Suresh Kumar Angurana

Background: Multisystem inflammatory syndrome in children (MIS-C) is associated with favorable short-term outcomes. However, long-term follow-up data are limited, particularly from low- and middle-income countries. Methodology: This prospective follow-up study was conducted at a tertiary care hospital in North India between November 2024 and September 2025, involving children (n = 51) admitted with MIS-C and discharged between September 2020 and August 2022. The data collected included the clinical profile, echocardiography (left ventricular ejection fraction and coronary artery diameters), Screen for Child Anxiety–Related Disorders scale, UCLA Post-traumatic Stress Disorder Reaction Index and Vineland Social Maturity Scale. Results: During the acute phase, common manifestations included fever, gastrointestinal and mucocutaneous involvement, shock, elevated inflammatory markers, myocardial dysfunction (47%) and coronary abnormalities (22%). At follow-up 44 (43–46) months after discharge, none had major clinical symptoms; all had normalized left ventricular ejection fraction, normal school performance and peer relationships; and only 12% had mild coronary artery abnormalities. There were no anxiety or post-traumatic stress disorder–related symptoms, and all had age-appropriate social maturity. Conclusions: Long-term outcomes of MIS-C are favorable, with complete recovery of myocardial function, persistent mild coronary abnormalities in a small proportion, no anxiety or post-traumatic stress disorder, and normal social maturation.

Paraquat is a “restricted-use herbicide” known to be lethally toxic to humans. Paraquat poisoning carries a high fatality rate due to its lethality, lack of an antidote, and limited management guidelines. We report a 12-year-old boy who ingested approximately 40 mL of paraquat, leading to acute kidney injury requiring dialysis and subsequent evolving pulmonary fibrosis. He was managed with immunomodulation, antioxidants, antifibrotic therapy (nintedanib), and oxygen support. At 4-month follow-up, he remained clinically stable on long-term oxygen therapy.

Background and objectives: Sepsis remains a leading cause of mortality among children worldwide. This study aimed to describe clinical profile, intensive care needs, outcome, and predictors of mortality among children with severe sepsis. Methods: This secondary analysis of FerriPedS study was conducted in a tertiary pediatric intensive care units (PICU) in North India involving 115 children (3 months-12 yr) with severe sepsis. Data were collected regarding clinical details, laboratory investigations, organ dysfunction, intensive care needs, and outcome. Univariate and multinomial logistic regression analyses were used to determine predictors of mortality among severity score, organ dysfunction, serum ferritin, and PICU needs. Results: The median (IQR) age was 3 (1-7) years, and common diagnoses were community acquired pneumonia (CAP) (n=45 39.1%), scrub typhus (n=16, 13.9%), and CNS infections (n=12,10.4%). Mortality was 27.8% (n=32). Non-survivors had significantly higher Pediatric Risk of Mortality-III (PRISM III) score, serum ferritin, and daily Pediatric logistic organ dysfunction-2 (PeLOD-2) score. Higher proportion of non-survivors had positive blood culture at admission, coagulopathy, hepatic dysfunction, shock, acute respiratory distress (ARDS), acute kidney injury (AKI), and multiple organ dysfunction score (MODS); and required invasive mechanical ventilation, vasoactive drugs, blood products, and RRT. On multinomial logistic regression analysis, PRISM-III, ARDS, and AKI were independent predictors of mortality. Interpretation and conclusions: In children with severe sepsis, the common etiologies included CAP, scrub typhus, and CNS infections, and characterised by high mortality (27.8%). PRISM-III score, ARDS, and AKI were independent predictors of mortality.

Fluid overload (FO) is associated with poor clinical outcomes among critically ill children. The objective of this trial was to assess the impact of a protocolized restrictive maintenance fluid strategy on FO among mechanically ventilated children. This open-label randomized controlled trial was conducted in the pediatric intensive care unit (PICU) of a tertiary care hospital in North India over 13 months (November 2023–November 2024). Hemodynamically stable mechanically ventilated children were randomized to protocolized restrictive (n = 64) (40–50% of maintenance fluids with diuretic infusion if FO% > 10%); and liberal/usual (n = 66) (70–80% of maintenance fluid) groups. The primary outcome was the proportion of children with cumulative FO% > 10% through day 5. Secondary outcomes were daily cumulative FO%; inferior vena cava variability index (∆IVC) and N-terminal pro-brain natriuretic peptide (NT-proBNP) levels at 48 h; safety parameters (requirement of fluid boluses or vasoactive drugs); 28-day ventilator and PICU-free days (VFDs and PFDs), and mortality. Significantly fewer children in the protocolized restrictive group had cumulative FO% > 10% than in the liberal/usual group (22% vs. 47%, p = 0.003). Also, the restrictive group had significantly lower daily cumulative FO% on the first 5 days. ∆IVC and NT-proBNP levels at 48 h, as well as safety parameters, were similar between the two groups. Protocolized restrictive and liberal/usual groups had similar VFDs [20 (8–24) vs. 16 (0–23), p = 0.076], PFDs [16 (3–21) vs. 13 (0–19), p = 0.071], and mortality (14% vs. 24%, p = 0.14). Conclusion: A protocolized restrictive maintenance fluid strategy resulted in significantly lower FO among hemodynamically stable, mechanically ventilated children without a measurable impact on short-term clinical outcomes. What is Known: • Fluid overload (FO) is associated with poor clinical outcomes among critically ill mechanically ventilated children. What is New: • Among hemodynamically stable, mechanically ventilated children, a restrictive maintenance fluid strategy may be a useful intervention to limit FO and improve clinical outcomes in LMICs.

Background: Outborn neonates referred to tertiary care facilities face significantly higher mortality than inborn neonates. Existing severity scores (SNAP-II, CRIB-II) are validated only for inborn populations in high-income settings. We developed and internally validated an explainable machine learning model for predicting 24-hour mortality in outborn neonates admitted to a pediatric emergency room in India, representative of referral systems across South and South-east Asia. Methods: This prospective cohort study enrolled consecutive outborn neonates admitted to the pediatric emergency room of a tertiary care institute in North India. The primary outcome was predicting all-cause mortality within 24 hours following admission and early stabilization (within first 6 hours). Eight machine learning algorithms were evaluated using 5-fold cross-validation. Model performance was assessed using precision-recall PR-AUC, ROC-AUC, calibration metrics, and decision curve analysis. SHAP values were calculated for model explainability. Reporting followed the TRIPOD+AI guidelines. Findings: Of the 600 neonates enrolled, 121 (20·2%) died within 24 hours. Random forest algorithm demonstrated superior discrimination on the test set: ROC-AUC 0·81 (95% CI 0·76–0·85), PR-AUC of 0·562 (0·334, 0·760) and accuracy of 0·831 (0·809, 0·852). Calibration was adequate (Brier score 0·13, expected calibration error 0·05). Decision curve analysis showed positive net benefit across threshold probabilities of 5-30%. Top five predictors were small for gestation at birth, abnormal pediatric assessment triangle at admission, seizures at admission and refractory shock and low blood pH or standard base excess in the 6 hours period from admission, reflecting early post-admission physiologic instability. Interpretation: An explainable random forest model can predict 24-hour mortality in outborn neonates with good discrimination and calibration. The identified predictors are routinely available and actionable in emergency referral settings. Such models may support triage and early resource allocation in overstretched neonatal emergency systems across South and Southeast Asia. External validation across diverse LMIC settings is warranted before clinical implementation