Concept of perioperative respiratory risk: a narrative review

Perioperative respiratory risk represents a major interdisciplinary challenge, as postoperative pulmonary complications (PPCs) are associated with increased mortality, prolonged hospital stay, and higher healthcare costs. Heterogeneity in the definitions of complications complicates the interpretation of the literature; initiatives such as Standardised Endpoints in Perioperative Medicine (StEP) and the European Society of Anaesthesiology working group on European definitions have proposed harmonised definitions and severity grading systems, thereby improving comparability across studies and clinical applicability of results. This narrative review synthesises data on the epidemiology of PPCs, patient- and procedure-related risk factors, methods of risk stratification and prevention, and presents contemporary prognostic models ranging from clinical risk scores to machine learning algorithms, with particular emphasis on validation requirements and reporting standards. Practical implications include the need to apply StEP definitions with severity grading alongside discrete registry-based outcomes according to the NSQIP (National Surgical Quality Improvement Program) classification, to perform external validation of predictive models, and to implement bundled preventive strategies throughout all phases of perioperative care in high-risk populations. Integration of standardised outcomes, validated risk models, and multistage preventive interventions, including the use of machine learning algorithms, provides a foundation for reducing the incidence of PPCs and improving long-term outcomes.