DVT, D-dimer and Clinical Decision Rules

April 2005

DVT, D-dimer and Clinical Decision Rules

Search Strategy: Utilizing OVID from 1996 until present you perform a search using the search terms deep venous thrombosis (including all subheadings 10,718 “hits”), pre-test probability (including the subheadings decision making, predictive value of test, diagnosis, probability, and sensitivity/specificity, 1,483,061 “hits”), and D-dimer (limiting to subheading diagnosis, 2117 “hits”). Combining the results for these three search terms, you are still left with 1,711 articles so you limit these to English language, systematic reviews, and core clinical journals. Among the 20 articles this leaves are three of the articles selected below. By serendipity you pick up this months Annals of Emergency Medicine and note article 2 below.

A 35 year old three-month post-partum ambulatory female presents to your ED complaining of left calf pain of two days duration. Her pregnancy was an uncomplicated, term, spontaneous vaginal delivery and she has no past history of blood clots, malignancy, recent surgery, or immobilization. She denies chest pain or dyspnea.

On physical exam you note the following vital signs: BP 120/80, P 76, R 18, T 36.7°, 98% on room air. She is obese with trace pedal edema bilaterally and no overlying cellulitis, calf edema, or tenderness along the deep venous system.

In formulating your differential diagnosis, you contemplate the possibility of DVT. Uncomfortable with the concept of pre-test probability or your innate ability to derive an estimate of disease prevalence in patients like this one, you turn to your ED computer to perform a literature search using EBM principles.


PICO Question

Population: Post-partum females with calf pain

Intervention: Evaluate DVT with or without empirical anti-coagulation

Comparison: Do not evaluate DVT

Outcome: Presence of DVT and/or PE within 3 months of index visit.


Years

First years: Usefulness of Clinical Prediction Rules for the Diagnosis of Venous Thromboembolism: A Systematic Review. Am J Med 2004; 117: 676-684.

Second years: Derivation and Validation of a Bayesian Network to Predict Pretest Probability of Venous Thromboembolism. Annals EM 2005; 45: 282-290.

Third years: Combined Use of Clinical Assessment and D-dimer to Improve the Management of Patients Presenting to the ED with Suspected DVT (EDITED Study). J Thrombosis & Haemostasis 2003; 1: 645-651.

Fourth years: Combined Use of Rapid D-dimer Testing and Estimation of Clinical Probability in the Diagnosis of DVT: Systematic Review. BMJ 2004; 329: 821-828.


Forms

Second, Third Year Residents: use the Clinical Decision Rule Critical Review Form

First, Fourth Year Residents: use the Meta-Analysis Critical Review Form


Articles

Article 1: Usefulness of Clinical Prediction Rules for the Diagnosis of Venous Thromboembolism: A Systematic Review. Am J Med 2004; 117: 676-684.
NO ANSWER

Article 2: Derivation and Validation of a Bayesian Network to Predict Pretest Probability of Venous Thromboembolism. Annals EM 2005; 45: 282-290.
ANSWER KEY

Article 3: Combined Use of Clinical Assessment & D-dimer to Improve the Management of Patients Presenting to the ED with Suspected DVT (EDITED Study), J Thrombosis & Haemostasis 2003; 1: 645-651.
ANSWER KEY

Article 4: Combined Use of Rapid D-dimer Testing and Estimation of Clinical Probability in the Diagnosis of DVT: A Systematic Review, BMJ 2004; 329: 821-828.
ANSWER KEY