Elsevier

Thrombosis Research

Volume 142, June 2016, Pages 1-7
Thrombosis Research

Full Length Article
Is D-dimer used according to clinical algorithms in the diagnostic work-up of patients with suspicion of venous thromboembolism? A study in six European countries

https://doi.org/10.1016/j.thromres.2016.04.001Get rights and content

Highlights

  • Majority of the physicians estimated pre-test probability by clinical decision scores.

  • Standardised clinical algorithms were often not followed.

  • Both D-dimer and imaging were regularly requested initially.

  • Results indicated waste of resources and potential hazardous actions.

Abstract

Introduction

Clinical algorithms consisting of pre-test probability estimation and D-dimer testing are recommended in diagnostic work-up for suspected venous thromboembolism (VTE). The aim of this study was to explore how physicians working in emergency departments investigated patients suspected to have VTE.

Materials and methods

A questionnaire with two case histories related to the diagnosis of suspected pulmonary embolism (PE) (Case A) and deep venous thrombosis (DVT) (Case B) were sent to physicians in six European countries. The physicians were asked to estimate pre-test probability of VTE, and indicate their clinical actions.

Results

In total, 487 physicians were included. Sixty percent assessed pre-test probability of PE to be high in Case A, but 7% would still request only D-dimer and 11% would exclude PE if D-dimer was negative, which could be hazardous. Besides imaging, a D-dimer test was requested by 41%, which is a “waste of resources” (extra costs and efforts, no clinical benefit). For Case B, 92% assessed pre-test probability of DVT to be low. Correctly, only D-dimer was requested by 66% of the physicians, while 26% requested imaging, alone or in addition to D-dimer, which is a “waste of resources”.

Conclusions

These results should encourage scientific societies to improve the dissemination and knowledge of the current recommendations for the diagnosis of VTE.

Introduction

When reviewing a patient with suspected venous thromboembolism (VTE), i.e. pulmonary embolism (PE) or deep venous thromboembolism (DVT), it is recommended first to estimate the pre-test probability of PE/DVT by means of clinical signs and symptoms [1], [2], [3], [4], [5], [6]. This pre-test probability can be assessed either by using clinical experience or by calculating the probability based on scores i.e. clinical decision rules like the Wells scores for PE or DVT [5]. The rationale behind estimating the probability of VTE before further testing is to be able to select a safe and efficient diagnostic work-up based upon the recommended clinical algorithm [1], [2], [3], [4], [5], [6], [7], [8]. In patients with a low to moderate pre-test probability, a negative D-dimer is sufficient to exclude VTE, while in a patient with a higher pre-test probability, a negative D-dimer cannot exclude VTE, and the patient has to be referred to radiologic imaging to be able to exclude or confirm VTE. Patients with low to moderate pre-test probability and a positive D-dimer result, should also be referred to imaging [1], [2], [3], [4], [5], [6], [7], [8]. By using pre-test probability scores followed by the clinical algorithm described above, studies have shown that VTE can be ruled out without imaging in 30–40% of the patients where VTE is suspected [1], [3], [4], [5], [7], [8], [9], [10]. However, recent studies suggest that such pre-test probability scores and the clinical algorithm are not implemented in the routine diagnostic work-up of VTE in some countries [11], [12], [13], [14], [15], [16], [17]. The aim of this study was to explore how clinicians from several countries working in emergency departments investigate patients with suspected VTE in relation to their own estimated pre-test probabilities of VTE, and to compare this practice with the recommended clinical algorithm.

Section snippets

Methods

Two case histories followed by a questionnaire were developed by the members of the Working Group on Postanalytical Phase (WG-POST) of the European Federation of Clinical Chemistry and Laboratory Medicine (EFLM) and the European Organisation for External Quality Assurance Providers in Laboratory Medicine (EQALM), and were piloted by 1–5 physicians in Italy, Turkey, Norway, Hungary and Croatia. The case histories were slightly modified after the results from the pilot. During the summer of 2012,

Results

Altogether, 542 physicians responded to the questionnaire. General practitioners, laboratory physicians and medical students were excluded, leaving 487 physicians for further analyses. The response rate varied from 4.1% to 42.7% in the different countries (median 20.8%). Numerically, most responses were from Turkey and Norway (68%) (Table 1). Background characteristics of the physicians are outlined in Table 1. Eight percent did not answer all questions in Case A and B, and these physicians did

Case history A – clinical actions based on a negative D-dimer

Given a D-dimer of 0.45 mg/L (negative), 65% of physicians chose the “correct” clinical action based upon their own stated pre-test probability (Fig. 2A, Part II, green colours), 25% chose an action with “waste of resources” (Fig. 2A, Part II, yellow colour) and 11% an action which could be “hazardous” for the patient (Fig. 2A, Part II, red colour).

Discussion

The aim of this study was to explore how clinicians working in emergency departments investigated patients with suspected VTE in relation to their own estimated pre-test probabilities of VTE.

It is important to underline that the present study deals with the actions of the physicians after they have chosen a pre-test probability, and not if the pre-test probability chosen was correct based on the case history. The reason for not addressing the last question is that the case history can be

Conclusion

The findings in this study suggest that although pre-test probability scores are used, the recommended clinical algorithm following such scores, are not systematically adhered to by a substantial number of physicians. This could result in under- and over-diagnosing of patients, which can put the patients into significant risk [26]. Not following the algorithms also lead to an extra burden on the health care system by applying both D-dimer and imaging in a significant amount of the patients

Acknowledgements

We would like to thank all the national societies who kindly offered to recruit their member physicians to this study (Norwegian Society of Internal Medicine, Turkish Society of Internal Medicine, Hungarian Society of Emergency Medicine, Croatian Society of Cardiology, Italian Academy of Emergency Medicine Care, all the participating physicians and in addition Thomas H Røraas for statistical support and Silvia Cattaneo (EFLM) for SurveyMonkey support. We also wish to thank Andreas Hillarp and

References (32)

  • L.A. Linkins et al.

    Use of different D-dimer levels to exclude venous thromboembolism depending on clinical pretest probability

    J. Thromb. Haemost.

    (2004)
  • British Thoracic Society

    British Thoracic Society guidelines for the management of suspected acute pulmonary embolism

    Thorax

    (2003)
  • CLSI

    Quantitative D-dimer for the Exclusion of Venous Thromboembolic Disease; Approved Guideline CLSI Document H59-A

    (2011)
  • D.M. Keeling et al.

    The diagnosis of deep vein thrombosis in symptomatic outpatients and the potential for clinical assessment and D-dimer assays to reduce the need for diagnostic imaging

    Br. J. Haematol.

    (2004)
  • S.V. Konstantinides et al.

    D. Task Force for the, C. Management of Acute Pulmonary Embolism of the European Society of, 2014 ESC guidelines on the diagnosis and management of acute pulmonary embolism

    Eur. Heart J.

    (2014)
  • NICE

    Venous Thromboembolic Diseases: The Management of Venous Thromboembolic Diseases and the Role of Thrombophilia Testing. Clinical Guideline 144

    (2012)
  • Cited by (0)

    View full text