Objective: To examine causes of death not captured by current risk factor adjustment following coronary angioplasty.
Background: NYS tracks and reports in-hospital/30-day deaths during or following all PCIs as observed and risk adjusted rates. The decline in observed mortality [OMR] has been credited as due to public reporting improving care, but also as care denial. Whether risk adjustment influences selection and captures mortality risk is controversial.
Procedure: NYS uses a multivariate risk factor equation to determine the risk adjusted rate. Adverse predictors include: Odds ratio: F_1.65, unstable_7.97, low EF_1.47 - 6.72, pre procedural MI_2.49- 8.03, chronic lung disease_2.04- 2.87, heart failure_1.87, PVD_1.87, renal failure_2.08-4.4, CAD extent_1.4. Emergent cases are higher risk than elective cases. At Stony Brook UH all PCI mortalities are reviewed to assess factors contributing to death. 4 yrs of deaths [unclassified preprocedure risks, major comorbidities contributing to death] are reported. Pts excluded from the NYS data base were excluded.
Results: 45 Patients died post PCI. Pre procedure risk factors not included as adjustments but that were felt to effect survival: severe aortic stenosis 13/45, coagulopathy 8/45, severe anemia 6/45, CTS turn downs 6/45, cancer not in remission 4/45, active systemic infection 4/45, post cardiac arrest not in coma 2/45, respiratory failure, active GI bleed, end stage liver disease [each 1/45]. PCI complications with adverse outcomes: coronary dissection 2/45, perforation 4/45, pericardial tamponade 4/45. Unsuccessful PCI 10/45 and need for IABP/Impella support 26/45 were associated with adverse outcomes. Post PCI diagnoses associated with mortality: CVA, hypoxic encephalopathy, intractable seizures 9/45, new diagnosis of cancer 2/45, sepsis 6/45 and pneumonia 4/45, persistent shock 16/45 and multi organ failure 12/45, malignant arrhythmias and PEA 12/45, anemia & GI bleeding 14/45, contrast induced nephropathy/AKI 13/45, lower extremity ischemia/bleeding 5/45, mechanical complications of MI 3/45. The decision to make the pt DNR/DNI 18/45 or comfort care/hospice 15/45 impacted on delivery of care and the likelihood of survival, particularly if only rescinded for the PCI.
Conclusions: Coronary PCI may be life saving. However, procedural urgency may limit knowledge of and optimization of risks. When possible, a more holistic discussion with the pt and family, giving careful consideration to the risks and benefits of angioplasty, goals of therapy and limitations of care should take place pre procedure. It is also important to be able to identify and aggressively treat the post procedure complications commonly associated with mortality.