Serum Potassium and Short-term Clinical Outcomes Among Hemodialysis Patients: Impact of the Long Interdialytic Interval.

BACKGROUND: Hyperkalemia is common among hemodialysis patients and is associated with morbidity and mortality. The long interdialytic interval is likewise associated with adverse outcomes. However, the interplay among serum potassium, dialysis cycle phase, and clinical outcomes has not been examined.

STUDY DESIGN: Retrospective observational study.

SETTING & PARTICIPANTS: 52,734 patients receiving in-center hemodialysis at a large dialysis organization during 2010 and 2011 contributed 533,889 potassium measurements (230,634 on Monday; 285,522 on Wednesday; 17,733 on Friday).

PREDICTOR: Serum potassium concentration, day of the week of potassium measurement.

OUTCOMES: Death, hospitalization, emergency department (ED) visit.

RESULTS: There was a significant association between higher serum potassium and risk of hospitalization within 96 hours that was of greater magnitude on Fridays (389 hospitalizations) than Mondays or Wednesdays (4,582 and 4,629 hospitalizations, respectively; P for interaction = 0.008). Serum potassium of 5.5 to <6.0 (vs the referent category of 4.0-<4.5 mEq/L) was associated with increased risk of hospitalization on Fridays, with an adjusted OR of 1.68 (95% CI, 1.22-2.30). However, serum potassium of 5.5 to <6.0 mEq/L was associated with only mild elevation of risk on Mondays and no significantly increased risk on Wednesdays (adjusted ORs of 1.12 [95% CI, 1.00-1.24] and 1.04 [95% CI, 0.94-1.16], respectively). Associations of elevated serum potassium (6.0-<6.5 mEq/L or greater) with death and ED visit were significant, but did not differ based on day of the week.

LIMITATIONS: There were insufficient observations to detect effect modification by day of the week for deaths, ED visits, and specific causes of hospitalizations. Confounding may have influenced results.

CONCLUSIONS: Higher serum potassium is associated with increased short-term risk of hospitalization, ED visit, and death. The association between serum potassium and hospitalization risk is modified by day of the week, consistent with a contribution of accumulated potassium to adverse outcomes following the long interdialytic interval. Further work is needed to determine whether directed interventions ameliorate this risk.