Abstract
Consideration for transcatheter aortic valve replacement (TAVR) necessitates an integrated risk assessment by members of the Heart Valve Team. The utility of the integrated risk assessment for predicting TAVR outcomes is not established. This article aims to compare the utility of the integrated risk assessment to that of the Society of Thoracic Surgeons Predicted Risk of Mortality (STS-PROM) score for predicting patient outcomes after TAVR. A total of 274 patients who underwent TAVR from January 2016 to August 2017 were included in this study. Patients were deemed intermediate or high risk by two surgeons on the Heart Valve Team based on an integrated risk assessment that incorporates the STS-PROM score, fragility measures, end-organ dysfunction, and surgeon evaluation. Patients were also deemed low, intermediate, or high risk based solely on their STS-PROM scores of <3%, ≥3% to <8%, and ≥8%, respectively. Differences in postoperative outcomes between intermediate- and high-risk groups as categorized by the integrated risk assessment versus STS-PROM were compared. There were no statistically significant differences in postoperative outcomes between patients who were deemed high and intermediate risk by the Heart Valve Team risk assessment. In contrast, postoperative complication rates were significantly higher in patients deemed high risk as compared with intermediate risk by STS-PROM. Integrated risk assessment by the Heart Valve Team is not superior to STS-PROM in predicting postoperative outcomes in patients undergoing TAVR.
Keywords: aortic valve disease, risk factors, heart team, transcatheter, valve replacement, risk score, outcomes
Transcatheter aortic valve replacement (TAVR) has been established as an alternative treatment modality in patients with severe aortic stenosis (AS) who are deemed high or intermediate risk for surgical aortic valve replacement (SAVR) or ineligible for surgery. 1 2 3 4 5 As such, surgical risk stratification by a heart team remains an important component for determination of eligibility and preoperative workup for patients being considered for TAVR.
The Society of Thoracic Surgeons (STS) developed a cardiac surgery risk model in 2008, the Predicted Risk of Mortality (PROM), which calculates the expected risk of postoperative mortality based on patient characteristics and preoperative risk factors. Although this model has been demonstrated to serve as an objective predictor of mortality in patients undergoing SAVR and concomitant SAVR plus coronary artery bypass grafting surgery, 6 7 it has been recognized that there are other patient factors which are not captured in the STS-PROM, such as frailty, major organ system dysfunction, and procedure-specific impediments which may add to the patient's surgical risk. 8 9 10 11 12 13 14 15 16 As such, in addition to the STS-PROM, the use of integrated risk assessment by the members of the Heart Valve Team is recommended for the risk assessment for all patients being considered for TAVR. While independent factors included in the integrated risk assessment by the Heart Valve Team, including frailty and end-organ dysfunction, have been found to be correlated with open surgical operative mortality, its utility for the comprehensive assessment for predicting TAVR outcomes is not established. 17
The objective of this study is to compare the utility of the integrated risk assessment by the Heart Valve Team to that of the STS-PROM score for predicting patient outcomes after TAVR.
Patients and Methods
This study was conducted with the approval of the Northwell Health System Institutional Review Board. As this is a retrospective study utilizing de-identified data that were collected for the New York State and STS databases, specific waiver of the need for individual patient consent was granted by the Institutional Review Board.
All patients who underwent elective TAVR for AS, from January 2016 to October 2017, were included in the study. Patients were deemed intermediate or high risk by two surgeons on the Heart Valve Team based on an integrated risk assessment that incorporates the STS-PROM score, fragility measures, end-organ dysfunction, and surgeon evaluation. Patients were also deemed low, intermediate, or high risk based solely on their STS-PROM scores of <3%, ≥3% to <8%, and ≥8%, respectively. Patients for whom the STS-PROM score and/or Heart Valve Team designation was unavailable were excluded. The following preoperative data were collected for each patient: age, gender, and risk factors and comorbidities (i.e., dialysis, hypertension, dyslipidemia, cerebrovascular disease, diabetes, chronic lung disease, heart failure, preoperative ejection fraction, preoperative albumin, prior myocardial infarction, prior cardiac intervention, prior coronary artery bypass grafting, and prior percutaneous coronary intervention procedure).
The clinical endpoints were composite complications, operating room extubation rate, and hospital length of stay. The composite complication endpoint represents the presence of at least one of the following complications: cardiac arrest, aortic dissection, annular dissection, stroke, major vascular complications, unplanned surgery, percutaneous coronary intervention, perforation, myocardial infarction, bleeding, and 30-day mortality.
All statistical analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC). Data analysis was performed retrospectively. Differences in postoperative outcomes between intermediate- and high-risk groups as categorized by the integrated risk assessment versus STS-PROM were compared. For each risk stratification system, logistic regression was used to examine the association between the stratification designated and the occurrence of at least one complication. Logistic regression was similarly used to examine the association between designated risk level and operating room extubation. The receiver operating characteristic (ROC) curve was plotted for each scoring system, for both composite complications and operating room extubation. The areas under these correlated ROC curves were then compared using the nonparametric approach of DeLong et al. 18 The association between each scoring system and hospital length of stay was examined using negative binomial regression. The Akaike Information Criterion (AIC) for each scoring system was calculated and compared.
Results
The preoperative characteristics based on STS-PROM and Heart Valve Team risk stratification for the 274 TAVR patients who were included in the study are listed in Tables 1 and 2 , respectively. Of the 274 patients, 70 patients were identified as intermediate risk and 204 patients were identified as high risk based on the integrated risk assessment by the Heart Valve Team. STS-PROM risk stratification identified 34 patients as low risk, 178 patients as intermediate risk, and 62 patients as high risk within the same cohort. Of the 204 patients identified as high risk by the Heart Valve Team, 62 were identified as high risk, 126 as intermediate risk, and 16 as low risk by STS-PROM alone.
Table 1. Baseline patient characteristics, stratified by STS-PROM.
| Preoperative characteristics | Low ( n = 34) | Intermediate ( n = 178) | High ( n = 62) | p -Value |
|---|---|---|---|---|
| Age, y | 76 (73–81) | 85 (81–88) | 87 (78–91) | <0.001 |
| Male | 15 (44.12) | 89 (50.00) | 26 (41.94) | 0.504 |
| Creatinine, mg/dL | 0.90 (0.80–0.99) | 1.07 (0.84–1.31) | 1.39 (1.04–2.18) | <0.001 |
| Dialysis | 0 (0.00) | 3 (1.69) | 10 (16.13) | <0.001 a |
| Hypertension | 29 (85.29) | 156 (87.64) | 57 (91.94) | 0.558 |
| Dyslipidemia | 29 (85.29) | 152 (85.39) | 54 (87.10) | 0.944 |
| Cerebrovascular disease | 8 (23.53) | 45 (25.28) | 24 (38.71) | 0.105 |
| Diabetes | 9 (26.47) | 49 (27.53) | 22 (35.48) | 0.461 |
| Chronic lung disease | 3 (8.82) | 26 (14.61) | 19 (30.64) | 0.006 |
| Heart failure | 7 (20.59) | 45 (25.28) | 25 (40.32) | 0.044 |
| Ejection fraction, % | 64 (57–72) | 65 (55–73) | 60 (50–69) | 0.081 |
| Albumin, g/dL | 4.3 (3.9–4.4) | 4.0 (3.7–4.3) | 3.9 (3.6–4.2) | 0.094 |
| Prior myocardial infarction | 5 (14.71) | 32 (17.98) | 21 (33.87) | 0.019 |
| Prior cardiac intervention | 16 (47.06) | 95 (53.37) | 44 (70.97) | 0.027 |
| Prior coronary artery bypass grafting | 1 (2.94) | 28 (15.73) | 17 (27.42) | 0.007 |
| Prior percutaneous coronary intervention | 12 (35.29) | 56 (31.46) | 32 (51.61) | 0.018 |
| Peripheral vascular disease | 6 (17.65) | 38 (21.35) | 11 (17.74) | 0.773 |
| Severe mitral regurgitation | 0 (0.00) | 12 (6.74) | 8 (12.90) | 0.063 a |
| Body mass index | 31.5 (26.5–35.0) | 26.9 (24.5–30.2) | 26.1 (22.8–30.0) | 0.002 |
Abbreviation: STS-PROM, Society of Thoracic Surgeons Predicted Risk of Mortality.
Notes: Continuous factors given as median (25th percentile to 75th percentile). Frequency and percent given for categorical factors and compared using the chi-square test except a Fisher's exact test.
Table 2. Baseline patient characteristics, stratified by heart valve team risk stratification.
| Preoperative characteristics | Intermediate ( n = 70) | High ( n = 204) | p -Value |
|---|---|---|---|
| Age, y | 82 (78–86) | 85 (78–89) | 0.005 |
| Male | 37 (52.86) | 93 (45.59) | 0.293 |
| Creatinine, mg/dL | 1.0 (0.9–1.3) | 1.12 (0.8–1.4) | 0.290 |
| Dialysis | 1 (1.43) | 12 (5.88) | 0.195 a |
| Hypertension | 61 (87.14) | 181 (88.73) | 0.722 |
| Dyslipidemia | 59 (84.29) | 176 (86.27) | 0.681 |
| Cerebrovascular disease | 19 (27.14) | 58 (28.43) | 0.836 |
| Diabetes | 25 (35.71) | 55 (26.96) | 0.165 |
| Chronic lung disease | 7 (10.00) | 41 (20.10) | 0.055 |
| Heart failure | 17 (24.29) | 60 (29.41) | 0.410 |
| Ejection fraction, % | 62 (55–73) | 64 (54–71) | 0.654 |
| Albumin, g/dL | 4.2 (4.0–4.3) | 3.9 (3.6–4.3) | 0.002 |
| Prior myocardial infarction | 14 (20.00) | 44 (21.57) | 0.782 |
| Prior cardiac intervention | 37 (52.86) | 118 (57.84) | 0.468 |
| Prior coronary artery bypass grafting | 5 (7.14) | 41 (20.10) | 0.012 |
| Prior percutaneous coronary intervention | 26 (37.14) | 74 (36.27) | 0.896 |
| Peripheral vascular disease | 7 (10.00) | 48 (23.53) | 0.015 |
| Severe mitral regurgitation | 3 (4.29) | 17 (8.33) | 0.261 |
| Body mass index | 28.7 (25.2–31.8) | 25.6 (23.7–31.1) | 0.045 |
Notes: Continuous factors given as median (25th percentile to 75th percentile). Frequency and percent given for categorical factors and compared using the chi-square test except.
Fisher's exact test.
Based on STS-PROM, high-risk patients were more likely to be older, have renal insufficiency, to require dialysis, have chronic lung disease, and have a history of heart failure. They were also more likely to have had prior myocardial infarction, to have prior cardiac interventions including percutaneous coronary interventions and/or coronary artery bypass grafting, and to have lower body mass indexes ( Table 1 ). Similarly, high-risk patients as stratified by the Heart Valve Team risk stratification also tend to be older, have a prior history of coronary artery bypass surgery, and have lower body mass index ( Table 2 ). In addition, high-risk patients as stratified by the Heart Valve Team also have lower preoperative albumin and are more likely to have peripheral vascular disease ( Table 2 ). There were no significant differences found between high- and intermediate-risk patients for other preoperative risk factors and comorbidities.
Postoperative patient outcomes based on STS-PROM and Heart Valve Team risk stratification are presented in Table 3 . Patients who were classified as high risk based on STS-PROM had significantly more complications after their TAVR than patients who were classified as intermediate risk (odds ratio [OR]: 5.51, 95% confidence interval [CI]: 1.91–15.89, p < 0.006). In contrast, there were no significant differences in complication rates between patients who were classified as high versus intermediate risk based on the Heart Valve Team risk stratification (OR: 1.65, 95% CI: 0.46–5.91, p < 0.445). There was a significant difference between the area under the ROC curve (AUC) for the predictive value of two scoring systems for post-TAVR complications ( Fig. 1 ), with STS-PROM being a significantly better predictor of post-TAVR complications (AUC: 0.696 vs. 0.542, p < 0.003).
Table 3. Patient outcomes, stratified by STS-PROM versus heart valve team risk score.
| Postoperative outcomes | Low | Intermediate | High | OR (95% CI) a | p -Value a |
|---|---|---|---|---|---|
| STS-PROM risk stratification | n = 34 | n = 178 | n = 62 | ||
| Complication rate, n (%) | 1 (2.94) | 6 (3.37) | 10 (16.13) | 5.51 (1.91–15.89) | 0.006 |
| OR extubation rate, n (%) | 30 (88.2) | 148 (83.15) | 37 (59.68) | 0.30 (0.16–0.57) | <0.001 |
| Hospital length of stay, days b | 3.2 (2.6–4.0) | 2.7 (2.4–2.9) | 4.5 (3.9–5.1) | <0.001 | |
| Heart valve team risk stratification | – | n = 70 | n = 204 | ||
| Complication rate, n (%) | – | 3 (4.29) | 14 (6.86) | 1.65 (0.46–5.91) | 0.448 |
| OR extubation rate, n (%) | – | 60 (85.71) | 155 (75.98) | 0.53 (0.25–1.11) | 0.091 |
| Hospital length of stay, days b | – | 2.6 (2.2–3.0) | 3.3 (3.0–3.6) | 0.007 |
Abbreviations: CI, confidence interval; OR, operating room; STS-PROM, Society of Thoracic Surgeons Predicted Risk of Mortality.
OR and p -value for STS-PROM risk stratification is for intermediate-risk patients versus high-risk patients.
Median (25th percentile to 75th percentile).
Fig. 1.
Receiver operating characteristic (ROC) of composite complication and mortality.
Under the STS-PROM classification, 148 (83.15%) intermediate-risk patients and 37 (59.68%) high-risk patients were extubated in the operating room. There was a significant difference in the operating room extubation rate between patients who were high versus intermediate risk based on STS-PROM (OR: 0.30, 95% CI: 0.16–0.57, p < 0.001). Under the Heart Valve Team risk stratification, 60 (85.71%) intermediate-risk patients and 155 (75.98%) high-risk patients were extubated in the operating room. In contrast to STS-PROM, there was no statistically significant difference in extubation rate between the high- and intermediate-risk groups under the Heart Valve Team risk stratification (OR: 0.53, 95% CI: 0.25–1.11, p = 0.091). There was a significant difference between AUC for the two scoring systems in operating room extubation rate ( Fig. 2 ), with STS-PROM being a better predictor of operating room extubation than Heart Valve Team risk stratification (0.638 vs. 0.555, p = 0.024).
Fig. 2.
Receiver operating characteristic (ROC) of operating room extubation rate.
There was a significant association between each stratification system and hospital length of stay. Patients in the high-risk category by STS-PROM had a significantly longer hospital stay than those in the intermediate-risk group (4.45 vs. 2.66 days, p < 0.001). Likewise, patients in the high-risk category by Heart Valve Team risk stratification also had significantly longer hospital stays when compared with intermediate-risk patients (3.32 vs. 2.59 days, p < 0.007). The AIC for the model using STS-PROM was smaller than that using the Heart Valve Team risk stratification (1065.76 vs. 1089.91), indicating that the STS-PROM model is superior to the Heart Valve Team model for predicting length of hospital stay.
Discussion
This single-center study showed that the risk stratification determined by the integrated assessment by the Heart Valve Team is not superior to that estimated by STS-PROM alone in predicting postoperative outcomes after TAVR.
When looking at risk algorithms, STS-PROM has been shown to be more effective at predicting perioperative and long-term mortality for high-risk TAVR patients than EuroSCORE and the Ambler risk score. 19 Although STS-PROM is developed to predict mortality for patients undergoing SAVRs, it has been shown to be just as effective in predicting both in-hospital and 30-day mortality for patients undergoing TAVR as the STS/ACC Transcatheter Valve Replacement registry risk calculator. 20 21 However, limitations of STS-PROM and other surgical risk models have been suggested. For instance, while these models consider baseline demographic and medical variables, they do not consider other anatomical factors, biological age, and level of frailty that may play a role in surgical risk and patient outcomes. 22 Specific factors that are not included in the STS-PROM score that may have an impact on patient outcomes include the following: liver cirrhosis, pulmonary hypertension, calcified and/or diseased aorta, recurrent pulmonary emboli, right ventricular failure, or cachexia. 23 In fact, both Arnold et al and the PARTNER trial found the STS-PROM alone to not be a predictor of death or poor outcomes following TAVR. 8 24 For this reason, a more integrated risk assessment by cardiovascular surgeons and interventional cardiologists has been adopted in patient selection methodology.
The use of an integrated Heart Valve Team approach is currently a key component of TAVR patient selection. It is a class I recommendation from both American and European professional societies and is a requirement for reimbursement from the Centers for Medicare and Medicaid Services. 17 The concept and methodology originated with the SYNTAX (Synergy Between PCI with Taxus and Cardiac Surgery) and PARTNER (Placement of Aortic Transcatheter Valves) trials as part of the patient recruitment methodologies, and it evolved to become incorporated into all cases of patients being considered for TAVR. However, despite its widespread use, there is no clear definition of the components of a Heart Valve Team's evaluation. More importantly, the data supporting its effectiveness are lacking. 17 Factors to be considered in the Heart Valve Team's assessment include STS-PROM or EuroSCORE, frailty (e.g., grip strength, walk test, graded exercise testing), laboratory indices (e.g., albumin, complete blood count, creatinine, and electrolytes), AS severity confirmed by echocardiography, other anatomical dimensions confirmed by imaging, significant comorbidities, and subjective assessment by the physicians. 23
Patient frailty is an incremental risk factor that is extensively used as part of the Heart Valve Team's evaluation. Frailty can be described as impaired physiologic ability secondary to changes in neurocognitive function and physical status, including but not limited to muscle wasting, malnutrition, and weakness. Traditionally, it was evaluated subjectively by surgeons and cardiologists, but more uniform and objective indicators of frailty, including gait speed, grip strength, weight loss, anemia, and albumin levels, are currently being used and studied in relation to SAVR and TAVR. 22 Various combinations of these objective frailty markers have been shown to be predictors of poor outcomes in patients, both independently and in combination with STS-PROM scores, albeit with disparate results. 8 11 14 24 25 A major limitation of the past literature is the lack of consistency of frailty markers used in their assessments. Thus, it is not clear which combination of markers is most reliable in predicting outcomes. In the FRAILTY-AVR study, Afilalo et al found that a scale encompassing lower-extremity weakness, cognitive impairment, anemia, and hypoalbuminemia was significantly associated with patient outcomes. 25 In contrast, Forcillo et al found that a composite of albumin level, grip strength, walk test speed, and activities of daily living was the best predictor of 30-day mortality. 11 Additionally, other studies have found no significant association between frailty markers and postoperative outcomes in TAVR patients. In their study of 1-year mortality in 732 consecutive TAVR patients, Greason et al identified no significant association between patient frailty and poor outcomes (OR = 1.02, 95% CI: 0.52, 2.02; p = 0.949). 15 Hermiller et al identified albumin levels ≤3.3 g/dL as significantly associated with 30-day and 1-year mortality in a derivation cohort of 2,482 TAVR patients and a validation cohort of 1,205 TAVR patients; however, they failed to find a significant association between other frailty markers, including grip strength, 5-m gait speed, unplanned weight loss, and early or late mortality in multivariable analysis. 26 There are additional limitations to using frailty as a predictor. First, some frailty markers are subjective and can be manipulated by patients to alter their own risk stratification. For instance, patients can put forth less effort in the grip test and the walk test and therefore increase their level of calculated risk. In addition, some frailty markers such as walk and grip tests may reflect orthopaedic limitations rather than frailty.
Another component of the integrated Heart Valve Team assessment is the subjective evaluation of patient risk and qualification for TAVR by the team of surgeons and cardiologists based on clinical intuition, or what is often referred to as the “eyeball test.” While this assessment likely involves consideration of patient characteristics that have been shown to be significantly associated with poor outcomes that are otherwise not captured in by formal risk scores, this component of the evaluation is by definition subjective in nature and subject to inherent variability among physicians. Studies by Jain et al and Pons et al comparing the performance of surgeon's subjective versus standard statistical methods of risk estimation in patients undergoing cardiac surgery found that statistical risk estimates serve as a better method to predict operative and long-term outcomes as compared with the physician's subjective risk estimate. 27 28 The results of our study are congruent with that of the earlier-mentioned studies showing that risk estimation by a statistical model such as the STS-PROM is superior to that of physician risk estimate by the Heart Valve Team.
Existing literature provides several prediction models for patient outcomes after TAVR, including traditional risk scores, newly developed risk scores, frailty markers, subjective physician assessments, and additional clinical variables. Our study provides support that STS-PROM risk stratification, although developed to estimate risk for open cardiac surgery, is applicable to TAVR patients for predicting postoperative morbidity and mortality. Furthermore, we are the first to show that the addition of the integrated risk assessment by the Heart Valve Team does not add any further predictive value for TAVR patients.
There are several limitations to this study that should be acknowledged. First, subjective evaluation by physicians is a major component of the integrated risk assessment, which presents the possibility of variance in stratification methodology from physician to physician. Second, the propensity to “upstage” the risk of patients by the Heart Valve Team to make them eligible for TAVR must be considered. Finally, as with all single-center studies, the results of this study may not be generalizable to other institutions.
Conclusion
The integrated risk assessment by the Heart Valve Team is not superior to STS-PROM in predicting postoperative outcomes in patients undergoing TAVR. While additional factors including frailty, end-organ dysfunction, and subjective surgeon evaluation may provide value in estimating long-term patient mortality outcomes, STS-PROM is sufficient for perioperative risk stratification for patients undergoing TAVR.
Footnotes
Conflict of Interest None.
References
- 1.Nishimura R A, Otto C M, Bonow R O et al. 2014 AHA/ACC guideline for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Thorac Cardiovasc Surg. 2014;148(01):e1–e132. doi: 10.1016/j.jtcvs.2014.05.014. [DOI] [PubMed] [Google Scholar]
- 2.Smith C R, Leon M B, Mack M J et al. Transcatheter versus surgical aortic-valve replacement in high-risk patients. N Engl J Med. 2011;364(23):2187–2198. doi: 10.1056/NEJMoa1103510. [DOI] [PubMed] [Google Scholar]
- 3.Arora S, Misenheimer J A, Jones W et al. Transcatheter versus surgical aortic valve replacement in intermediate risk patients: a meta-analysis. Cardiovasc Diagn Ther. 2016;6(03):241–249. doi: 10.21037/cdt.2016.03.04. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Leon M B, Smith C R, Mack M J et al. Transcatheter or surgical aortic-valve replacement in intermediate-risk patients. N Engl J Med. 2016;374(17):1609–1620. doi: 10.1056/NEJMoa1514616. [DOI] [PubMed] [Google Scholar]
- 5.Thourani V H, Kodali S, Makkar R Ret al. Transcatheter aortic valve replacement versus surgical valve replacement in intermediate-risk patients: a propensity score analysis Lancet 2016387(10034):2218–2225. [DOI] [PubMed] [Google Scholar]
- 6.O'Brien S M, Shahian D M, Filardo Get al. The Society of Thoracic Surgeons 2008 cardiac surgery risk models: part 2--isolated valve surgery Ann Thorac Surg 200988(1, Suppl):S23–S42. [DOI] [PubMed] [Google Scholar]
- 7.Shahian D M, O'Brien S M, Filardo Get al. The Society of Thoracic Surgeons 2008 cardiac surgery risk models: part 3--valve plus coronary artery bypass grafting surgery Ann Thorac Surg 200988(1, Suppl):S43–S62. [DOI] [PubMed] [Google Scholar]
- 8.Arnold S V, Reynolds M R, Lei Y et al. Predictors of poor outcomes after transcatheter aortic valve replacement: results from the PARTNER (Placement of Aortic Transcatheter Valve) trial. Circulation. 2014;129(25):2682–2690. doi: 10.1161/CIRCULATIONAHA.113.007477. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Asami M, Stortecky S, Praz F et al. Prognostic value of right ventricular dysfunction on clinical outcomes after transcatheter aortic valve replacement. JACC Cardiovasc Imaging. 2019;12(04):577–587. doi: 10.1016/j.jcmg.2017.12.015. [DOI] [PubMed] [Google Scholar]
- 10.Edwards F H, Cohen D J, O'Brien S M et al. Development and validation of a risk prediction model for in-hospital mortality after transcatheter aortic valve replacement. JAMA Cardiol. 2016;1(01):46–52. doi: 10.1001/jamacardio.2015.0326. [DOI] [PubMed] [Google Scholar]
- 11.Forcillo J, Condado J F, Ko Y A et al. Assessment of commonly used frailty markers for high- and extreme-risk patients undergoing transcatheter aortic valve replacement. Ann Thorac Surg. 2017;104(06):1939–1946. doi: 10.1016/j.athoracsur.2017.05.067. [DOI] [PubMed] [Google Scholar]
- 12.Green P, Arnold S V, Cohen D J et al. Relation of frailty to outcomes after transcatheter aortic valve replacement (from the PARTNER trial) Am J Cardiol. 2015;116(02):264–269. doi: 10.1016/j.amjcard.2015.03.061. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Hioki H, Watanabe Y, Kozuma K et al. Risk stratification using lean body mass in patients undergoing transcatheter aortic valve replacement. Catheter Cardiovasc Interv. 2018;92(07):1365–1373. doi: 10.1002/ccd.27547. [DOI] [PubMed] [Google Scholar]
- 14.Schoenenberger A W, Moser A, Bertschi D et al. Improvement of risk prediction after transcatheter aortic valve replacement by combining frailty with conventional risk scores. JACC Cardiovasc Interv. 2018;11(04):395–403. doi: 10.1016/j.jcin.2017.11.012. [DOI] [PubMed] [Google Scholar]
- 15.Greason K L, Eleid M F, Nkomo V T et al. Predictors of 1-year mortality after transcatheter aortic valve replacement. J Card Surg. 2018;33(05):243–249. doi: 10.1111/jocs.13574. [DOI] [PubMed] [Google Scholar]
- 16.Hebeler K R, Baumgarten H, Squiers J J et al. Albumin is predictive of 1-year mortality after transcatheter aortic valve replacement. Ann Thorac Surg. 2018;106(05):1302–1307. doi: 10.1016/j.athoracsur.2018.06.024. [DOI] [PubMed] [Google Scholar]
- 17.Coylewright M, Mack M J, Holmes D R, Jr, O'Gara P T. A call for an evidence-based approach to the Heart Team for patients with severe aortic stenosis. J Am Coll Cardiol. 2015;65(14):1472–1480. doi: 10.1016/j.jacc.2015.02.033. [DOI] [PubMed] [Google Scholar]
- 18.DeLong E R, DeLong D M, Clarke-Pearson D L. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics. 1988;44(03):837–845. [PubMed] [Google Scholar]
- 19.Dewey T M, Brown D, Ryan W H, Herbert M A, Prince S L, Mack M J. Reliability of risk algorithms in predicting early and late operative outcomes in high-risk patients undergoing aortic valve replacement. J Thorac Cardiovasc Surg. 2008;135(01):180–187. doi: 10.1016/j.jtcvs.2007.09.011. [DOI] [PubMed] [Google Scholar]
- 20.Arsalan M, Weferling M, Hecker F et al. TAVI risk scoring using established versus new scoring systems: role of the new STS/ACC model. EuroIntervention. 2018;13(13):1520–1526. doi: 10.4244/EIJ-D-17-00421. [DOI] [PubMed] [Google Scholar]
- 21.Codner P, Malick W, Kouz R et al. Mortality risk after transcatheter aortic valve implantation: analysis of the predictive accuracy of the Transcatheter Valve Therapy registry risk assessment model. EuroIntervention. 2018;14(04):e405–e412. doi: 10.4244/EIJ-D-18-00032. [DOI] [PubMed] [Google Scholar]
- 22.Van Mieghem N M, Serruys P W. The art of risk stratification in TAVI. Eur Heart J. 2013;34(25):1859–1861. doi: 10.1093/eurheartj/eht137. [DOI] [PubMed] [Google Scholar]
- 23.Mylotte D, Martucci G, Piazza N. Patient selection for transcatheter aortic valve implantation: an interventional cardiology perspective. Ann Cardiothorac Surg. 2012;1(02):206–215. doi: 10.3978/j.issn.2225-319X.2012.06.08. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Arnold S V, Afilalo J, Spertus J A et al. Prediction of poor outcome after transcatheter aortic valve replacement. J Am Coll Cardiol. 2016;68(17):1868–1877. doi: 10.1016/j.jacc.2016.07.762. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Afilalo J, Lauck S, Kim D H et al. Frailty in older adults undergoing aortic valve replacement: the FRAILTY-AVR study. J Am Coll Cardiol. 2017;70(06):689–700. doi: 10.1016/j.jacc.2017.06.024. [DOI] [PubMed] [Google Scholar]
- 26.Hermiller J B, Jr, Yakubov S J, Reardon M J et al. Predicting early and late mortality after transcatheter aortic valve replacement. J Am Coll Cardiol. 2016;68(04):343–352. doi: 10.1016/j.jacc.2016.04.057. [DOI] [PubMed] [Google Scholar]
- 27.Jain R, Duval S, Adabag S. How accurate is the eyeball test?: a comparison of physician's subjective assessment versus statistical methods in estimating mortality risk after cardiac surgery. Circ Cardiovasc Qual Outcomes. 2014;7(01):151–156. doi: 10.1161/CIRCOUTCOMES.113.000329. [DOI] [PubMed] [Google Scholar]
- 28.Pons J M, Borras J M, Espinas J A, Moreno V, Cardona M, Granados A. Subjective versus statistical model assessment of mortality risk in open heart surgical procedures. Ann Thorac Surg. 1999;67(03):635–640. doi: 10.1016/s0003-4975(98)01326-5. [DOI] [PubMed] [Google Scholar]