Correlates of Hemodynamic Instability During Non-Emergent Percutaneous Coronary Intervention: Refining High-Risk...

Previous studies using high-risk criteria to select patients for mechanical circulatory support (MCS) during percutaneous coronary intervention (PCI) have not consistently shown a benefit in reducing adverse outcomes. Identifying correlates for intra-procedural hemodynamic instability (HI) may improve patient selection for MCS.

Consecutive, hemodynamically stable patients undergoing non-emergent PCI between 2018 and 2022 were reviewed. High-risk patients, defined by left ventricular ejection fraction ≤ 35% with unprotected left main intervention or LVEF ≤ 35% with 3-vessel disease, were compared to the non-high-risk patients. The primary outcome was HI during PCI, a composite outcome defined by the occurrence of death, cardiac arrest, emergent MCS, or the need for sustained vasopressor support.

A total of 278 high-risk patients were compared to 2854 non-high-risk patients. The high-risk group was older with more comorbidities and poorer left ventricular ejection fraction (24.7% vs. 51.5%). The occurrence of HI was overall low but occurred more frequently in high-risk patients (4.3% vs. 2.2%, p = 0.025), mostly driven by sustained vasopressor need (75% vs. 66%, p = 0.023). Post-procedural adverse clinical events were more common in the high-risk group, including death (4.7% vs. 0.7%, p < 0.001). A predictive model for intraprocedural HI included: ejection fraction ≤ 25%, left main intervention, and atherectomy (AUC = 0.703), while 3-vessel disease, age and other clinical comorbidities were not strongly associated with HI.

The rate of HI during contemporary, non-emergent PCI is very low. While Traditional high-risk PCI criteria are associated with HI, prediction may be improved by including only very low EF, left main intervention and atherectomy. Further studies are needed to evaluate whether utilizing risk factors for HI could be a more effective strategy for selecting patients of MCS during PCI.