Dr. Ambreen Gul (AG), from Dialysis Clinic, Inc. in Albuquerque, New Mexico, discusses her abstract for the National Kidney Foundation’s 2016 Spring Clinical Meetings (SCM16), Relationships of Serum and Dialysate Potassium Concentrations to Hospitalization and Mortality Among Hemodialysis Patients, with Dr. Kenar Jhaveri, AJKD Blog Editor.

AJKDblog: Why don’t you tell us a little about your research and abstract being presented at the NKF 2016 Spring Meetings?

AG: Predialysis, both low (≤4.0 mEq/L) and high (≥5.6 mEq/L) serum potassium concentrations are associated with increased morbidity and mortality among hemodialysis (HD) patients. We conducted a retrospective, observational study of 32,000 HD patients treated at facilities operated by Dialysis Clinic, Inc. (DCI) to further assess the relationships among serum and dialysate potassium concentrations, hospitalization, and mortality.  We constructed time varying proportional hazards models adjusted for clinic, demographics, and time varying laboratory values lagged by 2 months. Both hyperkalemia (K > 5 mEq/L) and hypokalemia (K < 3.5 mEq/L), independent of the dialysate potassium concentrations, were associated with increased risk of hospitalization and mortality over an average of more than 2 years of follow-up.  There were no significant interactions between dialysate potassium concentration and serum potassium concentration on outcomes.

AJKDblog: Based on this study, how would you propose we dialyze these two patients: one with a K of 8 meq/L and another with a K of 2.5 meq/L in your outpatient unit?

AG: Our study was not specifically designed to assess the effects of the dialysate potassium on such extreme potassium values given how unusual they are in the outpatient dialysis setting.  Kovesdy et al. reported that among patients with predialysis hyperkalemia (≥5 mEq/L), a high dialysate potassium (> 3 mEq/L) was associated with increased mortality. In contrast, Karnik et al. have cautioned that a high serum to dialysate potassium gradient (use of a dialysate potassium of 0 to 1 mEq/L in patients with serum potassium ≥5 mEq/L) was associated with an increased risk of sudden death. In our study, serum potassium was measured monthly. Since serum potassium and, therefore, the serum to dialysate gradient may vary considerably during the month, our study cannot assess the risks encountered in a single treatment.

Ideally, we would dialyze a patient with serum potassium of 8 mEq/L in hospital with cardiovascular monitoring and serially obtain serum potassium concentrations.  Depending on the clinical situation, we might begin with a dialysate potassium concentration of4 mEq/L and gradually reduce it to 2 mEq/L, depending on the clinical situation and the repeated serum potassium levels. We would recommend careful monitoring of the serum calcium to avoid increasing the arrythmogenic potential associated with hypocalcaemia. It is important to repeat the serum potassium at regular intervals during the initial 24 to 48 hours post dialysis, including immediately post HD and again several hours later, since the rebound in serum potassium may be clinically significant.

The absence of an interaction in our study suggests that the specific dialysate concentration is less important than the severity of hyperkalemia.  In our opinion, chronic treatment of this patient begins with identifying the reason for hyperkalemia and addressing those reasons.  Additionally, there may be a role for longer dialysis duration.  Finally, whether outcomes are better with a lower potassium dialysate versus a slightly higher dialysate potassium concentration combined with a potassium binding resin is an important study that should be conducted.

Similar caveats apply to the patient with a serum potassium concentration of 2.5 mEq/L. Such a value is quite unusual and clinical correlations are necessary. In this setting, we would likely use a dialysate potassium of 4 mEq/L. Critically, it is important to identify the cause of the hypokalemia and treat this underlying cause.

AJKDblog: Where do you and your group go from here?

AG: Presently, we are examining clinical outcomes associated with a variety of concentrations of dialysate constituents. Given how long and in how many people we have been performing hemodialysis, how little we know about the optimal composition of the dialysate is really unfortunate.

—

All Spring Clinical Meeting abstracts are available in the May issue of AJKD.

Check out more AJKDblog coverage of the NKF’s 2016 Spring Clinical Meetings (#SCM16)!