Parameters used to discontinue dialysis in acute kidney injury recovery: a survey of United States nephrologists.

BACKGROUND: Despite advances in the approach to cure acute kidney injury (AKI), including definition, classification and treatment methods, there are no standard criteria to withdraw dialysis in the setting of improving AKI. We conducted this survey to elucidate parameters that United States (US) nephrologists used to determine when to stop dialysis with improving renal function in AKI. We hypothesized that there would be a difference in approach to weaning a patient off dialysis based on years in practice or the number of cases of AKI treated per year. METHODS: This was an anonymous electronic survey of practicing nephrologists who treated AKI. Data was de-identified and analyzed using descriptive statistics. RESULTS: The commonest criteria used to stop dialysis when renal function improved was, in decreasing order of importance, resolution in oliguria (51%), resolution of volume overload (29%), improvement in serum creatinine (26.7%) and resolution of hyperkalemia (21%). The most common reasons for re-starting dialysis within 28 days did not show a specific trend but respondents (20%) reported re-starting if estimated glomerular filtration rates (eGFR) declined. There was no significant pattern in approach to withdrawing dialysis or resuming dialysis based on the number of years in nephrology practice. However, responses of nephrologists who saw more than 20 AKI patients/year were significantly different in stopping dialysis with clinical stabilization of blood pressure (p < 0.001), improvement in respiratory parameters (p = 0.005), improvement in pre-dialysis blood urea nitrogen (BUN) levels despite the same dose of dialysis (p = 0.05) and resolution of oliguria (p = 0.025) compared to those who saw fewer cases. CONCLUSION: Resolution of oliguria was the commonest factor used to help deciding to stop dialysis in improving AKI. However, considerable variation was noted among US nephrologists who participated in this survey, regarding what criteria they used to withdraw dialysis in the setting of improving AKI. These results call for more studies in withdrawing dialysis in the setting of AKI that could lead to guideline formulation.