Elsevier

Clinical Radiology

Volume 63, Issue 3, March 2008, Pages 299-304
Clinical Radiology

Diagnostic value of CT for deep vein thrombosis: results of a systematic review and meta-analysis

https://doi.org/10.1016/j.crad.2007.09.010Get rights and content

Aim

To estimate the sensitivity and specificity of computed tomography (CT) for the diagnosis of deep vein thrombosis (DVT) in patients with suspected DVT and pulmonary embolus (PE).

Materials and methods

A search of the medical literature and citation lists was undertaken, and selected studies comparing CT to a reference standard in patients with suspected DVT or PE were retrieved. Data were analysed by random effects meta-analysis.

Results

Thirteen articles were included in the meta-analysis. Most compared CT to ultrasound in patients with clinically suspected PE. The sensitivity ranged from 71–100%, while specificity ranged from 93–100%. The pooled estimate of sensitivity was 95.9% (95% CI 93 to 97.8%) and the pooled estimate of specificity was 95.2% (93.6 to 96.5%). However, pooled estimates should be interpreted with caution as these were subject to significant heterogeneity (p = 0.025 and p < 0.001, respectively). Most studies only appeared to report proximal DVT. Too few data were available to estimate sensitivity for distal DVT.

Conclusions

CT has a similar sensitivity and specificity to ultrasound in patients with suspected PE where investigation of suspected DVT is required. There is insufficient research to determine the diagnostic accuracy of CT in patients with suspected DVT alone.

Introduction

Deep vein thrombosis (DVT) has an annual incidence of around 50 per 100,000 in western populations.1 If left untreated, DVT can lead to potentially fatal pulmonary embolus (PE) or can cause morbidity due to the development of post-thrombotic syndrome; therefore, accurate diagnosis is essential. Contrast venography has traditionally been considered the reference standard test for DVT, with near-perfect sensitivity and specificity, but, due to the invasive nature, a 10–20% failure rate and costs of the test,2, 3 it has gradually been replaced by ultrasound, which has a sensitivity of 97% for proximal DVT, 73% for distal DVT, and a specificity of 96%.4

However, there are a number of patients in whom ultrasound is technically impossible or where an indeterminate result is obtained, for whom alternative tests are needed. Computed tomography (CT) has the potential to fulfil this role. Furthermore, patients with venous thromboembolism (VTE) may present with either suspected DVT, PE, or both. At present, this would often require two separate diagnostic tests to diagnose both conditions. Given the increasing role of CT pulmonary angiography in diagnosing PE, it has been suggested that it would be more convenient and cost-effective to use CT venography in the same setting.

If CT is to be used as a definitive diagnostic test for DVT, it is necessary to evaluate whether CT is sufficiently accurate for this purpose. In order to estimate the sensitivity and specificity of CT in the diagnosis of DVT a systematic review and meta-analysis of studies using CT for the diagnosis of DVT was conducted.

Section snippets

Materials and methods

The aim of the present study was to identify all diagnostic cohort studies of patients with clinically suspected DVT or clinically suspected PE who underwent testing for DVT using CT followed by ultrasonography or venography. The following databases were searched: Medline, EMBASE, CINAHL, Web of Science, Cochrane Database of Systematic Reviews, Cochrane Controlled Trials Register, Database of Reviews of Effectiveness, and ACP Journal Club using the search strategy outlined in the Appendix.

Results

The literature search generated 2225 titles and abstracts for review. From these, 43 potentially relevant articles were selected for full review (kappa = 0.76). Eighteen of these articles were identified that met the inclusion criteria (kappa = 0.93). Two of these appeared to include some duplication of data and were excluded.7, 8 It was not possible to extract or analyse appropriate data from a further three articles, despite attempts to contact the authors.9, 10, 11 Review of the bibliographies

Discussion

The present meta-analysis indicates that CT imaging appears to have excellent sensitivity and specificity for proximal DVT, with pooled sensitivity and specificity close to that of ultrasound.4 However, the limitations of these data need to be appreciated. First, most of the studies used ultrasound as a reference standard. As ultrasound does not have perfect sensitivity and specificity this may lead to overestimation of the diagnostic accuracy of CT, if ultrasound and CT misclassify cases in

Acknowledgements

The authors thank Kathryn Paulucy for her clerical assistance and Angie Ryan for her help with the literature searches. This project was funded by the United Kingdom Health Technology Assessment R&D Programme (reference number 02/03/01).

References (25)

  • P.A. Loud et al.

    Combined CT venography and pulmonary angiography in suspected thromboembolic disease: diagnostic accuracy for deep venous evaluation

    Am J Roentgenol

    (2000)
  • M. Nicolas et al.

    Incremental lower extremity CT venography, a simplified approach for the diagnosis of phlebitis in patients with pulmonary embolism

    J Radiol

    (2001)
  • Cited by (89)

    • Diagnosis of deep vein thrombosis

      2020, Revista Clinica Espanola
    • Imaging of adult intestinal failure

      2019, Clinical Radiology
    View all citing articles on Scopus

    The views and opinions expressed therein are those of the authors and do not necessarily reflect those of the UK Department of Health.

    Present address: Carver College of Medicine, University of Iowa Hospitals and Clinics,Department of Radiology, 3895 JPP, 200 Hawkins Drive, Iowa City, IA 52242-1077, USA.

    View full text