74. pmid:31868999
- 29. Van Belle E, Vincent F, Labreuche J. Balloon-expandable versus self-expanding transcatheter aortic valve replacement. Circulation. 2020;141:243–59.
- 30. Liu X, Fan J, Guo Y, Dai H, Xu J, Wang L, et al. Wearable smartwatch facilitated remote health management for patients undergoing transcatheter aortic valve replacement. J Am Heart Assoc. 2022;11(7):e023219. pmid:35347997
- 31. Oh S, Kim JH, Hwang C-H, Hyun DY, Cho KH, Kim MC, et al. Comparison of outcomes after transcatheter aortic valve replacement between elderly (65-79 years) and super-elderly (≥80 years) patients. Medicine (Baltimore). 2022;101(26):e29816. pmid:35777026
- 32. Bohmann K, Burgdorf C, Zeus T, Joner M, Alvarez H, Berning KL, et al. The COORDINATE Pilot Study: Impact of a transcatheter aortic valve coordinator program on hospital and patient outcomes. J Clin Med. 2022;11(5):1205. pmid:35268296
- 33. Juo YY, Sanaiha Y, Khrucharoen U. Complete impact of care fragmentation on readmissions following urgent abdominal operations. J Gastrointest Surg. 2019;23:1643–51.
- 34. Li Y, Lu LX, Lu SF. The value of health IT Interoperability: Evidence from interhospital transfer of heart attack patients. Epub ahead of print. 2021.
- 35. Urbach DR, Baxter NN. Does it matter what a hospital is “high volume” for? Specificity of hospital volume-outcome associations for surgical procedures: Analysis of administrative data. Qual Saf Health Care. 2004;13(5):379–83. pmid:15465942
- 36. Vemulapalli S, Dai D, Hammill BG, et al. Hospital resource utilization before and after transcatheter aortic valve replacement: The STS/ACC TVT registry. J Am Coll Cardiol 2019; 73: 1135–46.
- 37. Ando T, Adegbala O, Villablanca P, Akintoye E, Ashraf S, Shokr M, et al. Incidence and predictors of readmissions to non-index hospitals after transcatheter aortic valve replacement and the impact on in-hospital outcomes: From the nationwide readmission database. Int J Cardiol. 2019;292:50–5. pmid:31053244
- 38. Gilard M, Eltchaninoff H, Donzeau-Gouge P. Late outcomes of transcatheter aortic valve replacement in high-risk patients: The FRANCE-2 registry. J Am Coll Cardiol. 2016;68:1637–47.
- 39. Kobayashi D, Otsubo T, Imanaka Y. The effect of centralization of health care services on travel time and its equality. Health Policy. 2015;119(3):298–306. pmid:25480458