LAUSR

Dear Editor, The paper by Robertson et al1 (2020) is of great interest because a fundamental question concerning patients and haematologists after the initial 3-6 months of anticoagulation after an unprovoked venous thromboembolism (VTE) is regarding the optimal duration of anticoagulation. This paper is therefore of interest as to whether the age-adjusted D-dimer could help in the decision to withhold long-term anticoagulation when used as part of the DASH score. The question they have posed is one we have informally discussed many times in thrombophilia clinics. The authors found a substantial number of patients, if age-adjusted D-dimer was used, that would potentially have had low enough rates of VTE recurrence to withhold long-term anticoagulation when compared to the original DASH score which used non-age-adjusted D-dimer. We previously published a retrospective study regarding the use of the DASH score at our centre, finding a recurrence rate > 5% per annum with a DASH score of ≤12,3. In view of the paper by Robertson et al, we looked back at the original cohort of patients from our database to determine the VTE recurrence rates in patients ≥50 years of age that had a DASH score of ≤1, using the same methods as in our previous publication but by updating the censor date to 1 November 2018 and which are summarized here2. An age of ≥50 years was chosen arbitrarily to reflect the patient age group studied in the ADJUST-PE study for the age-adjusted D-dimer of 500 ug/L to ×10 upper limit of their age (however, in that study fibrinogen equivalent units (FEU) were used which corresponds to approximately the same reference range as our local D-dimer assay)4. Medical charts of patients seen in the thrombophilia clinic between 1 January 2013 and 31 December 2016 with a first episode of unprovoked or hormone-provoked VTE were reviewed. The censor date was VTE recurrence, death or 1 November 2018 (whichever occurred first). D-dimer testing was performed around 1 month after stopping anticoagulation. Patients were tested for antithrombin deficiency (but not routinely other heritable thrombophilias) and antiphospholipid antibodies as these can influence treatment. The local D-dimer reference range is <230 ng/mL (and assay used is HemosIL D-Dimer HS, Werfen, corresponding to D-dimer Units (D-DU); 1 ng/mL DDU ≈2 ng/mL FEU) which was used in this study to score 0 for the D-dimer in the DASH score. This local reference range is also the exclusion value used when investigating patients for suspected VTE. The primary outcome was recurrent VTE, which was diagnosed when there was objective radiological evidence of a new VTE. Distal DVT was excluded as the clinical relevance of this is debated, and not all patients are scanned for this. No patient with a D-dimer > 230 ng/mL was included. Statistical analysis was performed using Microsoft Excel 2010 (Microsoft) & R programme (version 3.3.3, R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing). Cumulative incidence was calculated for recurrent VTE adjusting for the competing risk of death and is displayed with 95% confidence intervals (CI). We identified 127 patients aged ≥50 years (at the time of the index VTE) that had a DASH score ≤ 1, and the characteristics are presented in table 1. In total, there were 368 patient-years of follow-up available. The cumulative incidence of recurrent VTE at 1 year was 8% (95% CI