Potassium disorders at intensive care unit admission and functional outcomes after cardiac arrest.

Background: Abnormal serum potassium levels are commonly found in the intensive care unit (ICU) population. We aimed to determine the prevalence of potassium disorders at ICU admission and its association with functional outcomes in comatose patients resuscitated from cardiac arrest.

Methods: We performed a post hoc analysis of pooled data from four randomised clinical trials involving comatose post-cardiac arrest patients admitted to ICU after return of spontaneous circulation (ROSC). Reference serum potassium levels were defined as between 3.0 and 4.9  mmol/L. An unfavourable functional outcome was defined as a cerebral performance category of 3 to 5 at 180 days. We compared potassium disturbances categorically in a mixed effects logistic regression model including initial rhythm, delay from collapse to return of spontaneous circulation, bystander cardiopulmonary resuscitation, lactate and urea at ICU admission, with normokalaemia set as the reference group.

Results: We included 1133 patients (557 from the HYPERION, 346 from the TTH48, 120 from the COMACARE, and 110 from the Xe-HYPOTHECA trials) with a median age of 64 years (interquartile range 55-72) and a predominance of males (72 %); a total of 712 (64 %) patients had unfavourable functional outcomes. On admission, 221 patients (19.5 %) experienced hyperkalaemia and 35 (3.1 %) patients experienced hypokalaemia. Fewer patients in the normokalaemia group (513/877, 58.5 %) had an unfavourable functional outcome compared to the hypokalaemia (24/35, 68.6 %) and hyperkalaemia groups (180/221, 81.4 %; p < 0.001). Hyperkalaemia was associated with higher odds for an unfavourable functional outcome (adjusted odds ratio (OR) 1.85, 95 % confidence interval (CI) 1.10-3.12, p = 0.02), while hypokalaemia was not (OR 1.36 95 % CI 0.51-3.60, p = 0.53). The associations were not significant in a subgroup analysis adjusted for the modified cardiac arrest hospital prognosis score in 833 patients (OR 1.74, 95 % CI 0.91-3.34, p = 0.10 for hyperkalaemia and OR 1.48, 95 % CI 0.40-5.44, p = 0.55 for hypokalaemia).

Conclusions: Of the comatose patients admitted to ICU after cardiac arrest, one in five experienced a potassium disorder on ICU admission. Hyperkalaemia was associated with unfavourable functional outcomes at 180 days, while hypokalaemia was not.