Perioperative Risk Calculator (PROTECT)

Machine-learning Model for Perioperative Risk Calculation

The aim of this project is to develop a machine-learning model for calculating the risk of postoperative complications. In addition to the data collected during the premedication, the model will include all intraoperative values recorded in the Patient Data Management System (PDMS), which include not only vital and respiratory parameters, but also medication and doses, intraoperative events and times. Postoperative complications are defined according to their severity according to the Clavien-Dindo score (Dindo, Demartines et al., 2004) and are collected from the data available in the health information system (HIS).

The machine-learning model is created using an extreme-gradient boosting algorithm which has been updated with new data from the year 2021 to ensure accuracy of the model.

Study Overview

• Status: Active, not recruiting
• Conditions: Perioperative/Postoperative Complications

• Study Type: Observational
• Enrollment (Actual): 175559

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Child; Adult; Older Adult
• Accepts Healthy Volunteers: No

• Sampling Method: Non-Probability Sample

Study Population

Patients who underwent surgical interventions with general or regional anesthesia at Klinikum rechts der Isar, Munich after the implementation of an electronic patient data management system in May 2014

Description

Inclusion Criteria:

• all patients who underwent surgery with anesthesia

Exclusion Criteria:

• none

Study Plan

How is the study designed?

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
postoperative complications	Postoperative complications are classified by means of the Clavien-Dindo-Score. The Clavien-Dindo-Score describes classes of severity of postoperative complications: Grade I: any deviation from the normal postoperative course without the need for pharmacological treatment or surgical, endoscopic and radiological interventions Grade II: requiring pharmacological treatment Grade IIIa: requiring surgical, endoscopic or radiological intervention not under general anesthesia Grade IIIb: requiring surgical, endoscopic or radiological intervention under general anesthesia Grade IVa: single organ dysfunction Grade IVb: multiorgandysfunction Grade V: death of a patient	30 days

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
in-hospital mortality	mortality within hospital stay	30 days

Collaborators and Investigators

• Sponsor: Technical University of Munich
• Collaborators: Health Information Management, Belgium

Study record dates

Study Major Dates

• Study Start (Actual): May 1, 2014
• Primary Completion (Actual): September 30, 2022
• Study Completion (Estimated): December 31, 2024

Study Registration Dates

• First Submitted: July 8, 2019
• First Submitted That Met QC Criteria: September 16, 2019
• First Posted (Actual): September 17, 2019

Study Record Updates

• Last Update Posted (Actual): November 22, 2023
• Last Update Submitted That Met QC Criteria: November 21, 2023
• Last Verified: October 1, 2023

More Information

Terms related to this study

• Keywords: anesthesia; machine learning; decision support; risk calculator
• Additional Relevant MeSH Terms: Pathologic Processes; Postoperative Complications
• Other Study ID Numbers: 253/19

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No