The Clinical Application Study Of Risk Prediction Tools For Branch Occlusion In Interventional Treatment Of Coronary Artery Bifurcation Lesions

Validation of the V-RESOLVE(Visual estimation for Risk prEdiction of Side branch OccLusion in coronary bifurcation interVEntion)score systemObjectives:V-RESOLVE score is a scoring system to predict the risk of side branch(SB)occlusion after stenting in the main vessel(MV)of coronary bifurcation lesions based on visual estimation of the angiographic characteristics.The developed score is needed to be evaluated its clinical value.This study sought to validate the V-RESOLVE score system in a new cohort.Methods:From January to June 2013,1,286 patients with 1,820 bifurcation lesions undergoing elective intervention with provisional strategy were included.Angiographic data before MV stenting were reviewed,and the V-RESOLVE score was calculated.SB occlusion was defined as any decrease in Thrombolysis In Myocardial Infarction(TIMI)flow grade or the absence of flow in the SB after MV stenting.The statistical performance of the prediction model was assessed by its discrimination,calibration and clinical usefulness.Results:SB occlusion occurred in 222(12.20%)of 1,820 bifurcation lesions.The discrimination of the V-RESOLVE score for the validation cohort was good(C-statistic:0.80,95%confidence interval[CI]0.77 to 0.84).Regarding the calibration performance,the calibration-in-the-large was-0.03(95%CI:-0.181-0.12),while the combined predictive effect was slightly enlarged(calibration slope:1.25,95%CI:1.081-1.41)and,mainly attributed to the stronger predictive effect of the diameter stenosis of the SB before MV stenting.Stratified by the V-RESOLVE score,the SB occlusion rate was significantly higher in the high-risk group(26.18%)than in the non-high-risk group(3.48%)(P<0.01).Conclusions:The V-RESOLVE score system is a useful tool to help risk prediction for SB occlusion and decision-making in coronary bifurcation intervention.Using Baseline Angiographic Results by Visual Estimation for Risk PrEdiction of Side Branch OccLusion in Coronary Bifurcation InterVEntion:The Baseline V-RESOLVE Score SystemObjectives:The V-RESOLVE score is a validated risk prediction tool for side branch occlusion in coronary bifurcation intervention.The V-RESOLVE score requires 6 angiographic characteristics for calculation,including 2 characteristics at baseline and 4 characteristics before stenting in the main vessel.We sought to modify the V-RESOLVE score to a risk prediction score for SB occlusion totally based on visual estimation of the baseline angiographic characteristics.Methods:We enrolled all the 1545 patients with 1601 lesions in our previous study for development of the Baseline V-RESOLVE score,and other consecutive 1286 patients with 1820 lesions for validation.The Baseline V-RESOLVE score was obtained by adapting the 2 angiographic characteristics before stenting to corresponding pre-procedural angiographic characteristics and calculating the score with referrence to the algorithm of the V-RESOLVE score.SB occlusion was defined as any decrease in Thrombolysis In Myocardial Infarction(TIMI)flow grade or the absence of flow in the SB after MV stenting.The diagnostic performance was assessed and compared to the V-RESOLVE score.Risk stratification of the Baseline V-RESOLVE score was determined and its clinical usefulness was compared to that of the V-RESOLVE score.Statistical simulation of 30 observers was performed to test the influence of intro-and inter-observer variability on the performance of the Baseline V-RESOLVE score.Results:A strong correlation was showed between the established Baseline V-RESOLVE score and the V-RESOLVE score(Spearman coefficient 0.84,P<0.01).In the development cohort,the areas under the receiver operating characteristic(ROC)curves were similar between the Baseline V-RESOLVE score and the V-RESOLVE score(0.74 vs.0.76,P = 0.19),and the Hosmer-Lemeshow test suggested satisfactory goodness-of-fit(P = 0.71).Categorized by the the Baseline V-RESOLVE score of 14 as the cutoff,the high-risk group of the Baseline V-RESOLVE score demonstrated significantly higher rate of SB occlusion,compared to the non-high-risk group(17.31%vs.4.74%,P<0.01).When compared the Baseline V-RESOLVE score to the V-RESOLVE score,the integrated discrimination index(IDI)was-0.0181(95%confidence interval[CI]:-0,0363-0.0001,P = 0.052),suggesting insignificant decreasing of the mean prediction accuracy for SB occlusion,and the net reclassification improvement(NRI)was-0.0334(95%CI:-0.1203,0.0535,P =0.509),indicating an trend of lower accuracy of the risk stratification.In simulations the results of the 30 observers indicated that the areas under the ROC curves ranged from 0.66 to 0.74,and higher rate of SB occlusion in the high-risk group than the non-high-risk group regarding each simulated observer(all P<0.01).In the validation cohort,the areas under the ROC curves of the Baseline V-RESOLVE score was comparable to that of the V-RESOLVE score(0.73 vs,0.74,P = 0.85).The calibration-in-the-large was 0.13(95%CI:0.02-0.32),and calibration slope was 1.03(95%CI:0.85-1.20).Stratified by the V-RESOLVE score,the SB occlusion rate was significantly higher in the high-risk group(22.50%)than in the non-high-risk group(39/222,6.05%)(P<0.01).Conclusions:The Baseline V-RESOLVE score system was shown to be a powerful tool for risk prediction of SB occlusion in coronary bifurcation intervention.The interventional strategy could be rationally determined with the Baseline V-RESOLVE score using the pre-procedural coronary angiographic results.