# Predicting postoperative pain in children: an observational study using the pain threshold Index

## TL;DR

This study evaluates the pain threshold index's ability to predict postoperative pain in children and finds it more accurate when combined with other factors, especially in older children.

## Contribution

The study introduces a multivariable model combining PTI with clinical factors to improve postoperative pain prediction in children.

## Key findings

- PTI alone has limited predictive accuracy (AUC 0.659) for postoperative pain in children.
- A multivariable model combining PTI with gender, BMI, HR, and MAP improved predictive accuracy (AUC 0.768).
- PTI's accuracy increases with age, showing the highest AUC of 0.796 in children aged 9–16 years.

## Abstract

While the pain threshold index (PTI) holds potential as a tool for monitoring analgesia-pain equilibrium, its precision in forecasting postoperative pain in children remains unconfirmed. This study's primary aim was to assess the PTI's predictive precision for postoperative pain.

Children (aged 2–16 years) undergoing general surgery under general anesthesia were included. Within 5 min prior to the patient's emergence from surgery, data including PTI, wavelet index (WLI), heart rates (HR) and mean arterial pressure (MAP) were collected. Subsequently, a 15-min pain assessment was conducted following the patient's awakening. The accuracy of these indicators in discerning between mild and moderate to severe postoperative pain was evaluated through receiver operating characteristic (ROC) analysis.

The analysis encompassed data from 90 children. ROC analysis showed that PTI was slightly better than HR, MAP and WLI in predicting postoperative pain, but its predictive value was limited. The area under the curve (AUC) was 0.659 [0.537∼0.780] and the optimal threshold was 65[64–67]. Sensitivity and specificity were determined at 0.90 and 0.50, respectively. In a multivariable logistic regression model, a higher predictive accuracy was found for a multivariable predictor combining PTI values with gender, BMI, HR and MAP (AUC, 0.768; 95%CI, 0.669–0.866). Upon further scrutinizing the age groups, PTI's AUC was 0.796 for children aged 9–16, 0.656 for those aged 4–8, and 0.601 for younger individuals.

PTI, when used alone, lacks acceptable accuracy in predicting postoperative pain in children aged 2 to 16 years. However, when combined with other factors, it shows improved predictive accuracy. Notably, PTI appears to be more accurate in older children.