Abstract
Introduction: Venous thromboembolism (VTE), defined as deep vein thrombosis (DVT), pulmonary embolism (PE), or both, has been estimated to affect 300,000 to 600,000 individuals in the US each year. Incidence estimates suggest that African Americans have similar, or slightly higher, rates of VTE compared to Whites, while Asians appear to have a lower rate (Beckman, et al., Am J Prev Med, 2010;38:S495-S501). Acquired risk factors for VTE can be identified in ~50% of cases. Using a population-based approach, we used multiple case finding techniques to improve estimates of VTE occurrence in a geographically defined, racially diverse population.
Methods: Durham County, NC, has a population of ~280,000 individuals, characterized as: 52.4% female; 53% White, 38.8% African American, 4.9% Asian, 1% American Indian, and 2.2% ≥2 races; and 13.5% Hispanic (US Census Data, 2012). The County is served by three hospitals, two in the Duke University Health System (DUHS) that share an electronic medical record, and the Durham Veterans’ Administration Medical Center (VAMC). We used a combination of methods to systematically identify patients living in Durham County with a new diagnosis of VTE. These included direct review of imaging data in the Duke Picture Archive and Communication System (PACS); automated review of ICD9 diagnosis codes for VTE and CPT codes for imaging procedures used for diagnosing DVT and PE stored in the Duke Medicine Enterprise Data Warehouse via the Duke Enterprise Data Unified Content Explorer (DEDUCE); direct review of Duke autopsy reports; and review of Veterans Health Information Systems and Technology Architecture (VISTA) to identify cases at the VAMC. Individual patients and events were cross-referenced to avoid duplicate entries, and demographic, risk factor, and treatment data were collected by record review and entered into a REDCap database. Data for the 9 month period from March through November 2013 collected at the DUHS hospitals are included in this abstract.
Results: During the study period, we identified a total of 273 unique individuals with VTE, resulting in an estimated annual incidence of 1.3 per 1,000 individuals in Durham County. The mean age of the patients was 61.6 years (range, 19 to 99 years), and 153 (56%) were female. Mean BMI was 29.5±8.4. Racial distribution of VTE and estimated annual incidence is shown in the table below. Table 1RacePatients, n (%)Estimated annual incidence per 1,000 populationAsian2 (0.73%)0.19African American164 (60%)2.01White104 (38%)0.93Other3 (1.1%)0.45
Characteristics of some of the events and risk factors comparing African American and White patients are shown in the table below. Abstract 4256. Table 2Total (total n=273)African American (total n=164)White (total n=104)Pulmonary emboli141/271 (52%)91/163 (55.8%)48/103 (46.6%)History of prior VTE63/273 (23%)38/162 (23.2%)25/103 (24%)Surgery within the preceding 90 days103/238 (43%)55/151 (36.4%)34/92 (38.0%)Active cancer57/246 (23%)38/153 (24.8%)18/89 (20.2%)Catheter-related17/250 (7%)11/148 (7.4%)6/88 (6.8%)
None of these comparisons were statistically significant. Three African American patients had sickle cell disease. Hypertension (p=0.0392) and end-stage renal disease (p=0.0038) occurred more frequently in African American VTE patients compared to White VTE patients.
Most patients were treated with anticoagulant therapy at the time of diagnosis (n=236; 87%), including low-molecular weight heparin (n=167; 70.8%), unfractionated heparin (n=78; 33.1%), and rivaroxaban (n=21; 8.9%); some patients received more than 1 agent. 129 patients (54.7%) were started on warfarin at the time of diagnosis. A minority of patients were treated with thrombolytic therapy (n=7; 2.6%), thrombectomy/embolectomy (n=3; 1.1%), or placement of an IVC filter (n=19; 7%). Major bleeding occurred in 6 patients (2.2%).
These estimates are subject to limitations. They do not include Durham County residents who were diagnosed and treated completely outside the county, and cases from the Durham VAMC are pending.
Conclusions: Our VTE surveillance case finding approach resulted in an estimated annual VTE incidence of ~1.3 persons per 1,000 per year, with an incidence in African Americans that is approximately twice the incidence in Whites. Common risk factors were similar for African Americans and Whites, but hypertension and end-stage renal disease were more common comorbid conditions in African Americans.
Disclosures
No relevant conflicts of interest to declare.
Cancer-associated thrombosis: the when, how and why