Published: Apr 2, 2013
By Charles Bankhead

Full Story:  http://www.medpagetoday.com/Cardiology/VenousThrombosis/38215

 Action Points

Glucocorticoid use significantly increased the risk of venous thromboembolism (VTE) among current, new, and continuing users of the anti-inflammatory agents, data from a large case-control population showed.

Steroid users had two to three times the VTE rate of nonusers and former users. Cumulative doses ≥1,000 mg increased VTE risk by 60% to 98% as compared with a prednisone-equivalent cumulative dose ≤10 mg.

The VTE risk also applied to users of inhaled and intestinal glucocorticoids, Sigrun A. Johannesdottir, BSc, of Aarhus University Hospital in Denmark, and co-authors reported online in JAMA Internal Medicine.

However, “systemic glucocorticoids were associated with the greatest risk,” the authors said of their findings. “Patients initiating treatment with systemic glucocorticoids within 90 days before the index date had a threefold increased risk, corresponding to 11 extra VTE cases per 1,000 new users of systemic glucocorticoids annually.”

“Although residual confounding might partially explain the results, clinicians should be aware of this association,” they added.

Among the most commonly used drugs, glucocorticoids have been shown to increase levels of clotting factors and fibrinogen in experimental investigations. Other evidence of a possible link to VTE has come from the observation that patients with Cushing syndrome have an increased risk of VTE, possibly resulting from high levels of endogenous glucocorticoids.

Little clinical information exists regarding any associations between exogenous glucocorticoids and VTE, in part because most studies have focused on disease-specific patient populations, the authors noted in their introduction.

A single epidemiologic study of a general population showed a threefold increase in the risk of VTE among users of oral steroids compared with nonusers, and the magnitude of risk decreased with increasing duration of use (Arch Intern Med 2007; 167: 935-943).

Glucocorticoid use was not the primary exposure of interest in the epidemiologic study, and therefore the authors did not examine the possible influence of route of administration or cumulative dose. To address these issues, Johannesdottir and colleagues analyzed data from the Danish national health system, comprising the nation’s entire population.

Investigators identified all patients who had first-time inpatient or outpatient diagnosis (primary or secondary) of pulmonary embolism (PE) or deep-vein thrombosis (DVT) from 2005 through 2011. They categorized the VTE as provoked (risk factors, such as surgery, major trauma, or pregnancy) within 3 months of the VTE or unprovoked.

Glucocorticoid use was documented by a review of the national prescription database for the study period. Prescriptions for systemic, inhaled, and intestinal glucocorticoids were included in the study.

To the extent possible, investigators identified and accounted for potential confounding factors, including comorbid conditions, frailty and immobility, prescription and nonprescription drugs, and inflammatory conditions that may predispose to VTE and that are treated with glucocorticoids.

The analysis consisted of 38,765 cases of VTE and a control group of 387,650 age- and sex-matched individuals. Patients who had glucocorticoid prescriptions filled within 90 days of VTE were classified as current users. Prescriptions filled 91 to 365 days prior to the event defined recent users, and prescriptions filled >365 days from the event identified former users.

Current users were further categorized as new (first-ever prescription within 90 days of VTE) or continuing.

The results showed that current users had a VTE risk more than twice that of the control group (incidence rate ratio [IRR] 2.31). New users had a threefold increased risk (IRR 3.06), and continuing users had double the risk (IRR 2.02). Recent users had an 18% higher risk, which remained significant compared with nonusers (IRR 1.18, 95% CI 1.10 to 1.26).

The IRR declined to 0.94 among former users.

Oral glucocorticoids increased the VTE risk more than injected steroids did, the authors noted.

Analysis of the relationship between cumulative glucocorticoid dose and VTE risk produced an adjusted IRR of 1.00 for a prednisolone-equivalent dose ≤10 mg, 1.98 for a cumulative dose of 1,000 to 2,000 mg, and 1.60 for >2,000 mg.

New use of inhaled steroids doubled the risk of VTE (IRR 2.21) as did new use of intestinal-directed agents (IRR 2.17).

Analysis of patients with unprovoked VTE yielded results similar to those of the overall analysis. Subgroup analysis of provoked events showed that all comorbidities except asthma and cancer increased VTE risk compared with patients without the conditions.

The authors acknowledged several limitations of the study: lack of information about adherence, use of a relatively new database that might have introduced bias, and lack of information about glucocorticoid use during hospitalization and outpatient visits (such as intra-articular injection). The authors also had limited information about lifestyle factors.

“It is possible that our results present some overestimation of inflammatory conditions that might lead to VTE because of confounding by disease severity,” they noted. “We tried to mitigate this effect by multivariable analyses.”

“On the basis of our rule-out sensitivity analysis, we can reasonably conclude that there was no confounder capable of explaining away the findings, because disease severity or chronic severe inflammation would not have such strong independent effects on VTE,” they added.

The strong evidence of glucocorticoids’ association with VTE provides more reason for caution but should not rule out use of the drugs when indicated, journal editor Mitchell H. Katz, MD, wrote in a commentary.

“Given the already known serious adverse effects of glucocorticoids, establishing an elevated risk for venous thromboembolism with this study does not change the indications for glucocorticoids, but it should remind us to always make sure that the potential benefits of treatment outweigh the risks,” he said.

• Venous Thromboembolism Management

Primary source: JAMA Internal Medicine