The usefulness of the SOFA and APACHE II scoring systems for the early prediction of mortality in patients with dapsone poisoning

Abstract

The rate of mortality from dapsone poisoning is high because of the long absorption half-life of dapsone. This study aimed to evaluate the usefulness of the sequential organ failure assessment (SOFA) and acute physiology and chronic health evaluation II (APACHE II) scoring systems for the early prediction of mortality in patients with dapsone poisoning. This is a retrospective and observational study of consecutive patients diagnosed with dapsone poisoning. The SOFA and APACHE II scores were obtained within the first 24 h of admission. Patients were divided into survivor and non-survivor groups. In total, 106 patients were included. The SOFA scores of the survivor and non-survivor groups were 1 (0–8) and 4 (1–10), respectively ( p < 0.001). The APACHE II scores of the survivor and non-survivor groups were 9 (1–25) and 14 (3–23), respectively ( p < 0.001). Based on these scores and in-hospital mortality cases, the standardized mortality ratios for the APACHE II and SOFA were 1.00 (95% confidence interval (CI): 0.64–1.48) and 1.00 (95% CI: 0.64–1.49), respectively. In the model adjusted for clinically important variables and variables with significant differences between the survivor and non-survivor groups, the area under the curve of the SOFA (0.907; 95% CI: 0.834–0.955) was significantly higher than that of the APACHE II (0.793; 95% CI: 0.703–0.867) ( p = 0.008). The SOFA and APACHE II score systems had good discrimination and satisfactory calibration performance in patients with dapsone poisoning. However, the SOFA score was a more useful method in predicting mortality than the APACHE II score.