Association Between Hospitalization for Acute Medical Illness and VTE Risk: A Lower Efficacy of Thromboprophylaxis in Elderly Patients? Results from the Edith Case-Control Study

Background: Randomized controlled trials demonstrated that low-molecular-weight heparin (LMWH), unfractionated heparin (UFH), or fondaparinux, are efficient for venous thromboembolism (VTE) prevention in acutely ill medical patients. However, asymptomatic VTE, whose clinical relevance is unknown, represented the vast majority of outcome events of these trials.

Objectives: Using a case-control design, we aimed to estimate the association between admission to hospital for acute medical illness, prescription of pharmacological thromboprophylaxis during admission and symptomatic VTE.

Methods: 750 symptomatic non-surgical, non-cancer, and non-pregnancy-related VTE cases and their 750 age and sex-matched controls were analyzed. Patients hospitalized for an acute medical illness within three months prior to inclusion in the study were identified, and their hospital charts were reviewed. We considered that patients receiving pharmacological thromboprophylaxis were those prescribed either 1) LMWH at an approved preventive dose; 2) low dose UFH (5000 units bid or tid); or 3) fondaparinux 2.5mg once daily, for at least 6 days.

Results: Patients' mean age was 67.2 ± 17.1 years, and 664 (44.3%) were men. A total of 145 cases (19.3%) and 91 controls (12.1%) were hospitalized for acute medical illness at least once in the preceding 3 months prior to inclusion in the study (p<0.001). Hospitalization for acute medical illness in the preceding three months was associated with a 75% increase in VTE risk: OR 1.75 (95% CI: 1.32-2.33). Pharmacological thromboprophylaxis was prescribed in 46.2% of the hospitalized cases (n=67) and in 51.6% of the controls (n=47) (p=0.49). As compared with patients who were not hospitalized, there was a significant 2-fold increased risk of VTE when thromboprophylaxis was not prescribed during hospitalization: OR 2.01 (95% CI: 1.22-3.29), and a non-significant 61% increased risk when thromboprophylaxis was prescribed during hospitalization: OR 1.61 (95% CI: 0.55-1.69). Proportions of admitted patients and prescription of pharmacological thromboprophylaxis are shown in table.

In patients <75 years, VTE risk associated with hospitalization for acute medical illness was low and not significant (OR 1.42 (95% CI: 0.96-2.08) as compared with no hospitalization. When pharmacological thromboprophylaxis was not prescribed during admission, there was a significant 2-fold increase in VTE risk associated with hospitalization: OR 2.01 (95% CI: 1.11-3.62), whereas no association was found when thromboprophylaxis was prescribed: OR 0.93 (95% CI: 0.44-1.95). In patients ≥75 years, we observed an overall 2.25-fold increase in VTE risk associated with hospitalization for acute medical illness as compared with no hospitalization: OR 2.25 (95% CI: 1.46-3.47). VTE risk associated with hospitalization remained significant whether or not pharmacological thromboprophylaxis was prescribed during hospitalization: OR 2.69 (95% CI 1.28-5.66) and OR 2.02 (95% CI: 1.01-4.03) respectively. Adjustment for a priori VTE risk stratification of medical hospitalized patients did not modify estimates.

Conclusion: Our case-control study suggests that VTE prevention in acutely ill medical patients may be less effective in patients ≥75 years as compared with younger patients. This finding will be addressed in a randomized double blind controlled trial comparing enoxaparin 40mg once daily to placebo for the prevention of symptomatic VTE and fatal PE in patients ≥75 years admitted to hospital for an acute medical illness (NCT02379806).