Summary
Background
The Logistic European System for Cardiac Operative Risk Evaluation (EuroSCORE) and the Society of Thoracic Surgeons (STS) score are routinely used to identify patients at high surgical risk as potential candidates for transcatheter aortic valve implantation (TAVI).
Aims
To compare the new EuroSCORE II with the Logistic EuroSCORE and the STS score.
Methods
From October 2006 to June 2011, patients with severe symptomatic aortic stenosis who underwent a TAVI were enrolled prospectively.
Results
Among 272 patients, the EuroSCORE II was significantly lower and moderately correlated with the Logistic EuroSCORE (9 ± 8% vs. 23 ± 14%, P < 0.01; r = 0.61, P < 0.001), but similar to and poorly correlated with the STS (10 ± 9%, P = 0.10; r = 0.25, P < 0.001). Based on recommended high-risk thresholds (Logistic EuroSCORE ≥ 20%; STS ≥ 10%), a EuroSCORE II ≥ 7% provided the best diagnostic value. However, using the EuroSCORE II, Logistic EuroSCORE or STS score, only 51%, 58% and 37% of patients, respectively, reached these thresholds. Contingency analyses showed that agreements between the EuroSCORE II and the Logistic EuroSCORE or the STS score were modest or poor, respectively, with a risk assessment different in 28% and 36% of patients, respectively.
Conclusions
A EuroSCORE II ≥ 7% corresponded to a Logistic EuroSCORE ≥ 20% or STS score ≥ 10%, but correlations and agreements were at best modest and only approximately half of the patients reached these thresholds. Our results highlight the limits of current scoring systems and reinforce the European guidelines stressing the importance of clinical judgment in addition to risk scores.
Résumé
Contexte
Le Logistic EuroSCORE et le Society of Thoracic Surgeons (STS) score sont utilisés pour identifier les patients à haut risque chirurgical potentiellement candidats au remplacement valvulaire aortique percutané (TAVI).
Objectifs
Comparer l’EuroSCORE II avec le Logistic EuroSCORE et le STS score.
Méthodes
D’octobre 2006 à juin 2011, 272 patients traités par TAVI pour sténose aortique sévère symptomatique ont été inclus prospectivement.
Résultats
L’EuroSCORE II était plus faible et modérément corrélé au Logistic EuroSCORE (9 ± 8 % vs 23 ± 14 %, p < 0,01 ; r = 0,61, p < 0,001), mais non différent et médiocrement corrélé au STS (10 ± 9 %, p = 0,10 ; r = 0,25, p < 0,001). En se basant sur les seuils définissant le haut risque chirurgical (Logistic EuroSCORE ≥ 20 % ; STS ≥ 10 %), un EuroSCORE II ≥ 7 % a conduit à la meilleure valeur diagnostique. Cependant, le calcul des EuroSCORE II, Logistic EuroSCORE et STS score n’a permis d’atteindre ces seuils que dans 51 %, 58 % et 37 % des cas, respectivement. L’analyse de contingence a révélé que l’accord entre EuroSCORE II, Logistic EuroSCORE ou STS score était modéré ou médiocre avec une stratification du risque différente pour 28 % et 36 % des patients, respectivement.
Conclusions
Un EuroSCORE II ≥ 7 % correspond à un Logistic EuroSCORE ≥ 20 % ou un STS ≥ 10 %. Les corrélations et agréments étaient au mieux modestes et seulement la moitié des patients ont atteint ces seuils. Nos résultats confirment les limites des scores actuels et renforcent les recommandations européennes qui soulignent l’importance de l’évaluation clinique en complément de l’utilisation des scores.
Background
Since its introduction in 2002 , transcatheter aortic valve implantation (TAVI) has become increasingly popular for the treatment of severe aortic stenosis. It is now considered a valuable alternative to surgical aortic valve replacement for patients with contraindications to surgery or those considered to be at high surgical risk . In order to assess patients’ operative risk, several scoring systems have been developed. The European System for Cardiac Operative Risk Evaluation (EuroSCORE) and the Society of Thoracic Surgeons (STS) predicted risk of mortality score are the most commonly used. The European society recommends to consider performing a TAVI when the Logistic EuroSCORE is ≥ 20% and/or the STS score is ≥ 10% .
Recently, the EuroSCORE II has been proposed as an updated version of the Logistic EuroSCORE in order to provide a better assessment of the perioperative mortality risk of patients undergoing open heart surgery, especially heart valve surgery . However, comparisons between the new EuroSCORE II and the Logistic EuroSCORE and the STS score in the setting of TAVI are rare. There are currently limited data regarding the EuroSCORE II in high-risk patients and no threshold value has been proposed to define high-risk patients. Thus, the aims of the present study were to:
- •
compare the EuroSCORE II to the Logistic EuroSCORE and the STS score;
- •
determine a EuroSCORE II value corresponding to the Logistic EuroSCORE threshold of 20% and the STS score threshold of 10%;
- •
evaluate the agreement between these scoring systems with regards to the risk classification for surgery.
Methods
Study population
Consecutive patients with severe symptomatic aortic stenosis who underwent a TAVI using the Edwards-Sapien (Edwards Lifesciences, Irving, CA, USA) or the Medtronic CoreValve (Medtronic, Minneapolis, MN, USA) prostheses in our cardiology department were included in a prospective single-centre registry. All procedures were performed through retrograde transfemoral, subclavian, or retroperitoneal or anterograde transapical approaches, under local or general anaesthesia .
Patients were considered candidates for TAVI based on a Logistic EuroSCORE ≥ 20% or because of frailty, presence of severe comorbidities or contraindication to surgical aortic valve replacement. The decision to perform a TAVI was validated by our medico-surgical heart team. All patients provided signed informed consent for subsequent data collection and analysis for research purposes.
Clinical and echocardiographic characteristics
All patients were initially screened in order to confirm the diagnosis of severe symptomatic aortic stenosis. Severe aortic stenosis was defined as an aortic valve area < 1.0 cm 2 and/or a mean transvalvular gradient > 40 mmHg or a peak velocity > 4 m/sec . Data concerning history of coronary artery disease, peripheral artery disease, cerebrovascular stroke, renal impairment or chronic lung disease and previous cardiac surgery were collected. Porcelain aorta was defined as an extensive circumferential calcification of the ascending aorta. A comprehensive transthoracic echocardiography was performed before TAVI.
Risk score calculations
Predicted operative mortality was calculated using the EuroSCORE II, the Logistic EuroSCORE and the STS score. For each patient, the Logistic EuroSCORE and the STS score were calculated prospectively using web-based systems. The EuroSCORE II was calculated retrospectively, based on a prospective data collection of items, using an online calculator .
Statistical analysis
Variables are expressed as mean ± standard deviation (SD) or number (%). Comparisons between scores were performed using paired t -tests and linear regressions. The diagnostic value of the EuroSCORE II for the diagnosis of high operative risk (defined as Logistic EuroSCORE ≥ 20% or STS score ≥ 10%) was analyzed. Sensitivity, specificity, and positive and negative predictive values were determined for various thresholds of the EuroSCORE II. A contingency analysis was performed to assess the agreement between the different scores with regards to risk classification for surgery and expressed by kappa values. A high kappa value demonstrates that the risk classification of patients is close or equivalent using the two scoring systems. Statistical analyses were performed using JMP 7 software (SAS institute, Cary, NC, USA). P < 0.05 was considered statistically significant.
Results
Patient characteristics
From October 2006 to June 2011, 272 consecutive patients underwent a TAVI in the cardiology department of Bichat Hospital, Paris. Demographic, clinical and echocardiographic characteristics of the population are summarized in Table 1 . Briefly, mean age was 82 ± 9 years and 44% of patients were female. All patients had severe aortic stenosis. A large proportion of patients had severe symptoms, including 228 patients (84%) in New York Heart Association (NYHA) class III/IV. TAVI was performed using the transfemoral approach in 166 patients (61%), transapical in 88 patients (32%), subclavian in 15 (6%) and retroperitoneal in 3 patients (1%); under general anaesthesia in 202 patients (74%) or local anaesthesia in 70 patients (26%).
| All patients ( n = 272) | |
|---|---|
| Age (years) | 82 ± 9 |
| Female | 119 (44) |
| History of smoking | 91 (33) |
| Hypertension | 204 (75) |
| Diabetes mellitus | 69 (25) |
| Insulin-dependent diabetes mellitus | 17 (6) |
| Hypercholesterolaemia | 141 (52) |
| Coronary artery disease | 119 (44) |
| Prior heart valve surgery | 15 (6) |
| Prior cerebrovascular accident | 26 (10) |
| Carotid artery stenosis | 50 (18) |
| Peripheral artery disease | 46 (17) |
| Chronic pulmonary obstructive disease | 78 (29) |
| Atrial fibrillation | 103 (38) |
| Creatinine clearance (mL/min) | 47 ± 24 |
| Renal failure | 80 (29) |
| Porcelain aorta | 40 (15) |
| NYHA class III/IV | 228 (84) |
| Canadian Cardiovascular Society class ≥ 2 | 53 (19) |
| Mean aortic gradient (mmHg) | 51 ± 17 |
| Aortic valve area (cm 2 ) | 0.73 ± 0.19 |
| Left ventricular ejection fraction < 50% | 100 (37) |
| Systolic pulmonary artery pressure (mmHg) | 50 ± 15 |
Comparison between the EuroSCORE II, the Logistic EuroSCORE and the STS score
The mean Logistic EuroSCORE was 23 ± 14%, STS score 10 ± 9% and EuroSCORE II 9 ± 8%. The EuroSCORE II was significantly lower than the logistic EuroSCORE ( P < 0.01), but not significantly different to the STS score ( P = 0.10). The mean EuroSCORE II was significantly higher in patients with a Logistic EuroSCORE ≥ 20% ( n = 158 [58%]) than in those with a Logistic EuroSCORE < 20% ( n = 114) (12% vs. 5%; P < 0.0001). Similarly, the mean EuroSCORE II was significantly higher in patients with an STS score ≥ 10% ( n = 101 [37%]) than in those with an STS score < 10% ( n = 171) (12% vs. 8%; P = 0.0001). However, the EuroSCORE II was only moderately correlated with the Logistic EuroSCORE ( r = 0.61, P < 0.001) ( Fig. 1 A) and poorly with the STS score ( r = 0.25, P < 0.001) ( Fig. 1 B).
