Article Text

Download PDFPDF
  1. T Jaconelli,
  2. M Eragat,
  3. S Crane
  1. York, York District Hospital, York, UK


    Objectives & Background The standard work-up for patients suspected of having venous thromboembolism (VTE)-specifically deep vein thrombosis (DVT) and pulmonary embolism (PE) involves risk stratification, usually with the Wells score. This method of risk stratification dictates that patients who are deemed to have a low pre-test probability of VTE should have a serum D-dimer level taken. D-dimer values correlate positively with increasing age, this work looks at whether an age adjusted value can be used to safely exclude VTE.

    Table 1

    2x2 Table for Standard D-dimer Value

    Methods We designed a study protocol which aimed to identify the clinical outcomes of all patients at York Hospital who had a D-Dimer level measured in ED. Using retrospective data we estimated the proportion of patients who had investigations for suspected VTE, and the results of these investigations. The primary outcome was the diagnosis of VTE. We derived data for the sensitivity and specificity D-Dimer tests (in the diagnosis of VTE) using the current cut-off (230 ng/ml) and also for a proposed new cut-off 5× the age in those over 50 and 250 ng/ml up to the age of 50.

    Results From 01/11/13 to 21/03/14, 682 patients underwent D-dimer testing for suspected VTE (256 DVT and 426 PE). Those with a high wells score were excluded from analysis leaving 559 patients (156 DVT and 403 PE). Of this group 12 were diagnosed with DVT on ultrasound and 13 diagnosed with PE on CTPA; the rest had alternate diagnoses. 25 patients had positive D-dimers with the age adjusted value i.e. there were no false positives. With the age adjusted value 17 ultrasound scans would have been avoided and 28 patients would have PE safely ruled out.

    For the standard D-dimer the sensitivity was 100% (86.28–100 95% CI), specificity was 70.60% (66.53–74.43 95% CI) and the NPV was 100% (99.03–100). For the age adjusted value the sensitivity was 100% (86.28–100 95% CI), specificity was 79.03% (75.32&–82.40% 95% CI) and the NPV was 100% (99.13%–100% 95% CI).

    Conclusion The data shows that in patients suspected of having a VTE with a low pretest possibility, an age adjusted D-dimer increases the specificity of the test with no effect of the sensitivity when compared to the conventional D-dimer threshold. In practice this means that more patients can be safely discharged from the emergency department without further, unnecessary investigations. This may result in reduced hospital admissions, radiology time and iatrogenic harm from CTPA.

    • emergency departments

    Statistics from

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.