Accuracy of the PARIS score and PCI complexity to predict ischemic events in patients treated with very thin stents in unprotected left main or coronary bifurcations

Abstract

Background The PARIS risk score (PARIS‐rs) and percutaneous coronary intervention complexity (PCI‐c) predict clinical and procedural residual ischemic risk following PCI. Their accuracy in patients undergoing unprotected left main (ULM) or bifurcation PCI has not been assessed. Methods The predictive performances of the PARIS‐rs (categorized as low, intermediate, and high) and PCI‐c (according to guideline‐endorsed criteria) were evaluated in 3,002 patients undergoing ULM/bifurcation PCI with very thin strut stents. Results After 16 (12–22) months, increasing PARIS‐rs (8.8% vs. 14.1% vs. 27.4%, p < .001) and PCI‐c (15.2% vs. 11%, p = .025) were associated with higher rates of major adverse cardiac events ([MACE], a composite of death, myocardial infarction [MI], and target vessel revascularization), driven by MI/death for PARIS‐rs and target lesion revascularization/stent thrombosis for PCI‐c (area under the curves for MACE: PARIS‐rs 0.60 vs. PCI‐c 0.52, p‐for‐difference < .001). PCI‐c accuracy for MACE was higher in low‐clinical‐risk patients; while PARIS‐rs was more accurate in low‐procedural‐risk patients. ≥12‐month dual antiplatelet therapy (DAPT) was associated with a lower MACE rate in high PARIS‐rs patients, (adjusted‐hazard ratio 0.42 [95% CI: 0.22–0.83], p = .012), with no benefit in low to intermediate PARIS‐rs patients. No incremental benefit with longer DAPT was observed in complex PCI. Conclusions In the setting of ULM/bifurcation PCI, the residual ischemic risk is better predicted by a clinical risk estimator than by PCI complexity, which rather appears to reflect stent/procedure‐related events. Careful procedural risk estimation is warranted in patients at low clinical risk, where PCI complexity may substantially contribute to the overall residual ischemic risk.

Predictors of target lesion failure after treatment of left main, bifurcation, or chronic total occlusion lesions with ultrathin‐strut drug‐eluting coronary stents in the ULTRA registry

“…Yet, this finding should be interpreted in the context of the ULTRA study, in which all patients underwent high‐risk PCI procedures. A previous study showed that the classification of patients based on the standard definition of complex PCI has a modest accuracy for predicting adverse outcomes in high‐risk cohorts, where clinical risk estimators may be more accurate in predicting residual ischemic risk 32 . Similarly, in patients treated for coronary bifurcation lesions, we confirm the positive predictive value of both main and side branch lesion length along with main branch stenosis diameter and (smaller) side branch diameter in patients treated with ultrathin‐strut DES 18 .…”

Prediction of All-Cause Mortality Following Percutaneous Coronary Intervention in Bifurcation Lesions Using Machine Learning Algorithms

“…A multidimensional approach may be highly relevant in this setting, which is characterized by high anatomical and procedural complexity. Indeed, previous scores to predict adverse events focusing on either patient-related or anatomy/procedure-related factors performed only modestly in patients with bifurcation PCI [ 28 ]. Specifically, in a previous analysis of the RAIN registry, the PCI complexity definition proposed by Giustino et al [ 11 ] using validated and guideline-endorsed criteria (reflecting anatomy/procedure-related factors) was unable to discriminate post-procedural mortality (AUC 0.49), and the PARIS risk score [ 8 ] (reflecting patient-related factors) displayed only a modest discrimination capacity (AUC 0.65).…”

“…Specifically, in a previous analysis of the RAIN registry, the PCI complexity definition proposed by Giustino et al [ 11 ] using validated and guideline-endorsed criteria (reflecting anatomy/procedure-related factors) was unable to discriminate post-procedural mortality (AUC 0.49), and the PARIS risk score [ 8 ] (reflecting patient-related factors) displayed only a modest discrimination capacity (AUC 0.65). More importantly, these tools showed potential for an accurate event prediction when combined, thus suggesting that a comprehensive evaluation of clinical, anatomical, and procedural features may better reflect residual risk [ 28 ].…”

Prognostic and Practical Validation of ESC/EACTS High Ischemic Risk Definition on Long-Term Thrombotic and Bleeding Events in Contemporary PCI Patients

“…Several studies linked PCI procedural complexity features with a higher risk of subsequent adverse clinical events [2][3][4][5] . To facilitate risk stratification after PCI, the concept of complex PCI has been developed in the setting of using heterogeneous cohorts and different definitions, resulting in a variety of clinical outcome results [6][7][8][9][10][11][12] . Therefore, ambiguity about how to effectively identify patients at high ischemic risk (HIR) remains, and it is reflected in the varying inclusion criteria adopted by clinical trials in such patients, which may limit the interpretation, generalizability, and pooling of reported data 13,14) .…”