Development and Validation of Elective and Nonelective Risk Prediction Models for In-Hospital Mortality in Proximal Aortic Surgery Using the National Institute for Cardiovascular Outcomes Research (NICOR) Database.

BACKGROUND: To facilitate patient choice and the risk adjustment of consultant outcomes in aortic operations, reliable predictive tools are required. Our objective was to develop a risk prediction model for in-hospital mortality after operation on the proximal aorta. METHODS: Data for 8641 consecutive UK patients undergoing proximal aortic operation from the National Institute for Cardiovascular Outcomes Research database from April 2007 to March 2013 were analyzed. Multivariable logistic regression was used to identify independent predictors of in-hospital mortality. Model calibration and discrimination were assessed. RESULTS: In-hospital mortality was 4.6% in elective operations and 16.5% in nonelective operations. In the elective model, previous cardiac operation (adjusted odds ratio [OR] 4.1, 95% confidence interval [CI]: 3.0 to 4.7) and ejection fraction greater than 30% (adjusted OR 2.3, 95% CI: 1.7 to 3.1) were the strongest predictors of mortality (p < 0.001). The area under the receiver operating characteristic (AUROC) curve was 0.805 (95% CI: 0.802 to 0.807) with a bias-corrected value of 0.795. Model calibration was acceptable (p = 0.427) on the basis of the Hosmer-Lemeshow goodness-of-fit test. In the nonelective model, salvage operations (adjusted OR 9.9, 95% CI: 6.5 to 15.2) and previous cardiac operation (adjusted OF 3.9, 95% CI: 3.0 to 5.0) were the strongest predictors of mortality (p < 0.001). The AUROC curve was 0.761 (95% CI: 0.761 to 0.765) with a bias-corrected value of 0.756, and model calibration was also found to be acceptable (p = 0.616). CONCLUSIONS: We propose the use of these risk models to improve patient choice and to enhance patients' awareness of risks and risk-adjust aortic operation outcomes for case-mix.

Predicting angiographic outcome in contemporary percutaneous coronary intervention: a lesion-specific logistic model.

BACKGROUND: Previous angiographic lesion classification systems were derived from analysis of outcomes and lesion complexity in the early stent era. Advances in equipment design and techniques have altered the association between lesion and target vessel characteristics and procedural outcome in modern percutaneous coronary intervention (PCI). We evaluated the precise relationship between lesion characteristics and technical outcome on a lesion by lesion basis in a large dataset. We developed a multivariate model to predict technical failure in PCI. METHODS: Analysis of prospectively collected data on 10,800 lesions in 6,719 consecutive PCI cases between January 2000 and December 2004. Multivariate logistic regression was undertaken to identify predictors of angiographic outcome at each treated lesion (success/failure). Statistical model validation was carried out using data from a further 3,340 treated lesions in 1,940 consecutive cases. RESULTS: Independent variables associated with an increased risk of technical failure included total occlusion, severe calcification, proximal vessel tortuosity >90 degrees, lesion in a degenerate vein graft, and lesion angulation > or =90 degrees. The receiver operating characteristics (ROC) curve for the predicted probability of technical failure was 0.85. Failure occurred in 2.2% of treated lesions in the validation set (ROC curve 0.82, model predicted 2.5%). CONCLUSIONS: We have re-evaluated the association between lesion characteristics and technical outcome in modern PCI. We have thereby developed a contemporary prediction model for angiographic outcome at each treated lesion.

Impact of periprocedural creatine kinase-MB isoenzyme release on long-term mortality in contemporary percutaneous coronary intervention.

OBJECTIVE: To evaluate the incidence of periprocedural creatine kinase-MB (CK-MB) release and its impact on longterm mortality in contemporary percutaneous coronary intervention (PCI) at a tertiary referral center. METHODS: Retrospective analysis of 4,958 patients undergoing PCI with deployment of at least 1 stent at our center between January 1, 2003 and December 31, 2005. Patients admitted with acute ST-elevation myocardial infarction or cardiogenic shock (n = 617), and patients with no available CK-MB levels (n = 477) were excluded, leaving 3,864 patients for analysis. The outcome measure was all-cause mortality obtained from the National Strategic Tracing Service with patients followed up to June 30, 2006 (mean follow up 22 months). The association between CK-MB level and mortality was examined using Cox proportional hazards analysis. RESULTS: CK-MB elevation above the upper limit of normal (ULN) was detected in 29.4% patients. A total of 127 deaths were observed during follow up. By multivariate analysis, periprocedural CK-MB was independently associated with an increased risk of death (adjusted hazard ratio for every 10 units: 1.09; 95% CI: 1.05-1.12; p < 0.001). The relationship between the level of CK-MB and mortality was further examined by applying strata of CK-MB levels to the multivariate analysis (adjusted hazard ratio: 1.30, 1.76 and 2.26 for CK-MB levels of 1-3, 3-5 and > 5 the ULN, respectively). CONCLUSION: In the current era of PCI, periprocedural myonecrosis, evidenced by CK-MB elevation, is common and is associated with less favorable long-term mortality.

A study comparing the incidence and predictors of creatine kinase MB and troponin T release after coronary angioplasty. Does Clopidogrel preloading reduce myocardial necrosis following elective percutaneous coronary intervention?

BACKGROUND: To investigate the incidence and associated factors for enzyme release following percutaneous coronary intervention comparing assessment with creatine kinase MB (CK-MB) and troponin T (TnT). METHOD: TnT and CK-MB were measured post procedure in a consecutive series of 933 patients undergoing elective percutaneous coronary intervention between 1/4/2003 and 1/5/2004 at a single regional cardiac centre. RESULTS: CK-MB level significantly correlated to TnT levels (R=0.747, p<0.001) and a CK-MB level of above 3 times the upper limit of the local reference range (>3 x ULN) was predicted with 95% sensitivity (48% specificity) at a TnT level of 0.11. Multivariate predictors of >3 x ULN CK-MB release for uncomplicated percutaneous coronary intervention (n=898) were multi-vessel angioplasty (OR=2.51, 95% CI=1.57 to 4.01; p<0.001), saphenous venous graft angioplasty (OR=5.5, 95% CI=1.94 to 13.00; p=0.005) and lack of Clopidogrel preloading (OR=2.02, 95% CI=1.30 to 4.38; p=0.027). CONCLUSIONS: TnT was found to be a sensitive although not a highly specific marker of CK-MB release. In this study a TnT level above a threshold of 0.11 would identify 95% of the prognostically important 3-fold CK-MB releases. Replacing the >3 x ULN CK-MB threshold with a TnT level of 0.1 ng/l following percutaneous coronary intervention would increase the apparent rate of myocardial infarction from 11% to 20%. Lack of Clopidogrel preloading was independently associated with a >3 x ULN CK-MB release following uncomplicated elective percutaneous coronary intervention.

Surgeon specific mortality in adult cardiac surgery: comparison between crude and risk stratified data.

OBJECTIVE: As a result of recent failures in clinical governance the government has made a commitment to bring individual surgeons' mortality data into the public domain. We have analysed a database to compare crude mortality after coronary artery bypass surgery with outcomes that were stratified by risk. DESIGN: Retrospective analysis of prospectively collected data. SETTING: All NHS centres in the geographical north west of England that undertake cardiac surgery in adults. PARTICIPANTS: All patients undergoing isolated bypass graft surgery for the first time between April 1999 and March 2002. MAIN OUTCOME MEASURES: Surgeon specific postoperative mortality and predicted mortality by EuroSCORE. RESULTS: 8572 patients were operated on by 23 surgeons. Overall mortality was 1.7%. Observed mortality between surgeons ranged from 0% to 3.7%; predicted mortality ranged from 2% to 3.7%. Eighty five per cent (7286) of the patients had a EuroSCORE of 5 or less; 49% of the deaths were in this lower risk group. A large proportion of the variability in predicted mortality between surgeons was due to a small but differing number of high risk patients. CONCLUSIONS: It is possible to collect risk stratified data on all patients undergoing coronary bypass surgery. For most the predicted mortality is low. The small proportion of high risk patients is responsible for most of the differences in predicted mortality between surgeons. Crude comparisons of death rates can be misleading and may encourage surgeons to practise risk averse behaviour. We recommend a comparison of death rates that is stratified by risk and based on low risk cases as the national benchmark for assessing consultant specific performance.