Perioperative Predictive Value of Physical Activity on Short- and Long-term Morbidity and Mortality

Overview

Over 300 million surgeries are performed globally every year. Complications after surgery - infections, cardiovascular conditions, postoperative pulmonary complications and renal impairment - affect survival and quality of life. Age and co-morbidity are unmodifiable factors, contributing to increased risk of these perioperative complications. However, a modifiable risk factor is physical activity. This study aims to test if self reported physical activity has added predicted value, beyond established risk factors, for predicting perioperative morbidity and mortality.

Research question: This cohort study investigates if higher levels of self reported physical activity at preoperative assessment has added predicted value, beyond established risk factors, for predicting perioperative morbidity and mortality. Background: Previous studies of perioperative outcomes in high-income countries indicate that close to 20% had complications within 30 days after surgery, and that around 3% died within 1 yr after surgery. In multiple studies, postoperative complications massively increase risk of 1yr mortality. Whilst perioperative complications are under-reported, they affect length of stay and days at home up to 30 days after surgery (DAH30). DAH30 is a validated, patient-centered outcome measure with prognostic importance due to high sensitivity to changes in surgical risks and the impact of surgical complications. DAH365, the days at home up to one year after surgery, is also an important patient-centered outcome measure. Data collection: Age, sex, body mass index, co-morbid conditions (using ICD-codes and reported medication) as well as American Society of Anesthesiologists (ASA) physical status classification will be recorded. The predictor of interest: the Metabolic Equivalent of Task Score (MET-score), reported in the electronic health record by the attending anesthesiologist based on patient history in conjunction with the preoperative assessment. Analysis: The MET-score is the predictor of interest. The potential added predictive value of the MET-score, will be assed using a machine learning approach, by comparing it to other established predictive factors.

Conditions