POSTOPERATIVE COMPLICATIONS IN TONSILLECTOMY: PREDICTIVE FACTORS AND RISK STRATIFICATION IN PEDIATRIC PATIENTS

Abstract

Postoperative complications following pediatric tonsillectomy—most notably hemorrhage, respiratory compromise, dehydration, and unplanned readmissions—remain a pressing clinical challenge, affecting 10.0% of patients and contributing to increased morbidity, prolonged hospitalization, and higher healthcare costs. To address the need for reliable preoperative risk stratification, we conducted a retrospective cohort analysis of 1,200 children aged 2–16 years who underwent tonsillectomy or adenotonsillectomy at three tertiary-care centers between January 2022 and December 2023. We systematically extracted demographic (age, sex, race/ethnicity), clinical (body mass index percentile, preoperative apnea–hypopnea index, comorbid asthma or cardiac disease), surgical (technique, surgeon experience, operative duration, estimated blood loss), and perioperative (analgesic regimen, hospital length of stay) variables from electronic health records. Univariate screening (p < 0.10) and subsequent multivariable logistic regression with backward elimination (α = 0.05), complemented by LASSO regularization to guard against overfitting, identified five independent predictors of significant complications within 30 days: age under four years (adjusted OR 2.1, 95% CI 1.3–3.4), preoperative apnea–hypopnea index >15 events/hour (OR 3.2, 95% CI 2.0–5.1), underlying cardiac comorbidity (OR 4.5, 95% CI 2.6–7.8), surgeon experience under five years (OR 1.9, 95% CI 1.1–3.2), and use of total dissection technique (OR 1.6, 95% CI 1.0–2.6). The final model exhibited robust performance—area under the receiver–operator characteristic curve of 0.87 and satisfactory calibration (Hosmer–Lemeshow p = 0.21)—and enabled clear stratification into low- (predicted risk <5%, observed 2.3%), intermediate- (5–15%, observed 8.7%), and high-risk (>15%, observed 21.4%) groups. These findings support the utility of an evidence-based, objective risk calculator to inform individualized perioperative planning, such as targeted hemostatic measures, extended postoperative monitoring, and optimized analgesic protocols for those at highest risk. Adoption of this tool has the potential to standardize care, reduce practice variation, and allocate resources more effectively. Future prospective validation across diverse hospital settings—and exploration of additional biomarkers such as coagulation profiles or inflammatory mediators—will be essential to confirm generalizability and further refine predictive accuracy.