Hypophosphatemia is frequently observed in acetaminophen-induced hepatotoxicity and may be involved in the pathogenesis of hepatic failure. The aim of the study was to evaluate the prognostic value of serial measurements of serum phosphate in patients with severe acetaminophen poisoning. Prospectively, serial measurements of serum phosphate were performed in 125 patients with severe acetaminophen poisoning. The optimum threshold value of serum phosphate to discriminate nonsurvivors was identified. Prognostic value and speed of identification were compared with those of the King's College Hospital (KCH) criteria. Phosphate concentrations were significantly higher in nonsurvivors than in survivors at 48 to 72 hours after overdose (mean 2.65 ± 1.18 mmol/L vs. 0.68 ± 0.22 mmol/L, P < .001) as well as 72 to 96 hours after overdose (2.12 ± 0.22 mmol/L vs. 0.59 ± 0.23 mmol/L, P < .001). A threshold phosphate concentration of 1.2 mmol/L at 48 to 96 hours after overdose had sensitivity 89%, specificity 100%, accuracy 98%, positive predictive value 100%, and negative predictive value 98%. The phosphate criteria had higher sensitivity, accuracy, and positive and negative predictive values than the KCH criteria, and it identified patients significantly earlier after transferal [median 1 hour (range 1-38 hours) vs. 12 hours (2-192 hours), P < .05, respectively]. In nonsurvivors, the degree of hyperphosphatemia correlated with renal dysfunction (R = .55; P = .02). In conclusion, hyperphosphatemia after acetaminophen overdose is seen exclusively in nonsurvivors, which makes it a highly specific as well as sensitive predictor of nonsurvival. We propose that hyperphosphatemia is caused by renal dysfunction in the absence of hepatic regeneration, as the latter appears to be associated with lowering of serum phosphate.