OC 54.4 - Development and validation of a clinical prediction model for 90-day venous thromboembolism risk following total hip and total knee arthroplasty

Tuesday, July 12, 2022

11:30 AM – 11:45 AM

Location: ExCel Center, ICC Capital Suite Room 14-16

Presentation Format: In-Person

Presenting Author(s)

Banne Nemeth, MD, PhD

Postdoc
Department of Clinical Epidemiology and Department of Orthopedics, Leiden University Medical Center
Leiden, Zuid-Holland, Netherlands

Co-Author(s)

Mark J.R. Smeets, MD

PhD student
Department of Clinical Epidemiology, Leiden University Medical Center, The Netherlands
The Hague, Zuid-Holland, Netherlands
AP

Alma Becic Pedersen

Professor
Department of Clinical Epidemiology and Department of Clinical Medicine, Aarhus University
Aarhus, Midtjylland, Denmark
EK

Eskild Bendix Kristiansen

Statistician
Department of Clinical Epidemiology and Department of Clinical Medicine, Aarhus University
Aarhus, Midtjylland, Denmark
MW

Martin Whyte

Reader in metabolic medicine
University of Surrey
Surrey, England, United Kingdom
Lara N. Roberts, MD

Haematology Consultant
King's College Hospital NHS Foundation Trust
London, England, United Kingdom
Sd

Simon de Lusignan

Professor of primary care and clinical informatics
University of oxford
Oxford, England, United Kingdom
Sl

Saskia le Cessie

professor
Department of Clinical Epidemiology – LUMC – Netherlands and Department of biomedical data sciences – LUMC - Netherlands
Leiden, Zuid-Holland, Netherlands
SC

Suzanne C. Cannegieter, MD PhD

Professor
Department of Clinical Epidemiology, Leiden University Medical Center
Leiden, Zuid-Holland, Netherlands
Roopen Arya, MA PhD FRCP FRCPath

Professor of Thrombosis and Haemostasis
King's College Hospital NHS Foundation Trust
London, England, United Kingdom

Background: Patients undergoing total hip (THA) or total knee arthroplasty (TKA) are at increased risk for postoperative venous thromboembolism (VTE). About 1.0%-1.5% of patients still develop VTE despite thromboprophylaxis.

Aims: To develop and validate a clinical prediction model for 90-day VTE risk following THA/TKA.

Methods: For model development, data on all patients undergoing THA/TKA between 2010-2018 were used from the Royal College of General Practitioners Research and Surveillance Centre (data from >290 practices in the United Kingdom). Data were linked to the NHS Hospital Episode Statistics and Office of National Statistics. Patients with missing linkage between datasets or those on permanent therapeutic anticoagulants were excluded. For model validation, data were used from the Danish Hip and Knee Arthroplasty Registry from 2017-2019, where the same exclusion criteria applied. 40 candidate predictors were selected to develop a penalized logistic regression model. Discrimination and calibration analyses were performed to assess model performance.

Results: 64,770 patients were included in the development data, of whom 1.4% developed VTE within 90-days. The final prediction model consisted of age, sex, body mass index, history of coronary artery disease, history of lung disease, cystitis in the year before surgery, history of phlebitis, presence of varicose veins, history of VTE and type of procedure (THA/TKA). The Area Under Curve was 0.65 (95%Confidence Interval (CI) 0.63-0.67). 40,646 patients underwent THA/TKA in the validation cohort, 1.0% developed VTE. Here, the AUC was 0.63 (95%CI 0.61-0.66). The model showed good agreement between observed and predicted risks following recalibration of the model intercept (Figure 1 and Table 1).

Conclusion(s): This clinical prediction model for 90-day VTE risk following THA/TKA performed well in both development and validation data. This model may be used to identify patients at either low or high risk for VTE following THA/TKA.

Funding: The Netherlands Thrombosis Foundation and Sanofi.