A novel nomogram for predicting the prolonged length of stay in post-anesthesia care unit after elective operation

• Published in: BMC anesthesiology Vol. 23; no. 1; p. 404
• Main Authors: Fang, Fuquan, Liu, Tiantian, Li, Jun, Yang, Yanchang, Hang, Wenxin, Yan, Dandan, Ye, Sujuan, Wu, Pin, Hu, Yuhan, Hu, Zhiyong
• Format: Journal Article
• Language: English
• Published: England BioMed Central Ltd 07-12-2023; BioMed Central; BMC
• Subjects: Age; Ambulatory care; Analysis; Anesthesia; Blood transfusion; Body mass index; Calibration; Cancer; Care and treatment; Cholecystectomy; Delirium; Elective surgery; Elective Surgical Procedures; Enhanced recovery after surgery; Hospitals; Humans; Hypertension; Length of Stay; Logistic Models; Medical care; Medical personnel; Medical research; Medical societies; Medicine, Experimental; Naloxone; Nomogram; Nomograms; Nurses; Pain; Patients; Population; Post-anesthesia care unit; Predictions; Quality management; Recovery (Medical); Regression analysis; Risk factors; Surgery; Thyroid gland; Variables; Anesthesia; Enhanced recovery after surgery; Nomogram; Post-anesthesia care unit; Risk factors
• ISSN: 1471-2253; 1471-2253
• DOI: 10.1186/s12871-023-02365-w

Prolonged length of stay in post-anesthesia care unit (PLOS in PACU) is a combination of risk factors and complications that can compromise quality of care and operating room efficiency. Our study aimed to develop a nomogram to predict PLOS in PACU of patients undergoing elective surgery.
Data from 24017 patients were collected. Least absolute shrinkage and selection operator (LASSO) was used to screen variables. A logistic regression model was built on variables determined by a combined method of forward selection and backward elimination. Nomogram was designed with the model. The nomogram performance was evaluated with the area under the receiver operating characteristic curve (AUC) for discrimination, calibration plot for consistency between predictions and actuality, and decision curve analysis (DCA) for clinical application value.
A nomogram was established based on the selected ten variables, including age, BMI < 21 kg/m
, American society of Anesthesiologists Physical Status (ASA), surgery type, chill, delirium, pain, naloxone, operation duration and blood transfusion. The C-index value was 0.773 [95% confidence interval (CI) = 0.765 - 0.781] in the development set and 0.757 (95% CI = 0.744-0.770) in the validation set. The AUC was > 0.75 for the prediction of PLOS in PACU. The calibration curves revealed high consistencies between the predicted and actual probability. The DCA showed that if the threshold probability is over 10% , using the models to predict PLOS in PACU and implement intervention adds more benefit.
This study presented a nomogram to facilitate individualized prediction of PLOS in PACU for patients undergoing elective surgery.