Evaluating Risk Factors for Acute Kidney Injury after Catheterization in Critically Ill Infants with Congenital Heart Disease

Wednesday, May 22, 2019
Belmont Ballroom 2-3 (The Cosmopolitan of Las Vegas)
Reid C Chamberlain, M.D. , Duke University Hospital, Durham, NC
Melissa Ross , Duke University Medical School, Durham, NC
Dylan Eiger , Duke University Medical School, Durham, NC
Kevin D Hill, M.D. , Duke University Hospital, Durham, NC
Gregory A Fleming, M.D. , Duke University Medical Center, Durham, NC

Background:
Critically ill infants with congenital heart disease often require cardiac catheterization. Acute kidney injury (AKI) is a potential post-procedural complication and AKI is consistently associated with adverse outcomes in this high-risk patient population. Little is known about AKI incidence and risk factors after catheterization.

Methods:
Retrospective, single-institution study of all patients <1 year who underwent catheterization and admitted to an intensive care unit (2013-2018). Chart review performed for patient and procedural characteristics included creatinine levels for 5 days post-procedure. AKI was defined using Kidney Disease: Improving Global Outcomes (KDIGO) criteria. Factors associated with AKI after catheterization were determined using multivariate logistic regression.

Results:
Characteristics of 388 catheterizations performed on 223 patients are summarized in Table 1. Overall post-procedure AKI rates were 27.1% with 12.1% having ≥ stage 2 AKI. In multivariate analysis, risk factors for ≥ stage 2 AKI included increased weight (adjusted odds ratio [AOR] 1.24, 95% CI 1.04-1.48) and need for inotropic support at time of catheterization (AOR 2.21, 95% CI 1.06-4.62). Contrast volume and prior surgery were not associated with AKI in univariate or multivariate analysis.

Conclusions:
In critically ill infants who undergo catheterization, pre-procedure risk factors can help predict significant post-procedure AKI. Future studies with greater variability in contrast volumes administered will help improve the predictive AKI model and validate associations between AKI and serious adverse outcomes.