Use of Systemic Inflammatory Markers to Replace Chest Radiography in Clinical Decision Making for Adolescents with Tuberculosistext_resourcesIntroduction: In 2022, the World Health Organization issued new recommendations for shortened treatment regimens for children aged less than 16 years with non-severe Mycobacterium tuberculosis (TB) but it did not provide evidence-based guidelines for when chest x-rays (CXRs) are not available. Commonly assessed laboratory inflammatory markers and patient age could help clinicians determine treatment duration for patients in low-resource, high TB-burden settings. Methods: A retrospective cohort study was conducted using data on adolescents aged 10 to 15 treated over a ten-year period at a tertiary pediatric hospital in Buenos Aires, Argentina. Experienced pediatric pulmonologists read patient CXRs, which were used to categorize adolescents with non-severe or severe drug-susceptible intrathoracic TB per WHO guidelines. Adolescents were excluded if they had comorbidities that might affect inflammatory markers, previous treatment in the past two years, or, if categorized as non-severe, they had contraindications to shortened treatment (current pregnancy or malnutrition). Univariable logistic regressions were employed to identify potential predictors of non-severe TB from baseline laboratory values and patient age. Marginal R2 values were used to select variables for inclusion in the multivariate regression model. Multiple imputation by chained equations (MICE) was used to impute variables with <25% missingness, while back-wise step regression was used to select predictors for the final model. Lastly, the area under the curve (AUC) was calculated from the final model. Results: 525 patient charts were abstracted, of which 284 met eligibility criteria (93 had non-severe TB whereas 191 had severe TB). Factors independently associated with non-severe TB were age (adjusted OR 0.70, 95% Confidence Interval: 0.54-0.90), erythrocyte sedimentation rate (aOR 0.93, 95% CI: 0.91-0.95), and neutrophil percentage (aOR 0.95, 95% CI: 0.91-0.99). The AUC was 0.903 (95% CI: 0.861-0.945). Conclusions: A model using baseline age, erythrocyte sedimentation rate, and neutrophil percentage can be used to determine probability of non-severe intrathoracic TB with a high degree of discrimination.Pralea, Alexander T.AuthorCastillo Espinal, JessielAuthorMacha Marín, EdithAuthorPawluk, VictorAuthorTanzer, Joshua RayAuthorGonzález, Norma E.AuthorChiang, Silvia S.AuthorBrown University. Alpert Medical School. Scholarly Concentration Program. Global Healthresearch programPediatricsTuberculosisDiagnosisEnglish2025Global HealthAll rights reservedIn CopyrightAll Rights Reserved10.26300/ahj4-z498