• 
• 

Three Part Question

In [patients presenting with a swollen upper limb] can a [negative d-dimer] safely [exclude an upper extremity deep vein thrombosis]?

Clinical Scenario

A 27 year old lady presents to A&E with a one week history of worsening whole right upper limb swelling. She admits to lots of heavy lifting days prior to the swelling commencing. No acute injury can be found and she has no previous medical history. You are concerned she may have an UEDVT. Given she has no risk factors, you consider carrying out a d-dimer to exclude the diagnosis.

Search Strategy

Medline 1946 to week 2 2012 using OvidSP
Medline 1946 to week 2 2012 using OvidSP Embase 1980 to week 2 2012 using OvidSP PubMed search done on 9th January 2012 Ovid SP: ({Upper limb deep vein thrombosis.mp} OR {Upper limb DVT.mp} OR {Arm deep vein thrombosis.mp} OR {Arm DVT.mp} OR {Upper extremity deep vein thrombosis.mp} OR {Upper extremity DVT.mp}) AND (d dimer.mp) PubMed Advanced Search via Search Builder: ‘Upper Extremity Deep Vein Thrombosis’ (MeSH Terms) AND ‘d dimer’ (MeSH Terms).

Search Outcome

Ovid SP Embase 10 articles.
Ovid SP Medline 3 articles.
PubMed 3 articles.

1 article was relevant.
2 further articles were broad systematic reviews on diagnosis of UEDVTs. Both of these solely quote the same, Merminod et al paper, when discussing the role of d dimers.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Merminod et al 2006 Switzerland	52 consecutive patients, over 18 years old, with suspected UEDVT. 23 males, 24 outpatients and median age 61 years. Blood was taken from the antecubital fossa and d dimer was measured using enzyme linked immunosorbent assay test. Diagnosis was confirmed with the aid of duplex ultrasonography (DUS), except in 5 patients where computed tomography (CT) was used.	Prospective unblinded diagnostic study.	Primary outcome measured was presence of d dimer level above 500mcg/L	15/52 (29%) patients in total had UEDVT diagnosed. Cancer patients included in this study,and those with CVC in situ had similar prevalence: 8/23 patients(30%) and 6/18 patients(33%) respectively. This showed an overall 100% sensitivity (95% CI 78-100) and 14% specificity (CI 4-29).	Small sample size hence large 95% CI. Little detail on location and methods of patient recruitment. Not clear on how soon after suspected diagnosis the d dimer level was measured. No mention on why some patients had final diagnosis with DUS and others with CT. High number of cancer patients included thus low specificity.

Comment(s)

There are no papers looking at the accuracy of d dimers in low risk/primary UEDVT. This subgroup would probably report a better specificity due the exclusion of those with malignancy. Given that upto 80% UEDVTs are thought to be secondary (Mai et al, 2011) it will be difficult to recruit the number of patients required for such a study.

Clinical Bottom Line

There is not enough evidence at present to comment on the usefulness of d dimer in UEDVT.

References

1. Merminod T, Pellicciotta S, Bounameaux H, et al. Limited usefulness of d dimer in suspected deep vein thrombosis of the upper extremities. Blood Coagulation and Fibrinolysis 2006, 17:225-227.