PDF icon Download PDF
Thromb Haemost 1994; 72(05): 652-658
DOI: 10.1055/s-0038-1648936
Review Article
Schattauer GmbH Stuttgart

Factors Influencing the Reported Sensitivity and Specificity of Impedance Plethysmography for Proximal Deep Vein Thrombosis

Clive Kearon
The McMaster University and McMaster Medical Unit, Hamilton Civic Hospitals Research Center, Hamilton, Ontario, Canada
,
Jack Hirsh
The McMaster University and McMaster Medical Unit, Hamilton Civic Hospitals Research Center, Hamilton, Ontario, Canada
› Author Affiliations
Further Information

Publication History

Received 11 April 1994

Accepted 13 June 1994

Publication Date:
06 July 2018 (online)

 PDF Download  Permissions and Reprints
Preview

Summary

A number of recent studies have found a considerably lower sensitivity of the IPG for proximal DVT than originally reported. We reviewed the literature to try and identify reasons for the between study differences in sensitivity and specificity of the IPG for proximal DVT. A number of biases were identified which may have inflated the sensitivity reported by some earlier studies including: repeated IPG testing prior to venography; inclusion of data from patients used to derive the IPG discriminant line in the final calculation of sensitivity; and inclusion of patients with a known abnormal IPG in the study population. In addition, there is emerging evidence that, at least in some centers, the sensitivity of the IPG may have decreased due to a shift in the spectrum of proximal DVTs to smaller clots. Furthermore, rates of conversion during serial follow-up are considerably higher for IPG than for venous ultrasound, suggesting that IPG conversion is often due to extension of small missed proximal DVTs rather than just extension of calf vein clots. As the smaller proximal DVTs which IPG is prone to miss are likely to be clinically less important than more extensive clots, it is unclear whether the recently reported drop in sensitivity of the IPG places patients at increased risk. IPG users need to be aware of its limitations, and we recommend that additional testing is performed in patients with a high clinical suspicion of DVT who have a normal IPG, as the negative predictive value of IPG may be unacceptable in these circumstances.