Association between serum potassium levels and short-term mortality in out-of-hospital cardiac arrest survivors

Funding Acknowledgements Type of funding sources: Foundation. Main funding source(s): The Danish Heart Foundation The Danish Foundation TrygFonden Background Serum potassium levels outside the normal range are associated with increased short-term mortality in patients with acute cardiovascular disease such as acute heart failure. In survivors of out-of-hospital cardiac arrest (OHCA), the initial serum potassium concentration is often outside the normal range. However, it is unknown whether there is an association between initial serum potassium level and short-term mortality in OHCA survivors. Purpose The aim of this study was to describe potential associations between serum potassium level and 30-day survival in patients surviving OHCA. Methods We identified 1630 patients who had return of spontaneous circulation (ROSC) at hospital arrival, and a registered post-OHCA serum-potassium value, using Danish nationwide registry data from 2001-2015. Thirty-day survival was analyzed according to the following seven predefined potassium levels: 6.0 mmol/L. Thirty-day survival was estimated using a multivariable Cox regression (reference normokalemia 3.5–4.6 mmol/L). The multivariable model included age, sex, Charlson comorbidity index, witnessed status, performance of CPR and first registered heart rhythm. Results During 30-day follow-up, survival in the seven strata was 29 (55.8%), 52 (61.9%), 167 (63.3%), 543 (58.7%), 66 (35.5%), 12 (28.6%), and 18 (23.4%), respectively. Standardized average risk ratios (ARR) of 30-day survival was significantly lower for all groups with hyperkalemia compared with normokalemia: 4.7-5.5 mmol/L: (ARR: 0.86, 95% confidence interval (95% CI): 0.74–0.96); 5.5-6.0 mmol/L: (ARR: 0.76, 95% CI: 0.52–0.98); >6.0 mmol/L: (ARR: 0.64, 95% CI: 0.50–0.79). Survival did not differ significantly in patients with hypokalemia compared with normokalemia: Conclusions In OHCA survivors, hyperkalemia was associated with reduced 30-day survival compared with normokalemia. This association was independent of age, sex, Charlson comorbidity index, witnessed status, performance of CPR and first registered heart rhythm. Conversely, hypokalemia was not associated with reduced 30-day survival in this patient population.