Platelet indices as a predictor of mortality in critically ill children

Abstract

In the current study, we aimed to elaborate the utility of platelet indices as biomarkers for predicting critical illness and mortality in critically ill children. Two main groups were included; group I (patient group): Two hundred (200) critically ill children, group II: Two hundred (200) healthy children with comparable age and sex, were enrolled as a control group. All subjects were evaluated clinically and by CBC including platelet indices (MPV, PDW, and PCT). Patient group I were evaluated with SOFA and PRISM scores and subjected to full lab and other radiological and invasive procedures as indicated. Among 200 critically ill patient 33% of them died and 67% survived and were discharged. Interestingly, It was found that Platelet count and PCT were significantly lower in patient group as compared to the control group with p-value (< 0.001). Moreover, the non survivors of patient group showed significantly lower platelet count and PCT values than the survivors in the same group with pvalue (< 0.001). In contrast, MPV and PDW were found to be significantly higher in patient group as compared to the control group with p-value (< 0.001), while, the survivors of patient group showed significantly lower MPV and PDW with p-value (< 0.001) Worthy noting that PCT was inversely related to PRISM and SOFA score while, MPV and PDW were directly related to PRISM and SOFA score. PDW was the most sensitive parameter to discriminate between patient and control group with sensitivity 96% while, PCT was the most specific parameter with 99% specificity. In addition, PCT was the most sensitive parameter to predict mortality in patient group with sensitivity 86.57%. Finally, MPV and PDW were the most specific parameters to predict mortality in patient group with specificity 87.88%. In conclusion, this study proved that platelet count and platelet indices are valid predictors for assessment of critical illness and PICU mortality.