Procalcitonin level, neutrophil to lymphocyte count ratio, and mean platelet volume as predictors of organ dysfunction and mortality in children with sepsis

Background Procalcitonin (PCT) level is one of known biomarker in septic diagnosis, but limited studies report its benefit in predicting the outcomes of children with sepsis. Neutrophil to lymphocyte (NLR) and mean platelet volume (MPV) are simple biomarkers of inflammation that can be measured in routine hematological examination which role in predicting organ dysfunction remain unclear. Objective To understand the correlations between PCT level, NLR, and MPV, tested in the first day of admission with outcomes of septic children in intensive care unit. Methods This retrospective cohort study obtained the data from medical record of pediatric patients admitted in PICU and HCU since January 2019. The inclusion criteria were children aged 1 months to 18 years with sepsis; whie exclusion criteria were patients with congenital heart disease, hematologic disease, malignancy, and length of care in intensive care unit less than 3 days or more than 28 days. The PCT, NLR, and MPV levels were assessed in the first day of admission. Organ dysfunction was identified using qSOFA score more than 2 points.   Results Sixty-nine septic children were reviewed. Procalcitonin level in the first day of admission correlated significantly with qSOFA score in the third day of admission (R= 0.639; P=0.000); as well as with mortality (R=0.747; P=0.000). Receiver operating characteristic (ROC) curve of PCT level in the first day of admission had area under curve (AUC) of 0.922 to predict organ dysfunction (cut off 3.425; sensitivity 95.8%; specificity 52.4%) and AUC of 0.952 to predict mortality (cut off 21.165; sensitivity 96.4%; specificity 78%). Moreover, NLR in the first day of admission correlated significantly with qSOFA in the third day of admission (R=0.407; P=0.001), but did not correlate with mortality. The ROC of NLR to predict organ dysfunction was 0.829 (cut off 3.52; sensitivity 87.5%; specificity 66.7%). There was no correlation between MPV in the first day of admission with qSOFA score in the third day of admission neither with mortality. Linear regression test showed that PCT level and NLR in the first day of admission simultaneously had correlated with qSOFA score in the third day of admission (R=0.696; P= 0.000) and mortality (R=0.748; P=0.000). Meanwhile, PCT and MPV simultaneously had correlation with qSOFA score in the third day of admission (R=0.688; P=0.000) and mortality (R=0.733; P=0.000). Moreover, NLR and MPV simultaneously had correlation with qSOFA score in the third day of admission (R=0.453; P=0.002). All three independent variables (PCT level, NLR, and MPV) simultaneously correlated with qSOFA score in the third day of admission (R= 0.744; P=0.000) and mortality (R=0.739; P=0.000). Conclusion There are significant correlations between each, PCT level and NLR in the first day of admission with qSOFA score in the third day of admission as well as with mortality. There is no  correlation between MPV in the first day of admission with qSOFA score in the third day of admission, neither with mortality. There are significant correlations between PCV level and NLR with or without MPV with qSOFA score in the third day of admission as well as with mortality.