Deep venous thrombosis in medical-surgical critically ill patients: Prevalence, incidence, and risk factors Deborah Cook, MD; Mark Crowther, MD; Maureen Meade, MD; Christian Rabbat; Lauren Griffith, MSc; David Schiff, MD; William Geerts, MD; Gordon Guyatt, MD C ritically ill patients have an increased risk of lower ex- tremity deep venous throm- bosis (DVT) compared with other hospitalized patients (1). The clin- ical consequences of DVT have the poten- tial to be particularly serious yet paradox- ically are unrecognized in the intensive care unit (ICU). Concern about undiag- nosed DVT in the medical-surgical ICU setting is underscored by studies showing that 10% (2) to 100% (3, 4) of DVTs identified by screening ultrasound were clinically unsuspected, and it is possible that many mechanically ventilated pa- tients with sudden episodes of hypoten- sion, tachycardia, or hypoxia may have undetected pulmonary embolism (PE) (5). Unsuspected PE may also contribute to difficulty weaning patients from me- chanical ventilation (3). Finally, venous thromboembolism (VTE) remains one of the most common unsuspected autopsy findings in critically ill patients (6). The significance of VTE in critically ill pa- tients is enhanced by the observation that these patients, who have impaired cardio- pulmonary reserve, are likely to have se- rious consequences of even small pulmo- nary emboli (7). We hypothesized that DVT rates would be lower in a heterogeneous group of medical-surgical ICU patients compared with studies conducted in the last decade (2– 4), due to increased attention to thromboprophylaxis. The objectives of this study were to determine the preva- lence and incidence of proximal lower extremity DVT detected by screening compression ultrasonography among medical-surgical critically ill patients and to determine the baseline and time- dependent risk factors for DVT in this population. Objective: Critically ill patients may be at high risk of venous thromboembolism. The objective was to determine the prevalence, incidence, and risk factors for proximal lower extremity deep venous thrombosis among critically ill medical-surgical patients. Design: Prospective cohort. Setting: Closed university-affiliated intensive care unit. Patients: We enrolled consecutive patients >18 yrs of age expected to be in intensive care unit for >72 hrs. Exclusion criteria were an admitting diagnosis of trauma, orthopedic sur- gery, pregnancy, and life support withdrawal. Interventions: Interventions included bilateral lower extremity compression ultrasound within 48 hrs of intensive care unit admis- sion, twice weekly, and if venous thromboembolism was clinically suspected. Thromboprophylaxis was protocol directed and universal. We recorded deep venous thrombosis risk factors at baseline and daily, using multivariate regression analysis to determine indepen- dent predictors. Patients were followed to hospital discharge. Results: Among 261 patients with a mean Acute Physiology and Chronic Health Evaluation II score of 25.5 (8.4), the prevalence of deep venous thrombosis was 2.7% (95% confidence interval 1.1–5.5) on intensive care unit admission, and the incidence was 9.6% (95% confidence interval 6.3–13.8) over the intensive care unit stay. We identified four independent risk factors for intensive care unit-ac- quired deep venous thrombosis: personal or family history of venous thromboembolism (hazard ratio 4.0, 95% confidence interval 1.5– 10.3), end-stage renal failure (hazard ratio 3.7, 95% confidence interval 1.2–11.1), platelet transfusion (hazard ratio 3.2, 95% confi- dence interval 1.2– 8.4), and vasopressor use (hazard ratio 2.8, 95% confidence interval 1.1–7.2). Patients with deep venous thrombosis had a longer duration of mechanical ventilation (p  .03), intensive care unit stay (p  .005), and hospitalization (p < .001) than patients without deep venous thrombosis. Conclusions: Despite universal thromboprophylaxis, medical- surgical critically ill patients remain at risk for lower extremity deep venous thrombosis. Further research is needed to evaluate the risks and benefits of more intense venous thromboembolism prophylaxis. (Crit Care Med 2005; 33:1565–1571) KEY WORDS: critical care; deep venous thrombosis; prevalence; incidence; risk factors From the Departments of Medicine (DC, MC, MM, CR, GG), Clinical Epidemiology & Biostatistics (DC, MM, LG, GG), and Radiology (DS), McMaster University, Hamilton, Ontario, Canada; and the Departments of Med- icine, Health Policy, Measurement and Evaluation (WG), University of Toronto, Toronto, Ontario, Canada. Supported, in part, by grant MOP-49571 from the Canadian Institutes for Health Research. Dr. Cook is a chair of the Canadian Institutes for Health Research. Drs. Crowther and Meade hold research scholarships from the Canadian Institutes for Health Research. Drs. Cook, Meade, Rabbat, and Guyatt have re- ceived grants from Pfizer for studies other than the one published here, which were deposited into a univeristy research account. Dr. Cook has consulted for and received honoraria from Pfizer, which were deposited into a university research account. Dr. Crowther has received grants, honoraria, or consulting fees from Pfizer, Sanofi-Aventis, Leo Laboratories, Astra Zeneca, Bayer Healthcare, Novo Nordisc, and Calea. Dr. Geerts has consulted for and received honoraria or research funding from Aventis, Bayer, GlaxoSmithKline, Lilly, Pfizer, and Sanofi-Synthelabo. The remaining authors, Drs. Griffith and Schiff, have no financial interests to disclose. The authors assert that none of the disclo- sures listed here conflict with the study presented, which was funded by the Canadian Institutes for Health Research. Copyright © 2005 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/01.CCM.0000171207.95319.B2 1565 Crit Care Med 2005 Vol. 33, No. 7