학술논문
Venous Thromboembolism Among People With HIV: Design, Implementation, and Findings of a Centralized Adjudication System in Clinical Care Sites Across the United States.
Document Type
Article
Author
Crane, Heidi M.; Nance, Robin M.; Ruderman, Stephanie A.; Haidar, Lara; Tenforde, Mark W.; Heckbert, Susan R.; Budoff, Matthew J.; Hahn, Andrew W.; Drumright, Lydia N.; Ma, Jimmy; Mixson, L. S.; Lober, William B.; Barnes, Gregory S.; McReynolds, Justin; Attia, Engi F.; Peter, Inga; Moges, Tesfaye; Bamford, Laura; Cachay, Edward; Mathews, William C.
Source
Subject
Language
ISSN
1525-4135
Abstract
Background: People with HIV (PWH) are at increased risk for venous thromboembolism (VTE). We conducted this study to characterize VTE including provoking factors among PWH in the current treatment era. Methods: We included PWH with VTE between 2010 and 2020 at 6 sites in the CFAR Network of Integrated Clinical Systems cohort. We ascertained for possible VTE using diagnosis, VTE-related imaging, and VTE-related procedure codes, followed by centralized adjudication of primary data by expert physician reviewers. We evaluated sensitivity and positive predictive value of VTE ascertainment approaches. VTEs were classified by type and anatomic location. Reviewers identified provoking factors such as hospitalizations, infections, and other potential predisposing factors such as smoking. Results: We identified 557 PWH with adjudicated VTE: 239 (43%) had pulmonary embolism with or without deep venous thrombosis, and 318 (57%) had deep venous thrombosis alone. Ascertainment with clinical diagnoses alone missed 6% of VTEs identified with multiple ascertainment approaches. DVTs not associated with intravenous lines were most often in the proximal lower extremities. Among PWH with VTE, common provoking factors included recent hospitalization (n = 134, 42%), infection (n = 133, 42%), and immobilization/bed rest (n = 78, 25%). Only 57 (10%) PWH had no provoking factor identified. Smoking (46%), HIV viremia (27%), and injection drug use (22%) were also common. Conclusions: We conducted a robust adjudication process that demonstrated the benefits of multiple ascertainment approaches followed by adjudication. Provoked VTEs were more common than unprovoked events. Nontraditional and modifiable potential predisposing factors such as viremia and smoking were common. [ABSTRACT FROM AUTHOR]