Pharmacologic Prophylaxis for Venous Thromboembolism Among Hospitalized Patients with Acute Medical Illness: An Electronic Medical Records Study

Patients hospitalized with acute medical illness have an elevated risk of venous thromboembolism (VTE). ACCP guidelines list various risk factors including active cancer, congestive heart failure (CHF), respiratory failure, chronic obstructive pulmonary disease (COPD), sepsis, advanced age, history of VTE and immobility. These guidelines also recommend prophylactic anticoagulation using fondaparinux, low molecular-weight heparin, or low-dose unfractionated heparin. This study examined pharmacologic prophylaxis against VTE among hospitalized medically ill patients and assessed demographic and clinical correlates related to VTE prophylaxis.

A retrospective observational study was conducted with electronic medical records (EMR) in a large Midwestern U.S. public hospital and its network of outpatient clinics between 1999 and 2010. Patients >=40 years old were included if they were hospitalized with a primary diagnosis (ICD-9 code) of CHF, pneumonia, cancer, stroke, infectious diseases, sepsis, severe respiratory disorders (including COPD, asthma, and acute respiratory failure), inflammatory bowel disease (IBD), or neurologic disorders, or were hospitalized for VTE within the year before the index admission. Each patient's first qualifying hospitalization during the 12-year period was included in the analysis. Patients were excluded if the length of stay was <3 days, the primary diagnosis was VTE, if a hospital diagnosis of VTE was given in the past 30 days, or if IV heparin or a treatment dose of subcutaneous (SC) heparin (>15,000 units/day) or enoxaparin (>40 mg/day) was started within one day of admission. Patients treated with warfarin (and not heparin or enoxaparin) were also excluded. Fondaparinux was not on the formulary. Prophylaxis was defined as present if the first dose (during the hospital stay) of anticoagulation with SC heparin or enoxaparin was lower than the treatment dose levels defined above. All IV heparin was considered to be at treatment-level doses. SC heparin records with missing doses were considered to be at prophylaxis-level doses. Logistic regression was used to examine factors associated with VTE prophylaxis.

Of 13,029 patients eligible for the study, mean age was 58.6 years; 51% were women; 47% were white, 47% African-American, 6% other; 58% had public healthcare coverage, 15% commercial insurance, 27% self-pay. Among patients with a BMI stored in the EMR, 44% were obese, 23% overweight, and 28% normal. There was a history of cigarette smoking in 44% and atrial fibrillation in 8%. The primary hospital diagnoses were infectious diseases 21.2%, cancer 19.9%, severe respiratory disorders 14.5%, pneumonia 12.7%, CHF 12.2%, stroke 9.6%, sepsis 8.9%, IBD <1%, neurologic disorders <1%, and other <1%. Mean length of stay was 7.8 days (median 5 days).

The overall prophylaxis rate was 22.1% (14.8% for 1999–2002, 22.0% for 2003–2006, 28.5% for 2007–2010), including 11.8% with enoxaparin and 10.3% with SC heparin. Multivariable logistic regression showed that VTE prophylaxis was significantly associated with year of hospitalization (OR=2.2 for ‘07-’10, OR=1.7 for ‘03-’06 vs. ‘99-’02, P<.001), SC heparin in the 30 days before admission (OR=1.6, p=0.02), aspirin use (OR=1.4, p<.0001), a primary hospital diagnosis of sepsis (OR=1.4, p<.0001), self-pay status (OR=1.2 vs. commercial insurance, p=0.01), and age (for 10 year increments, OR=1.07, p=0.001). The following factors were associated with reduced likelihood of VTE prophylaxis: smoking (OR=0.9, p=.005), alcohol use (OR=0.9, p=.04), warfarin in the past 30 days (OR=0.5, p=0.02), and a primary diagnosis of stroke (OR=0.7, p<0.0001), infectious disease (OR=0.7, p<.0001), or IBD (OR=0.3, p=.003).

Pharmacologic VTE prophylaxis has increased significantly over the past 12 years but is still largely underused in patients hospitalized with acute medical illness. Multiple demographic, behavioral and clinical factors are associated with VTE prophylaxis during hospitalization. Further research is needed to examine if VTE prophylaxis is associated with reduced VTE risk and healthcare utilization.

Rosenman:Bristol-Myers Squibb: Research Funding; Pfizer: Research Funding. Liu:Pfizer: Employment. Phatak:Bristol-Myers Squibb: Employment. Qi:Pfizer: Research Funding; Bristol-Myers Squibb: Research Funding. Teal:Pfizer: Research Funding; Bristol-Myers Squibb: Research Funding. Nisi:Pfizer: Research Funding; Bristol-Myers Squibb: Research Funding. Liu:Pfizer: Employment, Equity Ownership. Ramacciotti:Bristol-Myers Squibb: Employment.