Abstract
Introduction: Acute appendicitis is a common paediatric emergency, with higher risks of complications than in adults. Accurately distinguishing uncomplicated appendicitis (UA) from complicated appendicitis (CA) is essential for guiding appropriate management, particularly as non‑operative treatment for UA gains acceptance. However, existing scoring systems and single biomarkers lack sufficient specificity, especially in paediatric populations. This study assesses whether common inflammatory markers can predict UA and CA in children, providing a supplementary diagnostic tool for resource‑limited settings where imaging may be limited. Methods: We conducted a retrospective analysis of paediatric patients aged 16 years or younger who underwent appendectomy at two tertiary hospitals. Data included demographic information, symptoms and pre‑operative blood markers such as C‑reactive protein (CRP), total leucocyte count (TLC), neutrophil count and platelet count. Inflammatory marker values were incorporated into a multivariate logistic regression model to predict the likelihood of CA. Diagnostic accuracy was evaluated using the receiver operating characteristic curve analysis, reporting sensitivity, specificity and optimal cutoffs for the identified markers. Results: Of 247 paediatric appendectomy cases, 60.7% were UA and 39.2% were CA. Elevated CRP (≥76.8 mg/L) had a specificity of 78.7% and a sensitivity of 60.3% for CA. Combined markers (TLC, neutrophils, lymphocytes and platelets) improved sensitivity to 70.1%. The logistic regression model incorporating neutrophil count, lymphocyte count and platelet count achieved a sensitivity of 78.4% and specificity of 52.7% at a threshold score of 1.20. This model effectively stratified risk, providing diagnostic support when advanced imaging is limited. Conclusion: Our study introduces a logistic regression model using routine blood markers to differentiate between UA and CA in paediatric patients. While not a replacement for imaging or clinical evaluation, the model offers a valuable adjunct in settings with limited resources, improving sensitivity for detecting complicated cases. Prospective studies are warranted to validate these findings in the real‑time clinical applications.
Keywords: Complicated appendicitis, inflammatory markers, paediatric appendicitis, predictive model