Paradoxical Association of Body Mass Index With Occurrence and Mortality of Venous Thromboembolism in Hospitalized Patients: A National Hospital-Based Cohort Analysis.

BACKGROUND

The association between body mass index (BMI) and venous thromboembolism (VTE) occurrence and mortality is controversial. This study aimed to investigate the association of BMI with VTE occurrence and mortality and explore the consistency across age, sex, and medical specialty.

METHODS

This study included 3 199 526 inpatients at potential risk for VTE from a nationwide hospital-based cohort. Participants were categorized as lean weight (BMI <18.5 kg/m2), normal weight (18.5 ≤BMI<24.0 kg/m2), overweight (24.0 ≤BMI <28.0 kg/m2), and obesity (BMI ≥28.0 kg/m2). Odds ratios (ORs) were calculated by logistic regression models.

RESULTS

In all participants, 59 002 (1.8%) developed VTE during hospitalization. After adjusting for the confounding factors, compared with normal weight, the risk of VTE occurrence reduced by 18% (odds ratio [OR], 0.82 [95% CI, 0.80-0.83]) in participants with overweight and 22% (OR, 0.78,[95% CI, 0.75-0.81]) in participants with obesity. The protective association of overweight/obesity was more pronounced in age ≥65 years, female patients, and surgery participants (Pfor interaction <0.05), whereas in internal medical participants, obesity was associated with a higher risk of VTE occurrence (OR, 1.11 [95% CI, 1.01-1.21]). A total of 636 (1.1%) patients with VTE died before discharge. No significant association was observed between overweight/obesity and post-VTE mortality. However, in the surgical subgroup, obesity was associated with a lower risk of post-VTE mortality (OR, 0.37 [95% CI, 0.17-0.80], Pfor interaction <0.05).

CONCLUSION

An inverse association between BMI and VTE occurrence was observed among at-risk inpatients, most pronounced in older, female, and surgical subgroups. Notably, obesity remained a risk factor in medical patients, whereas among surgical patients who developed VTE, it was paradoxically associated with lower mortality.