Predictors of Short-Term Readmissions in Patients with Coronary Artery Disease Treated with Percutaneous Coronary Intervention

Background: Readmission following percutaneous coronary intervention (PCI) is associated with poorer outcomes and increased healthcare costs. Identifying predictors of readmission may help establish preventative interventions. Methods: We retrospectively reviewed data for 1216 patients treated with PCI at Western Health between April 2013 and October 2016 enrolled in the Victorian Cardiac Outcomes Registry. Baseline, clinical and procedural characteristics were analysed using univariate and multivariate regression models to compare the group of patients readmitted at 30 days to those who were not readmitted. Results: The readmission rate was 11.7%. Compared to non-readmitted group (n = 1073), post PCI readmitted patients (n = 143) were more likely to have had ACS as initial presentation (100% vs. 71.2%, p < 0.001), out-of-hospital cardiac arrest (7% vs.1.6%, p < 0.001), cardiogenic shock (5.6% vs. 2.3%, p < 0.024), moderate to severe LV dysfunction (100% vs. 5.6%, p < 0.001), and higher peak CK (1833.3 ± 1659.3 vs. 644.1 ± 1246.5, p < 0.001). Adjunct device and thrombus aspiration during PCI were used more commonly in the readmitted group compared to the non-readmitted (23.1% vs. 6.3%, p < 0.001) and (23.1% vs. 3.8%, p < 0.001) respectively. Surprisingly, diabetes was more common in the non-readmitted group (25.8% vs. 17.5%, p = 0.03). Multivariable analysis identified STEMI (HR 47.6, 11.5 - 200, p < 0.001) and inter-hospital transfer (HR 2.14, 1.12 - 4.08, p < 0.021) to be independent predictors of 30-day re-hospitalisation. Conclusions: Patients post PCI who present initially with STEMI or who are transferred from other hospitals have significant higher risk of readmission at 30 days compared to other patients. Intensive follow-up strategies may ameliorate readmissions and requires further study.