Abstract Introduction Children with obstructive sleep apnea (OSA) are at increased risk for perioperative adverse events. Whilst polysomnography (PSG) remains the reference standard for OSA diagnosis, oximetry is a valuable screening tool; visual analysis of desaturation clusters derived from a tabletop oximeter is the traditional practice. However, wearable oximetry with automated analysis could be an alternative option. This study investigated the accuracy of a wrist-worn oximeter as an OSA screening tool for a paediatric surgical population. Method Children scheduled for adenotonsillectomy without significant co-morbidities underwent concurrent overnight PSG and oximetry (Nonin WristOx2 3150) preoperatively. To determine the obstructive apnea-hypopnea index (OAHI), PSG was scored manually, and oximetry data were auto-analysed by Nonin proprietary software to determine the 3% oxygen desaturation index (oximetric ODI3). Logistic regression assessed the predictive performance of ODI3 and covariates (age, gender, BMI z-score) for detecting any OSA and moderate-severe OSA. Results Seventy-six children (34 females, aged (mean±SD) 5.7±1.6years) were classified based on PSG-derived OAHI as having no OSA (n=31), mild (n=31), and moderate-severe OSA (n=14). The moderate-severe OSA (OAHI≥5 events/hr) prediction analysis showed that oximetric ODI3 (OR 1.38, 95% CI 1.15, 1.65, p=0.001) was the sole OSA predictor. At a cut-off of ODI3=5events/hr, sensitivity and specificity were 78.6% and 75.8%, respectively, capturing all severe OSA cases (ROC-AUC=0.857). For any OSA (OAHI≥ 1 event/hr), prediction analysis showed reduced accuracy in performance: 75.6% sensitivity, 61.3% specificity (ROC-AUC=0.7097). Conclusion Wrist-worn oximetry-derived automated ODI3 can reliably identify moderate-severe OSA. Easy acquisition and interpretation may expedite the preoperative identification of OSA at-risk children.