Authors
Qin Liao, Yang Yang, Changqin Li, Zhengdong Wang
Published in
Journal of perianesthesia nursing : official journal of the American Society of PeriAnesthesia Nurses. Aug 12, 2026. Epub Aug 12, 2026.
Abstract
This study aimed to investigate the prevalence and characteristics of perioperative sleep disturbances in children undergoing elective surgery, to identify associated risk factors, and to develop and validate a predictive nomogram for the early identification of high-risk patients.
This was a prospective cross-sectional study.
Between June and December 2024, 361 children aged 3 to 12 years scheduled for elective surgery under general anesthesia were enrolled using a two-stage mixed sampling method. Sleep was assessed using the Children's Sleep Habits Questionnaire (CSHQ) and caregiver sleep via the Pittsburgh Sleep Quality Index (PSQI). Univariate analysis was first performed to identify potential predictors, followed by least absolute shrinkage and selection operator (LASSO) logistic regression with 10-fold cross-validation for data-driven variable selection. Multivariable logistic regression was then conducted to determine independent risk factors, and statistically significant predictors were incorporated into the final nomogram. Model evaluation included assessment of discrimination (Area Under the Curve, AUC; concordance index, C-index), calibration (Hosmer-Lemeshow test, calibration curve), and clinical utility (Decision Curve Analysis, DCA). Internal validation was conducted using a 70%/30% stratified split (training/test set) and bootstrap resampling (1000 repetitions). Sensitivity analyses were conducted using alternative CSHQ cutoffs ( ≥48 and ≥54) to assess the stability of prevalence estimates and model performance.
The perioperative sleep disturbance detection rate was 90.0% (325/361) using the standard CSHQ cutoff of ≥41, with core symptoms of sleep resistance (56.0%), sleep anxiety (42.4%), and abnormal sleep duration (36.0%). Sensitivity analyses revealed that raising the CSHQ cutoff to ≥48 reduced prevalence to 67.3% (AUC = 0.684) and to ≥54 reduced prevalence to 36.8% (AUC = 0.606), indicating that the original cutoff provided optimal discriminative performance (AUC = 0.800). Using the standard cutoff, independent risk factors included preschool age (OR = 0.196, 95% CI: 0.045 to 0.850), female sex (OR = 0.470, 95% CI: 0.221 to 0.997), caregiver college education or higher (OR = 1.359, 95% CI: 1.001 to 1.844), and caregiver sleep disturbance (OR = 2.552, 95% CI: 1.031 to 6.320) (all P < .05). Caregiver sleep disturbance remained the only consistently significant predictor across all CSHQ thresholds (OR range: 2.143 to 2.552, all P ≤ .043). The nomogram demonstrated good discrimination (training set AUC = 0.800, 95% CI: 0.711 to 0.888; test set AUC = .771, 95% CI: 0.644 to 0.898; bootstrap-corrected C-index = 0.800) and calibration (mean absolute error = 0.014). DCA confirmed clinical utility within a 25% to 75% threshold probability range.
Perioperative sleep disturbances are highly prevalent in pediatric surgical patients, influenced by both child and caregiver factors. The prevalence is threshold-dependent (range: 36.8% to 90.0%), and caregiver sleep disturbance emerged as the most robust predictor. The validated nomogram provides a clinically useful tool for preoperative risk stratification, enabling early identification and targeted intervention for high-risk children. Multicenter external validation and establishment of perioperative-specific CSHQ thresholds are recommended.
PMID:
42593388
Bibliographic data and abstract were imported from PubMed on 13 Aug 2026.
Read full publication at:
Please sign in
to see all details.
Advertisement
Stats
- Recommendations n/a n/a positive of 0 vote(s)
- Views 8
- Comments 0