Background: Asymptomatic deep vein thrombosis (DVT) occurs in up to 11% of medical inpatients. The incidence of asymptomatic DVT among patients with inflammatory bowel disease (IBD) is unknown but may be even higher. D-dimer is effective for DVT screening, but its utility has not been studied in the IBD population.
Methods: Hospitalized and ambulatory patients with IBD during flares were recruited between 2009 and 2011. Those with clinical symptoms of venous thromboembolism or previous venous thromboembolism were excluded. We determined the prevalence of DVT among asymptomatic subjects using lower extremity Doppler ultrasound and assessed the performance characteristics of the D-dimer in this high-risk study population.
Results: We enrolled 101 hospitalized and 49 ambulatory patients with IBD during active flares. There were no cases of proximal DVT detected by lower extremity Doppler ultrasound. The 95% confidence interval (CI) for the rate of proximal DVT was 0% to 2%. D-dimer was elevated in 60% of subjects without DVT, occurring more frequently among hospitalized than ambulatory subjects [89% versus 65%, P = 0.01; adjusted odds ratio (aOR), 4.16, 95% CI, 1.58–10.9]. Other predictors of elevated D-dimer were incremental decade in age (aOR, 1.97; 95% CI, 1.24–3.14); ulcerative colitis versus Crohn’s disease diagnosis (aOR, 3.38; 95% CI, 1.29–8.84); and every 10-unit increase in C-reactive protein (aOR, 1.33; 95% CI, 1.09–1.62).
Conclusion: From this pilot study, there appears to be low prevalence of asymptomatic DVTs among patients with IBD during flares. The high prevalence of elevated D-dimer in DVT-negative patients limits its utility in IBD.