Hypokalemia is a biochemical signal of poor prognosis for acute paraquat poisoning within 4 hours.

Anecdotal reports have suggested that hypokalemia is an independent predictor of mortality in patients with acute paraquat poisoning. However, the clinical significance of hypokalemia has not been systematically investigated in the early stage of paraquat poisoning. This study aimed to evaluate whether initial hypokalemia is a good predictor of mortality in patients with acute paraquat poisoning within 4 h. We retrospectively analyzed patients who were admitted to the emergency department after paraquat poisoning between September 2012 and January 2015. Demographic, clinical, and laboratory data were recorded, and the prognostic significance of hypokalemia was analyzed. A total of 120 patients were included. The 60-day mortality was 71.7 %. Serum potassium concentrations were significantly lower in non-survivors (3.1 ± 0.5 mmol/L) than in survivors (3.8 ± 0.3 mmol/L, P < 0.001). Independent predictors of death were amount of paraquat ingested (hazard ratio 1.005; 95 % confidence interval 1.002-1.007), serum potassium (0.498, 0.277-0.897), bicarbonate (0.934, 0.876-0.995), and white blood cell count (1.032, 1.001-1.065). For receiver operating characteristic curve analysis, serum potassium had an area under the curve of 0.852 (95 % confidence interval 0.784-0.920, P < 0.001), and the best cutoff value was 3.5 mmol/L (sensitivity, 88.2 %; specificity, 75.6 % in predicting survivors). Kaplan-Meier analysis showed that decreased serum potassium concentrations were associated with an increase in 60-day mortality (P < 0.001). Hypokalemia may be a reliable predictor in evaluating prognosis in paraquat poisoning within 4 h. The mechanism is not clear, and further studies verifying the precise mechanism of hypokalemia are required.