EEG-derived pain threshold index for prediction of postoperative pain in patients undergoing laparoscopic urological surgery: a comparison with surgical pleth index.
Ruijing WangYixu DengShoujing ZhouJun ZhangPublished in: Journal of clinical monitoring and computing (2020)
Recently a novel pain recognition indicator derived from electroencephalogram(EEG) signals, pain threshold index(PTI) has been developed. The aim of this study was to determine whether PTI can be used for prediction of postoperative acute pain while surgical pleth index(SPI) applied as control. Eighty patients undergoing laparoscopic urological surgery under general anesthesia were enrolled. Data of SPI, PTI and a sedative index-wavelet index(WLI) were recorded within last 10 min at the end of surgery. The postoperative pain scores (NRS, numerical rating scale) were obtained. The Bland-Altman analysis was used for evaluation of consistency between PTI and SPI, whereas receiver-operating characteristic (ROC) curves was used for the mean values of PTI, SPI, and WLI to distinguish between mild (NRS 0-3) and moderate-severe (NRS 4-10) pain, and calculate their "best-fit" cut-off values. Data from 76 patients were included for final analysis. There was a good agreement between SPI and PTI values at the end of surgery. The ROC analysis showed a cut-off PTI value of 53 to discriminate between mild and moderate-to-severe pain, while SPI is 44 for this discrimination. Further analysis indicated that PTI had a best predictive accuracy reflected by highest area under curve (AUC)(0.772, 95% CI: 0.661-0.860)with sensitivity(62.50%) and specificity(90.91%) and a best positive predictive value(83.3%,95% CI: 68.4-98.2%). PTI obtained at the end of surgery, which have better predictive accuracy for postoperative pain than SPI, could differentiate the patients with moderate-to-severe pain from those with mild pain after they awaken from anesthesia.Clinical trial registration Chinese Clinical Trials Registry: ChiCTR1900024789.
Keyphrases
- postoperative pain
- chronic pain
- clinical trial
- pain management
- minimally invasive
- neuropathic pain
- patients undergoing
- coronary artery bypass
- randomized controlled trial
- end stage renal disease
- surgical site infection
- machine learning
- spinal cord injury
- ejection fraction
- intensive care unit
- big data
- study protocol
- chronic kidney disease
- artificial intelligence
- resting state
- phase ii
- extracorporeal membrane oxygenation