External validation of the DASH prediction rule: a retrospective cohort study

Area of Science:

• Cardiology
• Hematology
• Clinical Medicine

Background:

• The DASH prediction model aims to identify patients at low risk for venous thromboembolism (VTE) recurrence.
• Validation in an independent cohort is crucial for establishing its clinical utility.
• Previous studies have not independently validated the DASH score's performance.

Purpose of the Study:

• To validate the calibration and discrimination of the DASH prediction model.
• To evaluate the DASH score's accuracy in predicting VTE recurrence.
• To assess the DASH score's performance in patients aged over 65 years.

Main Methods:

• A cohort of 827 patients with unprovoked deep vein thrombosis or pulmonary embolism was analyzed.
• The DASH score was calculated using D-dimer levels post-treatment and patient characteristics.
• Recurrence events were tracked using time-dependent analysis over 12 and 24 months.

Main Results:

• The DASH score showed good prediction accuracy, with a 3.6% recurrence risk at 1 year for scores ≤ 1.
• The model performed better in younger (< 65 years) subjects (c-statistic 0.72) compared to older subjects (c-statistic 0.54).
• Elderly patients (> 65 years) with the lowest DASH scores still exhibited a recurrence risk exceeding 5%.

Conclusions:

• The DASH prediction model is validated, particularly for younger individuals.
• The model's predictive accuracy is limited in elderly patients (> 65 years).
• Clinical management strategies may need adjustment for older VTE patients, even with low DASH scores.