Population-Based Study Finds Persistent Gaps in Understanding and Identification of VTE Risk

An estimated 500,000 venous thromboembolism (VTE) events occur annually in the United States, about half of which are related to current or recent hospitalizations, according to results from a 6-year, population-based cohort study published in Blood. These rates remained relatively constant throughout the study period (2005-2010), even with the near-universal adoption of in-hospital VTE prophylaxis, as reported by the authors, led by John A. Heit, MD, from the Mayo Clinic in Rochester, Minnesota.

“VTE is a common and potentially fatal disease. It recurs frequently, [with subsequent pulmonary embolism (PE)] associated with an 11 percent 2-week case fatality rate,” Dr. Heit told ASH Clinical News. Despite the seriousness of this health problem, data on the total number of VTE events – and the number of potentially preventable events – are sparse and conflicting.

In their study, the researchers estimated age- and sex-specific VTE attack rates (defined as incident plus recurrent VTE and consisting of deep vein thrombosis [DVT] or PE [with or without DVT]) among patients living in Olmsted County, Minnesota, who were admitted to a Mayo Clinic hospital for ≥24 hours between 2005 and 2010. They also determined whether events were related (occurring during or within 92 days of hospital discharge) to current or recent hospitalization.

Patients were considered to be “receiving indicated VTE prophylaxis” if they had an electronic dispensed indication for sequential compression devices (SCDs), warfarin, unfractionated heparin (UFH), and/or low-molecular-weight heparin (LMWH), or if they had an electronic order stating that VTE prophylaxis was not needed. Patients who were hospitalized for chemical dependency or psychiatric evaluation or treatment were excluded.

During the study period, 855 adult patients developed a first lifetime VTE; the median age at VTE onset was 62 years (range = 2-100 years). Most patients (51.4%) had a DVT, 28.2 percent had a PE, and 20.5 percent had both PE and DVT. Over the same time period, 345 recurrent VTE events occurred among 281 patients: 57.4 percent were DVTs, 26.4 percent were PEs, and 16.2 percent were both DVTs and PEs.

“Hospitalization-related attack rates were [more than] 35-fold higher and in-hospital rates were [more than] 180-fold higher” than the VTE attack rates among community-dwelling residents with no recent hospitalizations, the authors reported. The VTE attack rates were as follows:

  • events related to hospitalization: 282 per 10,00 person-years (95% CI 257-308)
  • in-hospital VTE attack rate: 1,445 per 10,000 bed-years (95% CI 1,211-1,678)
  • events unrelated to hospitalization: 8.1 per 10,000 person-years (95% CI 7.5-8.7)

Overall, though, the in-hospital and community-dwelling VTE attack rates did not change significantly between 2005 and 2010 (p>0.25).

The in-hospital rate of receiving indicated VTE prophylaxis increased during the study period, from approximately 15-40 percent in 2005 to approximately 90 percent in 2010, which was deemed by the researchers as “a near-universally applied standardized VTE prophylaxis regimen.”

However, during the study period of 2005-2010, the annual hospitalization-related VTE attack rate did not change significantly (p=0.57; ranging from 251 to 306 per 10,000 person-years after adjusting for age and sex). The VTE attack rate for inpatients also remained similar after the researchers adjusted for age and sex (p=0.37; ranging from 1,155 to 1,751 per 10,000 bed-years).

According to Dr. Heit, “performance measures implemented by the Joint Commission and Centers for Medicare & Medicaid Services aimed at increasing the VTE prophylaxis rate to near-universal in-hospital prophylaxis have failed to reduce the occurrence of hospitalization-related VTE.”

Hospitalization for surgery or acute medical illness is the only indication for which guidelines recommend VTE prophylaxis, he added.

The researchers noted that people with active cancer, trauma and fracture, or leg paresis, as well as nursing home residents, are high-risk populations for VTE, but these characteristics have low predictive values for individuals. “Better risk-prediction tools … are needed,” they said.

Detailed electronic VTE prophylaxis data were available for the years 2008 to 2010, during which 15,533 adult residents were hospitalized, for a total of 25,617 hospitalizations. During that time period, 267 hospitalizations were complicated by an incident (n=171) or recurrent (n=96) VTE event either during hospitalization (n=68; 25%) or within 92 days of discharge (n=199; 75%).

The median duration of hospitalization was 3 days (range = 2-5 days), and most patients (66%) received pharmacologic prophylaxis with or without SCD (21% received SCD prophylaxis alone). The median duration of in-hospital anticoagulant-based prophylaxis (including UFH, LMWH, or warfarin with or without SCD) was 70 hours (range = 40-122 hours). An inadequate duration of prophylaxis may have contributed to the lack of improvement in VTE attack rates, Dr. Heit explained.

Upon discharge, 7.4 percent of patients received a prescription for warfarin, 0.9 percent for UFH, and 2.3 percent for LMWH. The median time from discharge to VTE event was 19.5 days (range = 7-46 days).

“These data should not be interpreted as showing that VTE prophylaxis (primary or secondary) is ineffective or that achieving near-universal in-hospital VTE prophylaxis should not be a goal,” Dr. Heit concluded. “Rather, we believe that better risk-assessment tools are needed that allow health-care providers to identify individual patients at high risk for VTE, such that longer duration of primary or secondary prophylaxis can be targeted to individuals who would benefit most.”

The study is limited in that many years have lapsed since the end of the study period, and “VTE attack rates may have changed accordingly” because of changes in prophylaxis regimens, the authors noted. They also said that Olmsted County residents have higher median income and education levels, and racial and ethnic minorities are underrepresented in the population, so the results may not be generalizable. As these are population-level data, further specifics about patient characteristics and indications for or against VTE prophylaxis were not available.

The authors report no relevant conflicts of interest.


Heit JA, Crusan DJ, Ashrani A, et al. Effect of near-universal hospitalization-based prophylaxis on annual number of venous thromboembolism events in the US. Blood. 2017 May 8. [Epub ahead of print]