Derivation and validation of Pd/Pa in the assessment of residual ischemia post-intervention: A prospective all-comer registry.

ABSTRACT

BACKGROUND: Measurement of post-percutaneous coronary intervention (PCI) fractional flow reserve (FFR) demonstrates residual ischemia in a large percentage of cases deemed angiographically successful which, in turn, has been associated with worse long-term outcomes. It has recently been shown that a resting pressure index, Pd/Pa, has prognostic value post stenting, however, its diagnostic value relative to FFR post-PCI has not been evaluated. METHODS: The diagnostic accuracy of Pd/Pa in identifying ischemia (FFR≤0.80) pre- and post-PCI was evaluated. Three patient subsets were analyzed. A reference pre-PCI cohort of 1,255 patients (1,560 vessels) was used to measure the accuracy of pre-PCI Pd/Pa vs. FFR. A derivation post-PCI group of 574 patient (664 vessels) was then used to calculate the diagnostic accuracy of post-PCI Pd/Pa vs. FFR. A final prospective validation cohort of 230 patients (255 vessels) was used to test and validate the diagnostic performance of post-PCI Pd/Pa. RESULTS: Median Pd/Pa and FFR were 0.90 (IQR 0.90-0.98) and 0.80 (IQR 0.71-0.88) in the reference pre-PCI model, 0.96 (IQR 0.93-1.00) and 0.87 (IQR 0.77-0.90) in the post-PCI derivation model, and 0.94 (IQR 0.89-0.97) and 0.84 (IQR 0.77-0.90) in the post-PCI validation model respectively. There was a strong linear correlation between Pd/Pa and FFR in all three models (p < 0.0001). Using ROC analysis, the optimal Pd/Pa cutoff value to predict a FFR ≤ 0.80 was ≤0.92 (AUC 0.87) in the pre-PCI model, ≤0.93 (AUC 0.85) in the post-PCI derivation model, and ≤ 0.90 (AUC 0.91) in the post-PCI validation model. Using a hybrid strategy of post-PCI Pd/Pa and post-PCI FFR when necessary (25% patients), overall diagnostic accuracy was improved to 95%. CONCLUSIONS: Pd/Pa has excellent diagnostic accuracy for identifying ischemia post-intervention. Using a hybrid strategy of post-PCI Pd/Pa first, and FFR afterwards, if required, adenosine administration can be avoided in over 75% of physiologic assessments post intervention.