Clinical and biochemical determinants of critical deterioration in paediatric scrub typhus: a retrospective study from a tertiary care centre in South India.
{"title":"Clinical and biochemical determinants of critical deterioration in paediatric scrub typhus: a retrospective study from a tertiary care centre in South India.","authors":"C Hari Prasath, P Anil Kumar, Penatal Sripooja","doi":"10.1093/tropej/fmag068","DOIUrl":null,"url":null,"abstract":"<p><p>Scrub typhus causes substantial morbidity and mortality in children across South Asia, yet admission-time predictors of critical deterioration remain poorly characterized in the paediatric population. This study aimed to identify independent predictors of critical deterioration in hospitalized children with confirmed scrub typhus. A retrospective observational cohort study was conducted at a tertiary-care teaching hospital in South India over 6 years. Children aged 1-14 years with immunoglobulin M enzyme-linked immunosorbent assay-confirmed scrub typhus were included. Critical deterioration, a pre-specified composite of shock, acute respiratory distress syndrome, multiple organ dysfunction syndrome, mortality, or high-flow respiratory support requirement, constituted the primary outcome. Independent predictors were identified by multivariable logistic regression, and discriminatory performance was assessed by area under the receiver operating characteristic curve (AUROC). One hundred and eighty-one children were enrolled (median age 6.0 years; 50.3% male). Critical deterioration occurred in 62 children (34.3%), with a case fatality rate of 12.9% among those affected. Serum albumin demonstrated the highest univariable discriminatory performance (AUROC 0.973), followed by neutrophil-to-lymphocyte ratio (NLR) (0.945), total bilirubin (0.929), and lactate dehydrogenase (0.902). Four variables independently predicted critical deterioration on multivariable analysis: NLR [adjusted odds ratio (aOR) 3.811, 95% confidence interval (CI): 2.119-6.857; P < .001), C-reactive protein (CRP) (aOR 1.052, 95% CI: 1.018-1.089; P = .003), doxycycline delay (aOR 1.829, 95% CI: 1.146-2.920; P = .011), and serum sodium (aOR 0.689, 95% CI: 0.523-0.909; P = .008). The model demonstrated excellent discrimination (apparent AUROC 0.986, 95% CI: 0.971-0.997; optimism-corrected AUROC 0.982, 95% CI: 0.978-0.987) and satisfactory calibration (Hosmer-Lemeshow P = .997). NLR, CRP, serum sodium, and doxycycline delay independently predict critical deterioration in paediatric scrub typhus at hospital admission, with serum albumin demonstrating the highest univariable discriminatory performance. Serial biomarker trajectories demonstrated sustained inflammatory and haemodynamic perturbation through 72 hours in deteriorating children, in contrast to the rapid biochemical resolution observed in those who recovered. These findings require external prospective validation before application to clinical risk stratification.</p>","PeriodicalId":17521,"journal":{"name":"Journal of Tropical Pediatrics","volume":"72 6","pages":""},"PeriodicalIF":1.1000,"publicationDate":"2026-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Tropical Pediatrics","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1093/tropej/fmag068","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"PEDIATRICS","Score":null,"Total":0}
引用次数: 0
Abstract
Scrub typhus causes substantial morbidity and mortality in children across South Asia, yet admission-time predictors of critical deterioration remain poorly characterized in the paediatric population. This study aimed to identify independent predictors of critical deterioration in hospitalized children with confirmed scrub typhus. A retrospective observational cohort study was conducted at a tertiary-care teaching hospital in South India over 6 years. Children aged 1-14 years with immunoglobulin M enzyme-linked immunosorbent assay-confirmed scrub typhus were included. Critical deterioration, a pre-specified composite of shock, acute respiratory distress syndrome, multiple organ dysfunction syndrome, mortality, or high-flow respiratory support requirement, constituted the primary outcome. Independent predictors were identified by multivariable logistic regression, and discriminatory performance was assessed by area under the receiver operating characteristic curve (AUROC). One hundred and eighty-one children were enrolled (median age 6.0 years; 50.3% male). Critical deterioration occurred in 62 children (34.3%), with a case fatality rate of 12.9% among those affected. Serum albumin demonstrated the highest univariable discriminatory performance (AUROC 0.973), followed by neutrophil-to-lymphocyte ratio (NLR) (0.945), total bilirubin (0.929), and lactate dehydrogenase (0.902). Four variables independently predicted critical deterioration on multivariable analysis: NLR [adjusted odds ratio (aOR) 3.811, 95% confidence interval (CI): 2.119-6.857; P < .001), C-reactive protein (CRP) (aOR 1.052, 95% CI: 1.018-1.089; P = .003), doxycycline delay (aOR 1.829, 95% CI: 1.146-2.920; P = .011), and serum sodium (aOR 0.689, 95% CI: 0.523-0.909; P = .008). The model demonstrated excellent discrimination (apparent AUROC 0.986, 95% CI: 0.971-0.997; optimism-corrected AUROC 0.982, 95% CI: 0.978-0.987) and satisfactory calibration (Hosmer-Lemeshow P = .997). NLR, CRP, serum sodium, and doxycycline delay independently predict critical deterioration in paediatric scrub typhus at hospital admission, with serum albumin demonstrating the highest univariable discriminatory performance. Serial biomarker trajectories demonstrated sustained inflammatory and haemodynamic perturbation through 72 hours in deteriorating children, in contrast to the rapid biochemical resolution observed in those who recovered. These findings require external prospective validation before application to clinical risk stratification.
期刊介绍:
The Journal of Tropical Pediatrics provides a link between theory and practice in the field. Papers report key results of clinical and community research, and considerations of programme development. More general descriptive pieces are included when they have application to work preceeding elsewhere. The journal also presents review articles, book reviews and, occasionally, short monographs and selections of important papers delivered at relevant conferences.