Abstract: Background. Transcatheter aortic valve implantation (TAVI) is a valid alternative to surgical aortic valve replacement for the treatment of symptomatic aortic stenosis. The EuroScore (ES) II, logistic EuroScore (log ES), and the Society of Thoracic Surgeons (STS) score are the most applied scores for surgical risk stratification. However, their predictive value for patients undergoing TAVI is still unclear. Aim. To evaluate the performance of STS, log ES and ES II as predictors of short-term and long-term mortality in patients undergoing TAVI. Methods. Between February 2008 and October 2017, a total of 384 patients underwent transfemoral TAVI at our institution and constituted the study population. Patients were divided into three groups based on the class of risk (low, intermediate, and high) calculated by each score. In-hospital complications, 30-day outcomes, and 5-year outcomes were assessed. Results. In-hospital mortality rate was 2.6% (n = 10). All scores over-estimated the risk of 30-day mortality, especially for the highest risk classes. At the end of follow-up (5 years), STS risk stratification was able to stratify all-cause and cardiovascular (CV) mortality (P<.01 and P=.02, respectively). Patients with intermediate ES II risk showed a lower survival rate (P=.04) while CV deaths did not differ between classes of risk. All-cause mortality and CV mortality curves did not diverge according to the patients’ risk profiles derived from log ES. Conclusion. Conventional surgical risk scores are not appropriate to predict 30-day mortality in patients undergoing transfemoral TAVI. STS assessment was the only risk score able to stratify long-term all-cause and CV mortality.
J INVASIVE CARDIOL 2019;31(10):E282-E288.
Key words: aortic stenosis, EuroScore II, logistic EuroScore, risk scores, STS
Aortic stenosis (AS) is the most common primary valve disease, and carries a high short-term mortality if not treated. Valve replacement is considered the first-line therapy for patients with severe AS presenting with symptoms and/or systolic left ventricular dysfunction not due to another cause.1-4
AS mainly affects elderly subjects, as its prevalence increases with age. Therefore, patients often present multiple comorbidities that increase the surgical risk. Nowadays, transcatheter aortic valve implantation (TAVI) is recommended in patients who are not suitable for surgical aortic valve replacement (SAVR) and in patients at increased surgical risk after adequate evaluation performed by the heart team. In fact, TAVI has been approved as an effective and less-invasive treatment option in high and intermediate operative risk patients.5-8 Accumulated operator experience and advances in device technology, especially new-generation devices, have led to the use of TAVI for selected patients with bicuspid AS or degenerated bioprosthetic surgical valves.9-13 More recently, new data on low-risk patients showed non-inferiority or superiority of TAVI in comparison with SAVR — findings that will certainly further broaden the indications for TAVI.14-15
The European System for Cardiac Operative Risk Evaluation II (EuroScore II),16 logistic EuroScore (log ES),17 and the Society of Thoracic Surgeons (STS) score18 are the most commonly used scoring systems to stratify surgical risk. However, while these scores have been shown to correctly predict SAVR outcomes, their effectiveness in short-term and long-term mortality prediction among patients undergoing TAVI is unclear. In the present study, we sought to evaluate in-hospital and long-term prognosis following TAVI procedure according to patients’ risk profiles.
This study included 384 consecutive patients who were treated with transfemoral TAVI between February 2008 and October 2017 at the St. Orsola-Malpighi University Hospital. All patients had severe symptomatic AS, defined as an aortic valve area <1.0 cm2 and/or a mean transvalvular gradient >40 mm Hg or a peak velocity >4 m/sec by a transthoracic echocardiogram evaluation.19 Patients with symptomatic degenerated bioprosthetic aortic valve were also included.
Feasibility of femoral access was assessed by computed tomography (CT) scan and/or peripheral angiography.20 All cases underwent a multidisciplinary evaluation in order to define the risk profile and were collectively discussed by the heart team.
EuroScore II, log ES, and STS score were calculated using the online calculators. Low risk was defined as a EuroScore II <4%, log ES <10%, and STS score <4%. Intermediate risk was defined as EuroScore II 4%-9%, log ES 10%-20%, and STS score 4%-8%. High risk was defined as a EuroScore II >9%, log ES >20%, and STS score >8%. Patients were divided into the three groups based on the risk classification.
The procedures were performed in the cardiac catheterization laboratory or hybrid operating room. Clinical outcomes were evaluated according to the Valve Academic Research Consortium (VARC)-2 consensus document.21 In-hospital mortality and major adverse cardiac and cerebrovascular events (MACCEs) were reported. Cardiovascular (CV) mortality was defined as composite of death from cardiac cause, sudden death, any death without other known cause, and fatal stroke. Long-term follow-up was performed by clinical outpatient evaluations and/or telephone interviews.
Written consent was obtained in all cases. This study conforms to the principles outlined in the Declaration of Helsinki and was approved by the local ethics committee.
Statistical analysis. Categorical variables are expressed as number and percentage, while continuous variables are expressed as mean ± standard deviation. Comparisons between categorical variables were performed with the Chi-square test; comparisons between continuous variables were performed with the Student’s t-test. Z-tests was used to compare two proportions. Kaplan-Meier curves were constructed to evaluate survival, and differences between subgroups were tested based on the log-rank test. For all comparisons, a P-value <.05 was considered significant. Cox regression analyses were performed to evaluate the association between surgical scores and long-term outcomes. Discrimination of the three surgical scores was assessed with receiver-operating characteristic (ROC) curves. The Stata software package version 14.2 (StataCorp) was used for all analyses.
Baseline clinical characteristics are summarized in Table 1. Mean age was 83.0 ± 6.7 years, 45.6% were male, and the mean body mass index was 26.3 ± 5.0 kg/m2. Seventy-nine patients (20.6%) had a previous myocardial infarction, 30 (7.8%) a previous stroke, and 35 (9.1%) had a permanent pacemaker at baseline. Porcelain aorta and hostile thorax were present in 5.2% and 0.8% of patients, respectively. Patients had undergone previous percutaneous coronary interventions in 33.9%, coronary artery bypass grafting in 10.2%, and aortic valve replacement (AVR) in 8.1%.
Median left ventricular ejection fraction was 59.2 ± 12.8%. Of the 384 patients, a total of 286 had tricuspid insufficiency, with an estimated systolic pulmonary artery pressure ≥50 mm Hg in 36 cases (12.6%). The average STS score was 5.9%, average log ES score was 15.7%, and average EuroScore II was 6.3%. A consistent number of patients had a low surgical risk for each score (38.0% for STS, 37.2% for log ES, and 43.2% for EuroScore II). Overall, a total of 173 patients (45.1%) were treated with the CoreValve (Medtronic), 208 (54.1%) with the Sapien valve (Edwards Lifesciences), and 2 (0.5%) with Portico valves (Abbott Vascular). One patient died before valve implantation due to a procedural vascular complication.
In-hospital outcomes and 30-day mortality. In-hospital mortality rate was 2.6% (n = 10), myocardial infarction rate was 2.3% (n = 9), and stroke rate was 0.8% (n = 3). Thirteen patients (3.3%) had a postprocedural life-threatening bleed and 14 patients (3.6%) had a major vascular complication. Forty-two patients (10.9%) required pacemaker implantation, while 54 patients (14.0%) developed a left bundle-branch block and 20 patients (5.2%) had a new-onset atrial fibrillation. No statistically significant correlation was found between in-hospital major complications and the grade of risk obtained by each scoring system (Figure 1).
Observed 30-day mortality was lower than predicted mortality calculated by any score (STS score, log ES, and EuroScore II), especially for the highest classes of risk (Figure 2).
Long-term follow-up. Kaplan-Meier survival curves according to the three risk subgroups are shown in Figure 3. At 5-year follow-up, a total of 88 patients died (22.9%). Risk stratification obtained throughout STS assessment was able to stratify both overall mortality and CV-related mortality (P<.01 and P=.02, respectively). Patients with intermediate EuroScore II risk showed a lower survival rate (P=.04) compared with the other risk groups, while CV deaths did not differ between classes of risk. All-cause mortality and CV mortality curves did not diverge according to the patients’ risk profiles derived from log ES. Furthermore, Cox regression analysis confirmed these results (Figure 4A). STS score, but not EuroScore II and log ES, provides an independent predictor of long-term mortality. The better performance of the STS score for predicting long-term mortality was confirmed by ROC curve analysis regarding 5-year mortality (Figure 4B). The area under the curve was 0.701 for STS score, indicating good discrimination power in this risk model.
STS score, log ES, and EuroScore II were designed and validated for the prediction of perioperative mortality in a conventional cardiac surgery setting. However, they have been systematically used in TAVI trials to select patient populations according to the surgical risk, and widely used during heart team discussions for decision-making, even if it is recommended that they should be integrated in a multiparametric evaluation. With new data showing at least non-inferior outcomes of TAVI vs SAVR for any surgical risk category, their potential utility in TAVI candidates remains elusive.
In our study of patients undergoing transfemoral TAVI, in-hospital mortality increased alongside with the surgical risk value estimated with the log ES and EuroScore II (without statistical significance), but not with the STS score. Furthermore, 30-day mortality was over-estimated by all risk scores for all risk classes, and the difference between predicted and observed death rates was striking, especially for high-risk patients. Although the difference between predicted and observed mortality rates with STS score was not statistically significant, it was still numerically (and clinically) meaningful. These results are in line with previous studies.22-24 Similarly, the rate of the main postprocedural complications (acute myocardial infarction, life-threatening bleeding, and major vascular complications) did not correlate with the degree of surgical risk derived from each score. In-hospital and 30-day outcome results can be explained by the fact that in TAVI patients, mortality and postprocedural complications are more commonly related to operator experience and anatomical abnormalities rather than patient comorbidities. Our results are consistent with a registry analysis by Hemmann et al25 that included a total of 426 TAVI procedures (274 transfemoral and 152 transapical) in which the observed mortality in the transfemoral group was lower than predicted mortality by any of the three scores (STS score, log ES, and EuroScore II).The same results are shown in a recent subanalysis of the SOURCE 3 registry,26 in which all-cause 30-day mortality was lower compared with what would be expected with surgery, in each risk class obtained by EuroScore II and log ES.
Regarding long-term outcome, patients with elevated STS risk showed significantly lower CV-related and overall survival rates at 5 years, whereas risk stratification obtained through log ES and EuroScore II assessment was not able to predict either overall or CV-related mortality. This expands previous observations at 1 year suggesting that only the STS score could significantly discriminate patients with good or compromised prognosis.27 In this single-center study, patients were categorized as “high risk” or “non-high risk” using a log ES threshold of 20%, a EuroScore II of 8%, and an STS score of 10%. Thirty-day and 1-year survival rates were significantly different between non-high risk and high-risk patients according only to STS score. In the study conducted by Hemmann et al,25 STS score was an independent predictor of long-term mortality in the multivariate analysis.
Our study confirms that surgical scores are not reliable in predicting procedural outcomes in patients undergoing TAVI, and this is due to several factors. First, the surgical scores present several limitations in risk assessment of elderly patients, who are the typical patients referred to TAVI. Specifically, they do not include indicators such as fragility or degree of disability, and echocardiographic variables such as low-flow AS. These conditions in elderly patients are related to in-hospital mortality.28 Moreover, they do not include anatomical variables that may influence outcomes in TAVI patients.
The reason for the STS score’s ability to stratify long-term outcomes probably relates to the fact that 5-year mortality is related to the presence of comorbidities (age, renal failure, coronary artery disease, chronic obstructive pulmonary disease, vascular diseases, and other cardiac valve disease) manifesting over time. However, additional factors, including procedural results (eg, presence of significant paravalvular leak or significant patient-prosthesis mismatch), can significantly affect long-term outcomes post TAVI.
Taken together, the results of our study, in combination with previous findings, call for the development of a dedicated TAVI scoring system able to predict more accurately both procedural and postprocedural outcomes. In this regard, the Society of Thoracic Surgeons/American College of Cardiology Transcatheter Aortic Valve Replacement score for in-hospital mortality appears quite promising.29,30
Our study confirms a disconnection between periprocedural mortality after TAVI observed and rates predicted by surgical risk scores. Development of TAVI-specific scores to accurately predict short-term and long-term mortality rates after TAVI seems desirable.
1. Charlson E, Legedza AT, Hamel MB. Decision-making and outcomes in severe symptomatic aortic stenosis. J Heart Valve Dis. 2006;15:312-321.
2. Baumgartner H, Falk V, Bax JJ, et al. 2017 ESC/EACTS guidelines for the management of valvular heart disease. Eur Heart J. 2017;38:2739-2791.
3. Nishimura RA, Otto CM, Bonow RO, 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:e1-e132.
4. Nishimura RA, Otto CM, Bonow RO, et al. 2017 AHA/ACC focused update of the 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 Clinical Practice Guideline. J Am Coll Cardiol. 2017;70:252-289.
5. Kapadia SR, Leon MB, Makkar RR, et al; PARTNER Trial Investigators. 5-year outcomes of trans-catheter aortic valve replacement compared with standard treatment for patients with inoperable aortic stenosis (PARTNER 1): a randomised controlled trial. Lancet. 2015;385:2485-2491.
6. Mack MJ, Leon MB, Smith CR, et al; PARTNER Trial Investigators. 5-year outcomes of transcatheter aortic valve replacement or surgical aortic valve replacement for high surgical risk patients with aortic stenosis (PARTNER 1): a randomised controlled trial. Lancet. 2015;385:2477-2484.
7. Leon MB, Smith CR, Mack MJ, et al; PARTNER 2 Investigators. Transcatheter or surgical aortic-valve replacement in intermediate-risk patients. N Engl J Med. 2016;374:1609-1620.
8. Reardon MJ, Van Mieghem NM, Popma JJ, et al; SURTAVI Investigators. Surgical or transcatheter aortic-valve replacement in intermediate-risk patients. N Engl J Med. 2017;376:1321-1331.
9. Yoon SH, Lefèvre T, Ahn JM, et al. Transcatheter aortic valve replacement with early- and new-generation devices in bicuspid aortic valve stenosis. J Am Coll Cardiol. 2016;68:1195-1205.
10. Bauer T, Linke A, Sievert H, et al. Comparison of the effectiveness of transcatheter aortic valve implantation in patients with stenotic bicuspid versus tricuspid aortic valves (from the German TAVI Registry). Am J Cardiol. 2014;113:518-521.
11. Mylotte D, Lefèvre T, Søndergaard L, et al. Transcatheter aortic valve replacement in bicuspid aortic valve disease. J Am Coll Cardiol. 2014;64:2330-2339.
12. Webb JG, Wood DA, Ye J, et al. Transcatheter valve-in-valve implantation for failed bioprosthetic heart valves. Circulation. 2010;121:1848-1857.
13. Dvir D, Webb JG, Bleiziffer S, et al; Valve-in-Valve International Data Registry Investigators. Transcatheter aortic valve implantation in failed bioprosthetic surgical valves. JAMA. 2014;312:162-170.
14. Mack MJ, Leon MB, Thourani VH, et al; PARTNER 3 Investigators. Transcatheter aortic-valve replacement with a balloon-expandable valve in low-risk patients. N Engl J Med. 2019;380:1695-1705 (Epub 2019 Mar 16).
15. Popma JJ, Deeb GM, Yakubov SJ, et al; Evolut Low Risk Trial Investigators. Transcatheter aortic-valve replacement with a self-expanding valve in low-risk patients. N Engl J Med. 2019;380:1706-1715 (Epub 2019 Mar 16).
16. Nashef SA, Roques F, Sharples LD, et al. EuroSCORE II. Eur J Cardiothorac Surg. 2012;41:734-744.
17. Roques F, Michel P, Goldstone AR, Nashef SA. The logistic EuroSCORE. Eur Heart J. 2003;24:881-882.
18. O’Brien SM, Shahian DM, Filardo G, et al; Society of Thoracic Surgeons Quality Measurement Task Force. The Society of Thoracic Surgeons 2008 cardiac surgery risk models: part 2 isolated valve surgery. Ann Thorac Surg. 2009;88:S23-S42.
19. Baumgartner H, Hung J, Bermejo J, et al. Echocardiographic assessment of valve stenosis: EAE/ASE recommendations for clinical practice. Eur J Echocardiogr. 2009;10:1-25.
20. Otto CM, Kumbhani DJ, Alexander KP, et al. 2017 ACC expert consensus decision pathway for transcatheter aortic valve replacement in the management of adults with aortic stenosis: a report of the American College of Cardiology Task Force on Clinical Expert Consensus Documents. J Am Coll Cardiol. 2017;69:1313-1346.
21. Kappetein AP, Head SJ, Genereux P, et al. Updated standardized endpoint definitions for transcatheter aortic valve implantation: the Valve Academic Research Consortium-2 consensus document. J Thorac Cardiovasc Surg. 2013;145:6-23.
22. Wang TKM, Wang MTM, Gamble GD, Webster M, Ruygrok PN. Performance of contemporary surgical risk scores for transcatheter aortic valve implantation: a meta-analysis. Int J Cardiol. 2017;236:350-355.
23. Martin GP, Sperrin M, Ludman PF, et al. Inadequancy of existing clinica prediction models for predicting mortality after transcatheter aortic valve implantation. Am Heart J. 2017;184:97-105.
24. Durand E, Borz B, Godin M, et al. Performance analysis of EuroSCORE II compared to the original Logistic EuroSCORE and STS scores for predicting 30-day mortality after transcatheter aortic valve replacement. Am J Cardiol. 2013;111:891-897.
25. Hemmann K, Sirotina M, De Rosa S, et al. The STS score is the strongest predictor of long-term survival following transcatheter aortic valve implantation, whereas access route (transapical versus transfemoral) has no predictive value beyond the periprocedural phase. Interact Cardiovasc Thorac Surg. 2013;17:359-364.
26. Wendler O, Schymik G, Treede H, et al. SOURCE 3: 1-year outcomes post-transcatheter aortic valve implantation using the latest generation of the balloon-expandable transcatheter heart valve. Eur Heart J. 2017;38:2717-2726.
27. Collas VM, Van De Heyning CM, Paelinck BP, Rodrigus IE, Vrints CJ, Bosmans JM. Validation of transcatheter aortic valve implantation risk scores in relation to early and mid-term survival: a single-centre study. Interact Cardiovasc Thorac Surg. 2016;22:273-279.
28. Schoenenberger AW, Stortecky S, Neumann S, et al. Predictors of functional decline in elderly patients undergoing transcatheter aortic valve implantation (TAVI). Eur Heart J. 2013;34:684-692.
29. Edwards FH, Cohen DJ, O’Brien SM, et al. Development and validation of a risk prediction model for in-hospital mortality after transcatheter aortic valve replacement. JAMA Cardiol. 2016;1:46-52.
30. 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;13:1520-1526.
From the Interventional Cardiology Unit, Cardio-Thoracic-Vascular Department, University Hospital of Bologna, Policlinico S. Orsola-Malpighi, Bologna, Italy.
Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Palmerini and Dr Saia report personal fees from Edwards Lifesciences. The remaining authors report no conflicts of interest regarding the content herein.
Manuscript submitted May 11, 2019, and accepted May 24, 2019.
Address for correspondence: Dr Francesco Saia, Interventional Cardiology Unit, Cardio-Thoracic-Vascular Department, Policlinico S.Orsola-Malpighi (Pavilion 23), Via Massarenti 9, 40138 Bologna, Italy. Email: email@example.com