Preliminary Validation of Clinical Assessment for Deep Vein Thrombosis in Orthopaedic Outpatients Daniel L. Riddle, PhD*; Marnix R. Hoppener, MSc†; Roderik A. Kraaijenhagen, MD†; Jodi Anderson, MS‡; and Philip S. Wells, MD‡ The purpose of our study was to determine if a previously published clinical decision rule designed to estimate the probability of proximal deep vein thrombosis in outpatients is valid when applied exclusively to outpatients with muscu- loskeletal disorders. We also sought to determine whether probability estimates differed for patients with or without trauma, fracture, or recent orthopaedic surgery. Data col- lected from outpatients with surgical and nonsurgical mus- culoskeletal disorders (n = 464) were extracted from the datasets of three previously published studies done on het- erogeneous groups of patients (n = 3424). Followup for all patients was 3 months. Testing of all patients for thrombo- embolic disease was done using validated diagnostic proce- dures. Probability estimates for orthopaedic outpatients were consistent with estimates from published studies. The proportion of patients who had venous thromboembolism was 5.6% (95% confidence interval, 3.5–8.7%) for the low probability group, 14.1% (95% confidence interval, 8.6– 22.4%) for the moderate probability group, and 47.4% (95% confidence interval, 35.3–60%) for the high probability group. Validity estimates for patients with and without re- cent trauma, surgery, or fracture differed, but not dramati- cally. The validity of the clinical decision rule as applied to outpatients with musculoskeletal disorders was supported. Level of Evidence: Prognostic study, Level II-1 (retrospective study). See the Guidelines for Authors for a complete de- scription of the levels of evidence. Lower-extremity deep vein thrombosis (DVT) is classified as being either proximal (popliteal vein and above) or distal (calf veins). Proximal DVT (PDVT) is the less com- mon but more dangerous form of lower extremity DVT because it is more likely than distal DVT to cause a life threatening pulmonary embolism (PE). 1,11 Orthopaedic surgery and orthopaedic injuries result in substantial risk for PDVT. 7 Most outpatients who had orthopaedic proce- dures have PDVT develop after hospital dis- charge. 11,26,31,35 Proximal deep vein thrombosis associ- ated with inhospital care for orthopaedic patients is well recognized, but PDVT associated with outpatient or posthospital care is less frequently discussed and may be under-recognized. 16 If outpatients with PDVT can be iden- tified early, the risk for severe morbidity and mortality can be decreased. 1,25 Some clinicians attempt to identify patients suspected of having PDVT by considering the patient’s signs, symp- toms, and associated risk factors. 24 However, 75–84% of patients who are suspected of having DVT do not have DVT when formal diagnostic testing is completed. 14,34 In response to the high rate of incorrect clinical diagnoses, numerous researchers have more closely examined the utility of the clinical examination for identifying patients with PDVT. 20,24,26,31,33 More sophisticated methods for combining risk factors and signs and symptoms into a clinical decision rule (CDR), a clinical tool that quantifies the contributions that history and physical examination data make toward diagnosis, have been described. 21 The CDR that generally has been accepted as being the most reliable and valid for diagnosing patients suspected of hav- ing a lower-extremity PDVT 9,10,13,17,19,23,27,28 is the CDR developed by Wells et al (Table 1). 2,4 Numerous investigators have reported on the validity of the CDR proposed by Wells et al, 31,33,34 however, a het- Received: May 5, 2004 Revised: July 26, 2004 Accepted: October 6, 2004 From the *Department of Physical Therapy, Medical College of Virginia Campus, Richmond, VA; the †Department of Vascular Medicine, University of Amsterdam, Amsterdam, the Netherlands; and the ‡Department of Medi- cine and Clinical Epidemiology Unit, University of Ottawa and the Ottawa Health Research Institute, Ottawa, Ontario, Canada. Each author certifies that he or she has no commercial associations (eg, consultancies, stock ownership, equity interest, patent/licensing arrange- ments, etc) that might pose a conflict of interest in connection with the submitted article. Each author certifies that his or her institution has approved the human protocol for this investigation and that all investigations were conducted in conformity with ethical principles of research. This study was funded, in part, by the Agency for Healthcare Research and Quality (#RO3 HS13059-01). Correspondence to: Daniel L. Riddle, PhD, PT, Department of Physical Therapy, P.O. Box 980224, Virginia Commonwealth University, Richmond, VA 23298-0224. Phone: 804-828-0234; Fax: 804-828-8111; E-mail: dlriddle@vcu.edu. DOI: 10.1097/01.blo.0000150347.36843.c4 CLINICAL ORTHOPAEDICS AND RELATED RESEARCH Number 432, pp. 252–257 © 2005 Lippincott Williams & Wilkins 252