[Skip to Content]
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
[Skip to Content Landing]
Figure 1.
Per Patient Change in Society of Thoracic Surgeons (STS) Score Between 2008/2009 and 2015
Per Patient Change in Society of Thoracic Surgeons (STS) Score Between 2008/2009 and 2015

Each line represents 1 individual patient. The blue line represents the mean STS score.

Figure 2.
Scatterplot of Society of Thoracic Surgeons (STS) Scores in 2008/2009 Compared With 2015
Scatterplot of Society of Thoracic Surgeons (STS) Scores in 2008/2009 Compared With 2015

One hundred four patients (58.7% of cohort) in the blue shaded area of the graph had an STS score of at least 10% in 2008/2009 but less than 10% when recalculated in 2015.

1.
Statistical Methodology for STS Risk Models and Performance Measures. http://www.sts.org/quality-research-patient-safety/statistical-methodology-risk-models-and-measures. Accessed August 22, 2016.
2.
Mack  MJ, Brennan  JM, Brindis  R,  et al; STS/ACC TVT Registry.  Outcomes following transcatheter aortic valve replacement in the United States.  JAMA. 2013;310(19):2069-2077.PubMedGoogle ScholarCrossref
3.
Capodanno  D, Barbanti  M, Tamburino  C,  et al; OBSERVANT Research Group.  A simple risk tool (the OBSERVANT score) for prediction of 30-day mortality after transcatheter aortic valve replacement.  Am J Cardiol. 2014;113(11):1851-1858.PubMedGoogle ScholarCrossref
4.
In-Hospital Mortality Risk Calculator TAVR. http://tools.acc.org/TAVRRisk/. Accessed August 22, 2016.
5.
Ribeiro  HB, Rodés-Cabau  J.  The multiparametric FRANCE-2 risk score: one step further in improving the clinical decision-making process in transcatheter aortic valve implantation.  Heart. 2014;100(13):993-995.PubMedGoogle ScholarCrossref
6.
Hermiller  JB  Jr, Yakubov  SJ, Reardon  MJ,  et al; CoreValve United States Clinical Investigators.  Predicting early and late mortality after transcatheter aortic valve replacement.  J Am Coll Cardiol. 2016;68(4):343-352.PubMedGoogle ScholarCrossref
Research Letter
April 2017

Society of Thoracic Surgeons Score Variance Results in Risk Reclassification of Patients Undergoing Transcatheter Aortic Valve Replacement

Author Affiliations
  • 1MedStar Washington Hospital Center, Washington, DC
  • 2Cardiovascular and Pulmonary Branch, Division of Intramural Research, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland
 

Copyright 2016 American Medical Association. All Rights Reserved.

JAMA Cardiol. 2017;2(4):455-456. doi:10.1001/jamacardio.2016.4132

The Society of Thoracic Surgeons (STS) score is the cornerstone of risk assessment for patients with symptomatic severe aortic stenosis undergoing transcatheter aortic valve replacement (TAVR) in the United States. The current STS score model was implemented in 2008, and the coefficients are reestimated every year using the last 3-year study sample in the STS database.1 Since 2007, high-risk patients have increasingly undergone TAVR, and advances in surgical technique, anesthesia, and critical care have led to improved surgical aortic valve replacement outcomes. Consequently, the STS database study sample has changed considerably. To evaluate the effect of these changes, we compared the STS scores of 177 consecutive patients undergoing TAVR in 2008/2009 in a tertiary referral hospital in the United States with their recalculated STS scores in 2015, using the updated online calculator.

Methods

The study was approved by the local institutional review board and in compliance with its regulation. Categorical variables are presented as numbers (percentages), and continuous variables are presented as mean (SD) values.

Results

In 2008/2009, of 177 patients, 141 (79.7%) and 34 (19.2%) were classified as high and extreme risk, respectively. Risk classification was unclear for 2 patients (1.1%). There was a high prevalence of clinically significant comorbidities, including chronic obstructive pulmonary disease (57 of 177 [32.2%]), peripheral vascular disease (57 of 174 [32.8%]), renal insufficiency (125 of 176 [71.0%]), prior coronary artery bypass graft (64 of 177 [36.2%]), and prior stroke or transient ischemic attack (45 of 175 [25.7%]). The mean (SD) STS score was 11.6% (3.9%) and 6.4% (2.8%) in 2008/2009 and 2015, respectively (Figure 1). The mean difference or bias between the STS score in 2008/2009 and the STS score in 2015 was −5.2%, with limits of agreement at −11.7% and 1.3%. Fifty-eight percent of patients classified as high risk in 2008/2009 would be reclassified as intermediate risk in 2015 based on the STS score alone (Figure 2). Forty percent of patients would not change classification. Mortality at 30 days and at 1 year was 11.9% and 27.1%, respectively.

Discussion

Certainly, risk assessment is influenced by additional factors, such as frailty, pulmonary hypertension, or liver disease, that we did not account for in this study. Nonetheless, the STS score remains the starting point for risk determination for all patients undergoing TAVR in the United States. Our data suggest that there may be substantial overlap between intermediate-risk populations in 2016 and high-risk populations studied in the original pivotal trials because of evolving STS scores. Consequently, caution should be taken when comparing results of clinical trials performed at different times.

Around the world, patients with lower STS scores are being treated with TAVR, and this so-called risk creep is apparent in the United States. Between November 2011 and March 2013, the mean STS score in the Transcatheter Valve Therapy Registry that captures all TAVR procedures in the United States was 7%.2 Our data suggest that part of this risk creep is simply owing to evolving STS scores. Recognizing the limitations of scores based on surgical aortic valve replacement, a number of alternative scores have been proposed,3-6 although they have yet to be widely adopted into clinical practice.

Following approval of TAVR for intermediate-risk patients in the United States in 2016, it is likely that, in the future, surgical aortic valve replacement rarely will be performed for patients with STS scores greater than 4%. Therefore, the use of any risk score based on historic outcomes of patients undergoing aortic valve replacement will become obsolete for evaluation of patients undergoing TAVR. This raises the important question of whether indications for use for TAVR devices in the United States should reflect STS score risk-defined patient populations. Possibly a better approach would be to allow the heart team to determine optimal treatment strategy for individual patients regardless of risk score.

Conclusions

In summary, changes to the STS score algorithm over time can result in significant reduction in STS scores for patients undergoing TAVR. This observation has important implications for the conduct and interpretation of contemporary clinical trials and device indications for use because there may be considerable unrecognized overlap in risk-defined patient populations. New risk stratification scores are warranted based on outcomes of patients undergoing TAVR.

Back to top
Article Information

Corresponding Author: Ron Waksman, MD, MedStar Washington Hospital Center, 110 Irving St NW, Ste 4B-1, Washington, DC 20010 (ron.waksman@medstar.net).

Published Online: November 16, 2016. doi:10.1001/jamacardio.2016.4132

Author Contributions: Drs Rogers and Koifman had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Rogers, Koifman, Torguson, Waksman.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Rogers, Koifman, Patel, Torguson, Waksman.

Critical revision of the manuscript for important intellectual content: Koifman, Gai, Torguson, Corso, Waksman.

Statistical analysis: Rogers, Koifman, Patel, Gai, Torguson.

Administrative, technical, or material support: Koifman, Corso.

Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Waksman is a consultant for Abbott Vascular, Biotronik, Boston Scientific, Medtronic, and St Jude Medical and is on the speakers bureau for AstraZeneca, Boston Scientific, Merck; grant support was provided by AstraZeneca, Biotronik, and Boston Scientific. No other disclosures are reported.

Additional Contributions: We thank Arie Steinvil, MD, Petros Okubagzi, MD, Sarkis Kiramijyan, MD, Sang Yeub Lee, MD, Christian Shults, MD, Lowell Satler, MD, Itsik Ben-Dor, MD, and Augusto Pichard, MD, all at the MedStar Washington Hospital Center, for their support and guidance. No compensation was received for their contributions.

References
1.
Statistical Methodology for STS Risk Models and Performance Measures. http://www.sts.org/quality-research-patient-safety/statistical-methodology-risk-models-and-measures. Accessed August 22, 2016.
2.
Mack  MJ, Brennan  JM, Brindis  R,  et al; STS/ACC TVT Registry.  Outcomes following transcatheter aortic valve replacement in the United States.  JAMA. 2013;310(19):2069-2077.PubMedGoogle ScholarCrossref
3.
Capodanno  D, Barbanti  M, Tamburino  C,  et al; OBSERVANT Research Group.  A simple risk tool (the OBSERVANT score) for prediction of 30-day mortality after transcatheter aortic valve replacement.  Am J Cardiol. 2014;113(11):1851-1858.PubMedGoogle ScholarCrossref
4.
In-Hospital Mortality Risk Calculator TAVR. http://tools.acc.org/TAVRRisk/. Accessed August 22, 2016.
5.
Ribeiro  HB, Rodés-Cabau  J.  The multiparametric FRANCE-2 risk score: one step further in improving the clinical decision-making process in transcatheter aortic valve implantation.  Heart. 2014;100(13):993-995.PubMedGoogle ScholarCrossref
6.
Hermiller  JB  Jr, Yakubov  SJ, Reardon  MJ,  et al; CoreValve United States Clinical Investigators.  Predicting early and late mortality after transcatheter aortic valve replacement.  J Am Coll Cardiol. 2016;68(4):343-352.PubMedGoogle ScholarCrossref
×