Purpose: Although a moderate proportion of cardiac arrest (CA) patients achieve a return of spontaneous circulation (ROSC), few survive to discharge, mostly with poor neurological development. As serum phosphate levels were described as elevated after cardiopulmonary resuscitation (CPR), we asked whether these elevations would predict a higher risk of mortality and impaired neurological outcome in CA patients following ROSC. Methods: Initial serum phosphate levels, survival, and neurologic status at discharge of 488 non-traumatic CA patients treated at a single German hospital after achieving ROSC were analyzed. The cut-off value of phosphate for mortality prediction was determined using the receiver operator characteristic (ROC) curve, and patients were divided accordingly for comparison. Results were validated by analyzing phosphate levels in a multi-centric cohort containing 3299 CA patients from the eICU database of the United States. Results: In the German cohort, ROC analysis showed a 90% specificity for phosphate levels >2.7 mmol/L to predict mortality (AUC: 0.76, p < 0.0001), and phosphate level elevations were associated with higher in-hospital mortality (crude odds ratio 3.04, 95% CI 2.32 to 4.08). Patients with initial phosphate levels >2.7 mmol/L had significantly higher mortality in both analyzed collectives (p < 0.0001). Similarly, patients from the German cohort who initially had higher phosphate levels also showed a higher proportion of impaired neurological status at discharge and morphological signs of brain injury. Conclusions: In CA patients following ROSC, initial serum phosphate levels >2.7 mmol/L predict higher mortality and impaired neurological outcome. Our data suggests that phosphate determination might improve the preciseness of the overall and neurologic prognostication in patients after CPR following ROSC.
Keywords: outcome; phosphate; prognosis; resuscitation.