| Background:As percutaneous coronary intervention(PCI) techniques improved, the scope of those patients who suitable for the treatment of PCI has expanded rapidly, and a growing number of elderly patients can also benefit from this technique. However, elderly patients tend to have extensive comorbidities and complex coronary lesions, which can lead to a higher rate of major adverse cardiovascular and cerebrovascular events(MACCE). Therefore, to identify the risk factors associated with MACCE in elderly patients after PCI and to construct a risk assessment model for PCI elderly patients plays a significant role for physicians in making a comprehensive assessment of PCI patients and carrying out risk management for PCI patients actively.Objective:To investigate and identify the risk factors associated with MACCE in elderly patients after PCI and to construct a risk assessment model of PCI patients.Methods:A total of 1007 consecutive elderly patients over 75 years old with coronary heart disease who underwent PCI in our hospital from January 2005 to December 2010 were involved as the derivation-set. Univariate and multivariate logistic regression analysis were performed to determine the independent risk factors associated with MACCE then a multivariate logistic regression model was constructed to predict MACCE after PCI for elderly patients. To develop a simple risk assessment model, a risk prediction score was first to be constructed. The risk factors identified through multivariate modeling were assigned an integer coefficient, integers were chosen to be approximately proportional to the estimated continuous coefficients from the logistic model. Then a risk score-probability curve was drawn, risk-stratified was divided. The final risk assessment model was validated by the 527 elderly patients over 75 years old in the validation-set who underwent PCI in our hospital between January 2011 and December 2012. The area under the receiver operating characteristic(ROC) curve and the Hosmer-Lemeshow goodness of fit statistic were calculated to assess the performance and calibration of this risk assessment model.Results:(1) Of the 1007 patients in the derivation-set, 40 patients(4.0%) had MACCE; of the 527 patients in the validation-set, 17 patients(3.2%) had MACCE after the procedure;(2) Multivariate logistic regression analysis indicated that 7 independent predictors with associated risk weights in parentheses were as follows: urgent PCI(3 scores), renal insufficiency(2 scores), left main disease(2 scores), diabetes mellitus(1 score), acute myocardial infarction(1 score), type C lesion(1score), and ≥3 stents placed(1 score).(3) The point values for risk factor scores range from 1 scores to 3 scores, and the total risk factor scores range from 0 score to 11 scores. With the risk factor scores increasing, the estimated rates of MACCE increase significantly. The predicted probabilities of MACCE for each risk score ranged from 0.002 for a patient with a score of 0 to 0.967 for a patient with the highest possible scores of 11.(4) Based on the integer risk score, the total risk factor scores ? 2 were classified as low risk, the total risk factor scores 3~5 were classified as moderate risk, the total risk factor scores ? 6 were classified as high risk, Observed rates of MACCE in these strata were 0.8% for low risk, 8.9% for moderate risk, 46.3% for high risk.(5) In both the derivation-set and validation-set, the model had a good performance in terms of discrimination(the area under ROC curve was 0.91 and 0.89, respectively) and calibration(Hosmer-Lemeshow P-value 0.658 and 0.586, respectively).Conclusion:We have developed a risk assessment model to predict and assess in-hospital MACCE in elderly patients after PCI. This model has been validated that it has a good performance in terms of discrimination and calibration. It is also a simple, useful tool for physicians to implement in clinical practice to estimate the in-hospital MACCE risk after PCI and to prevent MACCE. |
PDF Full Text Request