Predictive Time-to-Event Model for Major Medical Complications After Colectomy

University of British Columbia

Status: Unknown

Conditions: Complications, Postoperative; Predictive Model; Colectomy; Inflammatory Bowel Diseases; Diverticulitis; Colorectal Cancer

Treatments: Other: No Intervention

Study type: Observational

Funder types: Other

Identifiers

NCT05150548

H21-02670

Details and patient eligibility

About

Purpose: The purpose of this study is to create prediction models for when major complications occur after elective colectomy surgery.

Justification: After surgery, patients can have multiple complications. Accurate risk prediction after surgery is important for determining an appropriate level of monitoring and facilitating patient recovery at home.

Objectives: Investigators aim to develop and internally validate prediction models to predict time-to-complication for each individual major medical complications (pneumonia, myocardial infarction (MI) (i.e. heart attacks), cerebral vascular event (CVA) (i.e. stroke), venous thromboembolism (VTE) (i.e. clots), acute renal failure (ARF) (i.e. kidney failure), and sepsis (i.e. severe infections)) or adverse outcomes (mortality, readmission) within 30-days after elective colectomy.

Data analysis: Investigators will be analyzing a data set provided by the National Surgical Quality Improvement Program (NSQIP). Descriptive statistics will be performed. Cox proportional hazard and machine learning models will be created for each complication and outcome outlined in "Objectives". The performances of the models will be assessed and compared to each other.

Full description

Background: Planned (elective or time sensitive) colectomy are performed for indications including cancer, inflammatory bowel disease (IBD), and diverticulitis. After colectomy, patients are at risk of a variety of major medical complications, including pneumonia, myocardial infarction (MI), cerebral vascular event (CVA), venous thromboembolism (VTE), acute renal failure (ARF), and sepsis. However, different complications tend to happen at different times after surgery. Accurate risk prediction, not only whether a complication may occur in a patient, but also when, is crucial for patient education, monitoring, and disposition planning. While several studies have explored the incidence and binary risk prediction for major complications after surgeries, there has been scarce literature on time-to-complication prediction modeling in recent population cohort data.

Objectives

1. To develop and internally validate Cox proportional hazards models to predict time-to-complication for each individual major medical complication captured in the American College of Surgeons National Surgical Quality Improvement Program (NSQIP) dataset (pneumonia, myocardial infarction (MI), cerebral vascular event (CVA), venous thromboembolism (VTE), acute renal failure (ARF), and sepsis) or adverse outcomes (mortality, readmission), that started within 30-days after elective colectomy.
2. To develop and internally validate machine learning models to predict time-to-complication for major medical complications and adverse outcomes (same as in objective 1) within 30-days after elective colectomy in NSQIP. The best machine learning model for each complication will be compared to the Cox proportional hazards model in terms of discrimination, and calibration.

Methods: Investigators will conduct a time-to-event survival analysis in a retrospective cohort using NSQIP®, a prospectively-collected multicentre dataset with more than 150 clinical variables within 30 days after surgery. This dataset includes information on whether the patient was diagnosed with major complications (in- or out-of-hospital) as well as the number of postoperative days to the diagnoses of complications, as defined by a standardized criteria within the NSQIP operations manual. The general dataset will be linked with the NSQIP® Procedure Targeted Colectomy dataset, which contains additional colectomy-specific variables.

The study will be reported according to the Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD) guidelines and Guidelines for Developing and Reporting Machine Learning Predictive Models in Biomedical Research.

Enrollment: 130,000 estimated patients

Sex: All

Ages: 18+ years old

Volunteers: Accepts Healthy Volunteers

Inclusion criteria

• undergoing elective colectomy
• data has been collected in the NSQIP® Procedure Targeted Colectomy dataset from 2014-2019

Exclusion criteria

• American Society of Anesthesiologists (ASA) Physical Status (PS) V (defined as "5-Moribund") (ASA PS 6 - organ donation is not included within NSQIP)
• undergoing urgent or emergency surgery
• indication for colectomy consisting of "Acute diverticulitis", "Enterocolitis (e.g. C. Difficile)", and "Volvulus" due to the non-elective nature of these pathologies
• patient with disseminated cancer
• wound infection (i.e. potentially recent surgery)
• systemic sepsis
• ventilator-dependence preoperatively

Trial design

130,000 participants in 1 patient group

Entire Cohort

Description:

Patients undergoing elective colectomy with data that has been collected in the NSQIP® Procedure Targeted Colectomy dataset from 2014-2019 with American Society of Anesthesiologists (ASA) Physical Status I-IV. Patients will not be included in this cohort with urgent or emergency colectomy or indication for colectomy consisting of "Acute diverticulitis", "Enterocolitis (e.g. C. Difficile)", and "Volvulus", patients with disseminated cancer, wound infection, systemic sepsis or ventilator-dependence preoperatively.

Treatment:

Other: No Intervention