Acute Kidney Injury in Extracorporeal Membrane Oxygenation Patients: National Analysis of Impact of Age

Abstract

<b><i>Background:</i></b> The aim of this study was to determine epidemiology and outcomes of acute kidney injury (AKI) in patients on extracorporeal membrane oxygenation (ECMO) and to assess if age modifies the effect of AKI on mortality. <b><i>Methods:</i></b> Using National (Nationwide) Inpatient Sample Database for hospitalizations in the USA from 2003 to 2014, we identified adult patients on ECMO support. Using International Classification of Diseases 9th Revision, we assessed the rates of AKI and AKI requiring dialysis (AKI-D) among them and associated survival. We used a multivariable logistic regression to identify risk factors of and differential effect of age on mortality from AKI. <b><i>Results:</i></b> AKI was seen in 63.9% of 17,942 ECMO hospitalizations: 21.9% of those with AKI required dialysis. The percentage of those with AKI increased steadily. Mortality was higher in those with AKI, with highest in those with AKI-D (70.8% vs. 61.7%; <i>p</i> < 0.001). While both age and AKI were independent predictors of mortality, age was neither a risk factor for AKI nor did it modify the effect of AKI on mortality. <b><i>Conclusions:</i></b> AKI is common and is increasing among patients on ECMO support. Patients on ECMO have high mortality and AKI is an independent predictor of mortality. Though age is also an independent predictor of mortality in patients on ECMO, it is neither a predictor of AKI nor does not modify the relationship between AKI and mortality.

Health Disparities in Extracorporeal Membrane Oxygenation Utilization and Outcomes: A Scoping Review and Methodologic Critique of the Literature*

“…Overall, 54 studies (58%) found no statistical association between race and/or ethnicity and outcome (37–76), while 36 (39%) observed adverse outcomes in populations from underrepresented/diverse racial or ethnic backgrounds (Black/African American, Asian, Hispanic, or “other”/“non-Caucasian”/“non-White”). Of citations identifying adverse outcomes, 8 (22%) used univariate (77–84) and 28 (78%) used multivariate analyses. Of multivariate analyses, diverse populations, compared with nondiverse/White patients, were independently associated with mortality (26, 27, 29, 30, 32, 35, 85–100), adverse neurologic outcomes (101), or prolonged ECMO duration (34).…”

“…Eight studies reported improved survival in patients with private relative to government insurance (28, 77, 88, 89, 106, 123, 128, 129). Eight found no survival difference (55, 65, 81, 92, 130–133), with variable cost, complication, readmission, and LOS results (31, 33, 56, 70, 125, 126, 128, 134), including one with higher renal replacement in privately insured patients (83).…”