Abstract 1934: PCI Failure in the Occluded Artery Trial (OAT): Frequency, Mechanisms, and Outcome
Background: The Occluded Artery Trial (OAT) was predicated on achieving infarct-related coronary patency in PCI assigned pts. Eligibility defined an anatomically complex cohort (occlusions up to 28 d post-MI). Operators from 202 sites in 25 countries participated. We report procedural outcomes and describe mechanisms of PCI failure.
Methods: Of 2201 enrolled (2166 ± 35 substudy), 1101 were assigned PCI of which 1093 (99.3%) underwent core angio analysis using standard quantitative and qualitative variables. Technical reasons for PCI failure were adjudicated (site-reported PCI data were used in 8 cases not submitted for core analysis). Clinical and angio factors associated with PCI failure were modeled with logistic regression (significance - p<0.01, 20 analysis).
Results: PCI was attempted in 1090 of 1101 (99.0%) pts; stenting was attempted in 961 of 1090 (88.2%). A GPI was administered in 727 (66.7%). Baseline TIMI flow was grade 0 in 893/1080 (82.7%). Collateral flow was visible in 945/1076 (87.8%). PCI was successful in 953 of 1090 attempts (87.4%) and was similar in US (88.0%) and non-US sites (87.3%). Mechanisms of PCI failure available in 135 of 137 cases included failure to cross with wire (57%), inability to re-enter true lumen beyond occlusion (25%), residual stenosis ≥50% (20%), and extensive thrombus (10%). Of 50 characteristics examined, those independently predictive of failure were prior PCI (p<0.001) and TIMI grade 0 (p=0.007) with trends noted for Killip class ≥2 (p=0.013) and male gender (p=0.04). Life-table estimates of death, MI or class IV HF at 60 months was similar after PCI success and failure (18.3% vs 18.6%, p=NS).
Conclusions: In the largest trial to date testing PCI in occluded coronary arteries, success rates were similar in both US and non-US sites. As in chronic total occlusions, PCI failure was most commonly due to inability to cross the occluded segment with a guide wire or the inability to re-enter the true lumen once beyond the occluded segment.