Perioperative medicine & outcomes
Predictive validity of chronic obstructive pulmonary disease phenotypes in inpatient elective surgery: a population-based study
Authors' conclusion
COPD phenotypes are predictive of postoperative survival and improve perioperative risk stratification.
Study context
- Population
- Older adults (≥65 yr) with chronic obstructive pulmonary disease (COPD) undergoing inpatient elective surgery, Ontario, Canada
- Participants
- 116,757 adults
- Exposure
- COPD phenotypes vs COPD alone; baseline model with vs without COPD phenotype
- Outcome
- Survival in the year after surgery (Cox models: added predictive performance of COPD phenotype over baseline model)
- Follow-up
- Year after surgery
Methodological detailspopulation-based, nested Cox proportional hazards models
- Population-based
- Nested Cox proportional hazards models
Main finding
As reported in the abstract
Decreased postoperative survival (hazard of death) by COPD phenotype vs COPD alone
- Advanced COPD · markedly decreased survival
- aHR for death (decreased survival) 5.59
- COPD with frailty · markedly decreased survival
- aHR for death (decreased survival) 3.56
- COPD with frequent exacerbation · moderately decreased survival
- aHR for death (decreased survival) 1.45
- COPD with cardiovascular comorbidity · moderately decreased survival
- aHR for death (decreased survival) 1.35
Observational data: associations, not effects.
Further findings
Model performance: baseline model with vs without COPD phenotype
- Improved model fit with COPD phenotype (likelihood ratio test)
- P < 0.001
- Discrimination, C-index, with vs without phenotype
- 0.775 vs 0.720
- Calibration, integrated calibration index, with vs without phenotype
- 0.035 vs 0.043
- Improved net benefit across all decision thresholds
Original publication
Predictive validity of chronic obstructive pulmonary disease phenotypes in inpatient elective surgery: a population-based study. Br J Anaesth. Published 31 July 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →