Although anaesthesiologists strive to avoid hypoxemia during surgery, reliably predicting future intraoperative hypoxemia is not currently possible. Here, we report the development and testing of a machine-learning-based system that, in real time during general anaesthesia, predicts the risk of hypoxemia and provides explanations of the risk factors. The system, which was trained on minute-by-minute data from the electronic medical records of over fifty thousand surgeries, improved the performance of anaesthesiologists when providing interpretable hypoxemia risks and contributing factors. The explanations for the predictions are broadly consistent with the literature and with prior knowledge from anaesthesiologists. Our results suggest that if anaesthesiologists currently anticipate 15% of hypoxemia events, with this system’s assistance they would anticipate 30% of them, a large portion of which may benefit from early intervention because they are associated with modifiable factors. The system can help improve the clinical understanding of hypoxemia risk during anaesthesia care by providing general insights into the exact changes in risk induced by certain patient or procedure characteristics.

One Sentence Summary: We present a new machine learning based system called Prescience that provides interpretable real-time predictions to help anesthesiologists prevent hypoxemia during surgery.Abstract: Hypoxemia causes serious patient harm, and while anesthesiologists strive to avoid hypoxemia during surgery, anesthesiologists are not reliably able to predict which patients will have intraoperative hypoxemia. Using minute by minute EMR data from fifty thousand surgeries we developed and tested a machine learning based system called Prescience that predicts real-time hypoxemia risk and presents an explanation of factors contributing to that risk during general anesthesia. Prescience improved anesthesiologists' performance when providing interpretable hypoxemia risks with contributing factors. The results suggest that if anesthesiologists currently anticipate 15% of events, then with Prescience assistance they could anticipate 30% of events or an estimated additional 2.4 million annually in the US, a large portion of which may be preventable because they are attributable to modifiable factors. The prediction explanations are broadly consistent with the literature and anesthesiologists' prior knowledge. Prescience can also improve clinical understanding of hypoxemia risk during anesthesia by providing general insights into the exact changes in risk induced by certain patient or procedure characteristics. Making predictions of complex medical machine learning models (such as Prescience) interpretable has broad applicability to other data-driven prediction tasks in medicine.peer-reviewed)

BACKGROUND: Opioids have been a central component of routine adult and pediatric anesthesia for decades. However, the long-term effects of perioperative opioids are concerning. Recent studies show a 4.8%–6.5% incidence of persistent opioid use after surgery in older children and adults. This means that >2 million of the 50 million patients undergoing elective surgeries in the United States each year are likely to develop persistent opioid use. With this in mind, anesthesiologists at Bellevue Clinic and Surgery Center assembled an interdisciplinary quality improvement team focused on 2 goals: (1) develop effective anesthesia protocols that minimize perioperative opioids and (2) add value to clinical services by maintaining or improving perioperative outcomes while reducing costs. This article describes our project and findings but does not attempt to make inferences or generalizations about populations outside our facility. METHODS: We performed a large-scale implementation of opioid-sparing protocols at our standalone pediatric clinic and ambulatory surgery facility, based in part on the prior success of our previously published tonsillectomy and adenoidectomy protocol. Multiple Plan-Do-Study-Act cycles were performed using data captured from the electronic medical record. The percentage of surgical patients receiving intraoperative opioids and postoperative morphine preintervention and postintervention were compared. The following measures were evaluated using statistical process control charts: maximum postoperative pain score, postoperative morphine rescue rate, total postanesthesia care unit minutes, total anesthesia minutes, and postoperative nausea and vomiting rescue rate. Intraoperative analgesic costs were calculated. RESULTS: Between January 2017 and June 2019, 10,948 surgeries were performed at Bellevue, with 10,733 cases included in the analyses. Between December 2017 and June 2019, intraoperative opioid administration at our institution decreased from 84% to 8%, and postoperative morphine administration declined from 11% to 6% using analgesics such as dexmedetomidine, nonsteroidal anti-inflammatory drugs, and regional anesthesia. Postoperative nausea and vomiting rescue rate decreased, while maximum postoperative pain scores, total anesthesia minutes, and total postanesthesia care unit minutes remained stable per control chart analyses. Costs improved. CONCLUSIONS: By utilizing dexmedetomidine, nonsteroidal anti-inflammatory drugs, and regional anesthesia for pediatric ambulatory surgeries at our facility, perioperative opioids were minimized without compromising patient outcomes or value.