Validating Assessment Models for Predicting Risk of Venous Thromboembolism

Identifying patients with higher risk of venous thromboembolism (VTE) is crucial to guide prophylaxis. As such, researchers sought to evaluate the validity of the Padua Prediction Score and the Caprini risk assessment model (RAM) for estimating the risk of VTE in inpatients with acute exacerbation of chronic obstructive pulmonary disease (AECOPD). According to lead author, Chen Zhou, PhD, and colleagues, the Caprini RAM demonstrated superior accuracy compared to the Padua RAM in identifying inpatients with AECOPD with higher risk of VTE.

The study was performed at seven medical centers in China, where 3,288 inpatients with AECOPD were enrolled. Caprini and Padua RAM scores were determined at initial admission. The primary outcome was the three–month incidence of VTE.

During the three–month follow–up, 128 patients (3.9%) developed VTE. Based on the Caprini risk modelling, the study population was categorized as follows: high, 53.6%; moderate, 43.0%, and low, 3.5%. The incidence of VTE aligned with the Caprini risk modelling, with the following rates: high, 6.1%; moderate, 1.5%, and low, 0%.

Notably, under the Padua RAM stratification, only 10.9% of the study population was classified as high risk, with the remaining 89.1% categorized as low risk. The corresponding incidence rates of VTE for the Padua groups were 7.9% and 3.4%, respectively. The investigators reported that the Caprini RAM had a higher area under curve relative to the Padua RAM (0.713  0.021 vs. 0.644 ± 0.023, P = 0.029).

The study’s authors ultimately concluded that the Caprini RAM classifications were more valid than the Padua groupings in identifying patients at higher risk of VTE, and suggested that the use of Caprini RAM and “might better guide thromboprophylaxis in these patients.”