TEMPORAL TRENDS IN ACUTE KIDNEY INJURY INCIDENCE AND IN-HOSPITAL MORTALITY AFTER CORONARY REVASCULARIZATION: A UNITED STATES STUDY

Coronary artery bypass grafting (CABG) and percutaneous coronary intervention (PCI) both carry high risk of acute kidney injury (AKI). Our previous studies have shown that CABG was associated with higher incidence of post-procedural AKI compared with PCI, and in-hospital mortality after CABG was higher than that of PCI. Since the temporal trends of AKI incidence and associated mortality have been changing, it is important to study these temporal changes which could help us better understand the risk profile migration over the years in the coronary revascularization modalities.

We generated a propensity-matched cohort of 274,464 hospitalizations that had first-time CABG or PCI for multivessel coronary disease in 2004 to 2012 from the National Inpatient Sample (NIS). Patients who received concomitant valvular repair or both CABG and PCI during the same admission, history of organ transplant, CKD stage V, or ESRD on dialysis were excluded. Both groups were propensity score matched for age, gender, race, payer, prior MI, unstable angina, heart failure, CVA, peripheral arterial disease, valvular disease, atrial flutter or fibrillation, CKD, diabetes, HTN, dyslipidemia, smoking, cirrhosis, COPD, systemic cancer, obesity, and anemia. The odds ratios were estimated by the random intercept logistic regression model.

The temporal trends of AKI incidence in both CABG and PCI groups had been increasing over the years from 5.9% in 2004 up to 14.2% in 2012 for CABG and from 2.7% in 2004 up to 8.8% in 2012 for PCI. Compared with PCI, CABG was associated with 1.7 to 3.2 times higher incidence of post-procedural AKI in each individual year from 2004 to 2012. Although CABG had higher likelihood to develop AKI throughout the study period than PCI, the odds had been decreasing gradually (OR 3.29 in 2006; OR 1.73 in 2012). The temporal trends of in-hospital mortality in CABG-AKI group had been decreasing from 16.5% in 2004 to 6.5% in 2012 whereas in the PCI-AKI group, in-hospital mortality remained stable, 14.1% in 2004 to 13.1% in 2012. Compared to PCI-AKI group, the likelihood of in-hospital death for CABG-AKI group in 2004 was 20% higher, but after 2004, the odds reversed. From 2005 to 2012, the odds of in-hospital death in patients with post-CABG AKI became 23%–54% lower than the PCI-AKI group. When overall in-hospital mortality was compared irrespective of kidney function, CABG was associated with higher mortality from 2004–2010, but in 2011 and 2012, PCI was associated with higher mortality.

Both CABG and PCI were associated with increasing temporal trends in AKI incidence. CABG was associated with a higher incidence of post-procedural AKI and in-hospital mortality in earlier years. However, over time, CABG-AKI showed a decreasing trend in in-hospital mortality, whereas the mortality rate in PCI-AKI remained stable. The increasing comorbidity burden in the PCI population, improvements in CABG techniques, and decline in CABG volume compared to PCI could contribute to these observations. Further study is needed to analyze the factors influencing these changing trends in AKI and mortality in coronary revascularization over time in this United States study. This study was previously presented at the American Society of Nephrology (ASN) Kidney Week meetings in 2018. Re-submission of this combined abstract is permitted by the organizers of the original meetings, and since then, additional information has been analyzed and will be presented at this meeting.