Abstract

Introduction: Venous thromboembolism (VTE) that includes deep vein thrombosis (DVT) and pulmonary embolism (PE) is a common yet challenging diagnosis requiring the utilization of expensive imaging tests like duplex ultrasounds (DUS), ventilation perfusion (V/Q) scans and CT pulmonary angiograms (CTPA) for confirmation. Several management studies have shown that a low pre-test probability (PTP) score and a negative D-dimer (D-d) can safely rule out VTE in outpatients leading to increased efficiency and potential cost savings. However real-life data on the use of D-d have shown that it is often used inappropriately (in non-low probability patients, low sensitivity assays used, etc) and the availability of a simple blood test may lower the threshold for suspecting VTE with a resultant increase in the number of radiological investigations.

Methods: Our objectives were to introduce D-d with mandatory PTP testing within our Emergency Department (ED) to ensure it was utilized only in low clinical probability patients and assess if using D-d appropriately would result in decreased imaging tests and potential cost savings. After consulting with clinicians from the ED, thromboembolism, radiology and laboratory departments, a high sensitivity D-d (HemosIL D-dimer, Instrumentation Laboratory) along with the Wells score (for DVT and PE) was introduced formally to our ED on April 2, 2007. Separate algorithms were developed to investigate DVT and PE allowing ED physicians to compute the PTP score and then follow a suggested diagnostic work-up. Low PTP patients could have a D-d ordered, but only if the score sheet accompanied the sample to the laboratory. D-d samples without a score sheet were not processed until the sheet was sent. Score sheets were systematically filed in the laboratory. Patients with moderate and high PTP proceeded without a D-d to imaging. We defined 2 cohorts of patients

  1. a pre-intervention cohort that included all ED visits for suspected VTE in the year prior to the introduction of mandatory PTP and D-d

  2. a post-intervention cohort that included all ED visits for suspected VTE in the year after the introduction of mandatory PTP and D-d.

Data was obtained from 2 sources: the laboratory provided all PTP forms for review and a list of all ED patients with D-d assays (2) health records provided a list of all ED patients with procedure codes for DUS, CTPA and V/Q scans in our time frame. The lists were merged and duplicate entries removed. Retrospective chart review was carried out for all entries (n=3499) to exclude those in whom D-d testing or imaging was not done for VTE (n=1714). Patients with multiple visits were reviewed to determine if their visit represented the same or recurrent episode of VTE.

Results: From April 2, 2006 to April 1, 2007 (the pre intervention phase) there were 41,193 ED visits of which 789 visits had associated investigations for VTE. Mandatory PTP and D-d testing for low PTP patients began on April 2, 2007. From April 2, 2007 to April 1, 2008 (the post intervention phase) there were 40,680 ED visits of which 996 visits had associated investigations for VTE. The proportion of visits associated with VTE investigations increased from 1.9% to 2.3% (p<0.001) after introducing PTP plus D-d. The total number of D-d ordered understandably increased significantly from 85 pre intervention to 449 post intervention. Imaging tests for VTE (DUS, CTPA and V/Q scans) were unchanged from 818 pre intervention to 783 post intervention (2% vs. 1.9%, p= 0.25). Of the 449 D-d ordered in the post intervention period, 95% (n=428) had a PTP form as “mandated” by our algorithm. 405 (95%) were scored as low probability, 2% were moderate-high probability and 3% were illegible or inaccurate. Hence, compliance with our mandatory algorithm occurred in 90% of visits. Only 53% of low PTP patients had a normal D-d result.

Conclusions: We have shown that a simple strategy of mandating D-d use with a standardized PTP assessment via the laboratory to ensure appropriate use of D-d testing is possible, and associated with a high compliance rate among ED physicians. However, the availability of a simple blood test did lower the threshold at which VTE is considered increasing the number of patients investigated and resulting in no reduction in imaging for VTE or potential cost savings. Having a standardized approach to VTE investigation may be valuable to ED physicians and is being assessed as a qualitative outcome of this study.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author