Risk Factors and Outcomes of Postoperative Pulmonary Embolism in Southeast-Asian Patients Undergoing Major Surgery: A Nested Case-Control Study.

INTRODUCTION: Postoperative pulmonary embolism (PE) remains a rare but catastrophic perioperative complication. Anesthesia-related factors, perioperative physiological instability, and delayed postoperative mobilization may contribute to its development and associated mortality. However, data on contemporary outcomes in Asian surgical populations are limited. METHODS: A retrospective nested case-control study of adult patients undergoing major surgery was performed. The source cohort comprised all 113,660 adults who underwent major surgery at Ramathibodi Hospital from January 2020 to December 2024. Patients with radiologically confirmed PE were identified and individually matched at a 1:3 ratio with age-matched controls. PE diagnosis was triggered by clinical suspicion based on presenting symptoms or hemodynamic signs. The independent risk factors for PE were identified using multivariate logistic regression. Survival was analyzed using the Kaplan-Meier method. RESULTS: Among 113,660 patients, 140 developed PE (an incidence of 0.123%). Independent predictors included active cancer (adjusted odds ratio [AOR]: 11.74), emergency surgery (AOR: 3.37), arthroplasty and fracture surgery (AOR: 2.38), immobilization ≥3 d (AOR: 2.33), and increasing Caprini score (AOR: 1.44). Early ambulation was independently associated with lower odds of PE (AOR: 0.093; 95% CI: 0.035-0.247), although this association may partly reflect confounding by indication. The 90-d mortality was markedly higher in the PE group (32.86% vs 2.14%, P < 0.001), with 26 of 46 deaths classified as PE-associated; PE-associated case fatality was highest among those diagnosed within 48 h. CONCLUSIONS: Postoperative PE was rare but carried substantial early mortality. Active cancer, emergency surgery, arthroplasty and fracture surgery, prolonged immobilization, and increasing Caprini score were independently associated with PE. Early ambulation was associated with lower odds, although this estimate is susceptible to confounding by indication and reverse causation and should not be interpreted as a treatment effect. These data support systematic preoperative Caprini stratification and warrant prospective evaluation of structured early-mobilization pathways as an adjunct to guideline-concordant thromboprophylaxis.