ABSTRACT
PURPOSE: Given the high burden of health care-associated infections (HAIs) in resource-limited settings, there is a tendency toward overdiagnosis/treatment. This study was designed to create an easy-to-use, dynamic, bedside risk stratification model for classifying children based on their risk of developing HAIs during their pediatric intensive care unit (PICU) stay, to aid judicious resource utilization. MATERIALS AND METHODS: A prospective, observational cohort study was conducted in the 12-bed PICU of a large Indian tertiary care hospital between January and October 2011. A total of 412 consecutive admissions, aged 1 month to 12 years with PICU stay greater than 48 hours were enrolled. Independent predictors for HAIs identified using multivariate regression analysis were combined to create a novel scoring system. Performance and calibration of score were assessed using receiver operating characteristic curves and Hosmer-Lemeshow statistic, respectively. Internal validation was done. RESULTS: Age (<5 years), Pediatric Risk of Mortality III (24 hours) score, presence of indwelling catheters, need for intubation, albumin infusion, immunomodulator, and prior antibiotic use (≥4) were independent predictors of HAIs. This model, with area under the ROC curve of 0.87, at a cutoff of 15, had a negative predictive value of 89.9% with overall accuracy of 79.3%. It reduced classification errors from 29.8% to 20.7%. All 7 predictors retained their statistical significance after bootstrapping, confirming the internal validity of the score. CONCLUSIONS: The "Pediatric Risk of Nosocomial Sepsis" score can reliably classify children into high- and low-risk groups, based on their risk of developing HAIs in the PICU of a resource-limited setting. In view of its high sensitivity and specificity, diagnostic and therapeutic interventions may be directed away from the low-risk group, ensuring effective utilization of limited resources.
Subject(s)
Cross Infection/diagnosis , Intensive Care Units, Pediatric , Sepsis/diagnosis , Child , Child, Preschool , Female , Humans , Infant , Logistic Models , Male , Pediatrics/methods , Point-of-Care Testing , Prospective Studies , ROC Curve , Reproducibility of Results , Risk , Sensitivity and SpecificitySubject(s)
Hypotonic Solutions , Isotonic Solutions , Central Nervous System Infections , Child , Humans , Infusions, IntravenousABSTRACT
OBJECTIVE: To study the clinical profile of children with scrub typhus and its association with hemophagocytic lymphohistiocytosis. METHODS: Children presenting with unexplained fever and multi-systemic involvement between May to December 2011 were tested for scrub typhus using IgM ELISA kits. Occurrence of Hemophagocytic lymphohistiocytosis in IgM positive cases of scrub typhus was studied. RESULTS: Of the 35 children with unexplained fever and multi-systemic involvement, 15 children (9 boys) tested positive for scrub typhus. Thrombocytopenia, hypoalbuminemia and raised hepatic transaminases were observed in all children. Out of seven children evaluated for hemophagocytic lymphohistiocytosis. 3 met the criteria for hemophagocytosis. Two children (one with hemophagocytic lymphohistiocytosis) died. CONCLUSIONS: Scrub typhus is a common cause of unexplained fever in children in northern India. Hemophagocytic lymphohistiocytosis can occasionally complicate scrub typhus in children.
