- Email: [email protected]
COMBINED ANATOMICAL AND CLINICAL FACTORS FOR THE RISK STRATIFICATION OF PATIENTS UNDERGOING PERCUTANEOUS CORONARY INTERVENTION: THE LOGISTIC CLINICAL SYNTAX SCORE
E1374 JACC March 27, 2012 Volume 59, Issue 13
Chronic CAD/Stable Ischemic Heart Disease COMBINED ANATOMICAL AND CLINICAL FACTORS FOR THE RISK STRATIFICATION OF PATIENTS UNDERGOING PERCUTANEOUS CORONARY INTERVENTION: THE LOGISTIC CLINICAL SYNTAX SCORE ACC Oral Contributions McCormick Place North, N227b Saturday, March 24, 2012, 8:15 a.m.-8:30 a.m.
Session Title: Sex, Scoring, Sadness, Statins, Stents and Surgery Abstract Category: 2. Chronic CAD/Stable Ischemic Heart Disease: Clinical Presentation Number: 910-4 Authors: Vasim Farooq, Yvonne Vergouwe, Lorenz Raber, Pascal Vrancx, Hector Garcia, Salvatore Brugaletta, A. Kappetein, Marie-angèle Morel, Ton de Vries, Michael Swart, Keith Dawkins, Marco Valgimigli, Stephan Windecker, Ewout W. Steyerberg, Patrick Serruys, Thorax Center, Rotterdam, The Netherlands Background: The SYNTAX Score (SXscore) is an important prognostic factor for PCI patients. The incorporation of clinical factors may further augment the utility of the SXscore to risk stratify individual patients. Methods & Results: Patient-level pooled data from 7 contemporary coronary stent trials (LEADERS, RESOLUTE, SIRTAX, SYNTAX, ARTS II, STRATEGY & MULTISTRATEGY) were used to develop the Logistic Clinical SXscore (LCSS) (Fig.1) - to predict 1-year risk of all-cause Death & MACE. A Core Model (composed of the SXscore, age, creatinine clearance & left ventricular ejection fraction) & Extended Model (incorporating the Core Model & best performing clinical variables) were developed & cross-validated by the omission of each of the coronary stent trials in turn. The LCSS demonstrated significantly more predictive ability for 1-year all-cause Death, with an area under the receiver operator curve (AUC) for the Core & Extended Models of 0.75 & 0.79 respectively, compared to 0.66 for the SXscore in isolation. Validation plots demonstrated a good agreement between predicted & observed risks. For 1-year MACE the SXscore & LCSS demonstrated comparable discriminative abilities (AUC: SXscore 0.61, Extended Model 0.62), secondary to the SXscore being the predominant determinant of all-cause revascularisation. Conclusion: Compared to the SXscore the LCSS significantly enhances the risk stratification of PCI patients for Death at 1-year, and allows for a more personalised assessment of long-term patient risk.
Chronic CAD/Stable Ischemic Heart Disease COMBINED ANATOMICAL AND CLINICAL FACTORS FOR THE RISK STRATIFICATION OF PATIENTS UNDERGOING PERCUTANEOUS CORONARY INTERVENTION: THE LOGISTIC CLINICAL SYNTAX SCORE ACC Oral Contributions McCormick Place North, N227b Saturday, March 24, 2012, 8:15 a.m.-8:30 a.m.
Session Title: Sex, Scoring, Sadness, Statins, Stents and Surgery Abstract Category: 2. Chronic CAD/Stable Ischemic Heart Disease: Clinical Presentation Number: 910-4 Authors: Vasim Farooq, Yvonne Vergouwe, Lorenz Raber, Pascal Vrancx, Hector Garcia, Salvatore Brugaletta, A. Kappetein, Marie-angèle Morel, Ton de Vries, Michael Swart, Keith Dawkins, Marco Valgimigli, Stephan Windecker, Ewout W. Steyerberg, Patrick Serruys, Thorax Center, Rotterdam, The Netherlands Background: The SYNTAX Score (SXscore) is an important prognostic factor for PCI patients. The incorporation of clinical factors may further augment the utility of the SXscore to risk stratify individual patients. Methods & Results: Patient-level pooled data from 7 contemporary coronary stent trials (LEADERS, RESOLUTE, SIRTAX, SYNTAX, ARTS II, STRATEGY & MULTISTRATEGY) were used to develop the Logistic Clinical SXscore (LCSS) (Fig.1) - to predict 1-year risk of all-cause Death & MACE. A Core Model (composed of the SXscore, age, creatinine clearance & left ventricular ejection fraction) & Extended Model (incorporating the Core Model & best performing clinical variables) were developed & cross-validated by the omission of each of the coronary stent trials in turn. The LCSS demonstrated significantly more predictive ability for 1-year all-cause Death, with an area under the receiver operator curve (AUC) for the Core & Extended Models of 0.75 & 0.79 respectively, compared to 0.66 for the SXscore in isolation. Validation plots demonstrated a good agreement between predicted & observed risks. For 1-year MACE the SXscore & LCSS demonstrated comparable discriminative abilities (AUC: SXscore 0.61, Extended Model 0.62), secondary to the SXscore being the predominant determinant of all-cause revascularisation. Conclusion: Compared to the SXscore the LCSS significantly enhances the risk stratification of PCI patients for Death at 1-year, and allows for a more personalised assessment of long-term patient risk.