Non-steroidal anti-inflammatory drug use and risk of venous thromboembolism

M. Schmidt, C. F. Christiansen, E. Puhó, R. J. Glynn, K. J. Rothman, H. T. Sørensen

Research output: Contribution to journalArticle

31 Citations (Scopus)

Abstract

Background:The association between the use of non-selective non-steroidal anti-inflammatory drugs (NSAIDs) or cyclooxygenase-2-selective inhibitors (COX2Is) and the risk of venous thromboembolism (VTE) remains unclear. Objectives:To examine this association. Patients/Methods:We conducted a population-based case-control study in northern Denmark (population of 1.7million). Using the National Patient Registry, we identified patients with a first hospital VTE diagnosis during 1999-2006 (n=8368) and their comorbidities. For each case, we selected 10 controls (n=82218) matched by age and sex. From the prescription database, we ascertained the use of NSAIDs at the time of diagnosis (current use) or before (recent use), and comedications. Current use was further classified as new use (first-ever prescription redemption within 60 days before diagnosis date) or long-term use. We used odds ratios from a logistic regression model to estimate incidence rate ratios (IRRs) with 95% confidence intervals (CIs). Results:As compared with no use, the adjusted IRR associating current non-selective NSAID use with VTE was 2.51 (95% CI 2.29-2.76), and that for current COX2I use was 2.19 (95% CI 1.99-2.41). Recent users had substantially smaller increases than current users. The adjusted IRRs among long-term users were 2.06 for non-selective NSAIDs (95% CI 1.85-2.29) and 1.92 for COX2Is (95% CI 1.72-2.15). Similarly increased risks were found for unprovoked VTE (occurrence in the absence of pregnancy, cancer, major trauma, fracture or surgery within 3 months preceding the VTE), deep vein thrombosis, pulmonary embolism, and individual NSAIDs. Conclusions:The use of non-selective NSAIDs or COX2Is was associated with a two-fold or more increased risk of VTE.

Original languageEnglish
Pages (from-to)1326-1333
Number of pages8
JournalJournal of Thrombosis and Haemostasis
Volume9
Issue number7
DOIs
Publication statusPublished - Jul 2011

Fingerprint

Venous Thromboembolism
Anti-Inflammatory Agents
Cyclooxygenase 2 Inhibitors
Confidence Intervals
Pharmaceutical Preparations
Prescriptions
Incidence
Logistic Models
Denmark
Pulmonary Embolism
Venous Thrombosis
Population
Registries
Case-Control Studies
Comorbidity
Odds Ratio
Databases
Pregnancy
Wounds and Injuries
Neoplasms

Keywords

  • Cardiovascular disease
  • Case-control study
  • Cyclooxygenase-2 inhibitors
  • Epidemiology
  • Non-steroidal anti-inflammatory drugs
  • Venous thromboembolism

ASJC Scopus subject areas

  • Hematology

Cite this

Non-steroidal anti-inflammatory drug use and risk of venous thromboembolism. / Schmidt, M.; Christiansen, C. F.; Puhó, E.; Glynn, R. J.; Rothman, K. J.; Sørensen, H. T.

In: Journal of Thrombosis and Haemostasis, Vol. 9, No. 7, 07.2011, p. 1326-1333.

Research output: Contribution to journalArticle

Schmidt, M. ; Christiansen, C. F. ; Puhó, E. ; Glynn, R. J. ; Rothman, K. J. ; Sørensen, H. T. / Non-steroidal anti-inflammatory drug use and risk of venous thromboembolism. In: Journal of Thrombosis and Haemostasis. 2011 ; Vol. 9, No. 7. pp. 1326-1333.
@article{536ead92061f43f6962657c40fdb5201,
title = "Non-steroidal anti-inflammatory drug use and risk of venous thromboembolism",
abstract = "Background:The association between the use of non-selective non-steroidal anti-inflammatory drugs (NSAIDs) or cyclooxygenase-2-selective inhibitors (COX2Is) and the risk of venous thromboembolism (VTE) remains unclear. Objectives:To examine this association. Patients/Methods:We conducted a population-based case-control study in northern Denmark (population of 1.7million). Using the National Patient Registry, we identified patients with a first hospital VTE diagnosis during 1999-2006 (n=8368) and their comorbidities. For each case, we selected 10 controls (n=82218) matched by age and sex. From the prescription database, we ascertained the use of NSAIDs at the time of diagnosis (current use) or before (recent use), and comedications. Current use was further classified as new use (first-ever prescription redemption within 60 days before diagnosis date) or long-term use. We used odds ratios from a logistic regression model to estimate incidence rate ratios (IRRs) with 95{\%} confidence intervals (CIs). Results:As compared with no use, the adjusted IRR associating current non-selective NSAID use with VTE was 2.51 (95{\%} CI 2.29-2.76), and that for current COX2I use was 2.19 (95{\%} CI 1.99-2.41). Recent users had substantially smaller increases than current users. The adjusted IRRs among long-term users were 2.06 for non-selective NSAIDs (95{\%} CI 1.85-2.29) and 1.92 for COX2Is (95{\%} CI 1.72-2.15). Similarly increased risks were found for unprovoked VTE (occurrence in the absence of pregnancy, cancer, major trauma, fracture or surgery within 3 months preceding the VTE), deep vein thrombosis, pulmonary embolism, and individual NSAIDs. Conclusions:The use of non-selective NSAIDs or COX2Is was associated with a two-fold or more increased risk of VTE.",
keywords = "Cardiovascular disease, Case-control study, Cyclooxygenase-2 inhibitors, Epidemiology, Non-steroidal anti-inflammatory drugs, Venous thromboembolism",
author = "M. Schmidt and Christiansen, {C. F.} and E. Puh{\'o} and Glynn, {R. J.} and Rothman, {K. J.} and S{\o}rensen, {H. T.}",
year = "2011",
month = "7",
doi = "10.1111/j.1538-7836.2011.04354.x",
language = "English",
volume = "9",
pages = "1326--1333",
journal = "Journal of Thrombosis and Haemostasis",
issn = "1538-7933",
publisher = "Wiley-Blackwell",
number = "7",

}

TY - JOUR

T1 - Non-steroidal anti-inflammatory drug use and risk of venous thromboembolism

AU - Schmidt, M.

AU - Christiansen, C. F.

AU - Puhó, E.

AU - Glynn, R. J.

AU - Rothman, K. J.

AU - Sørensen, H. T.

PY - 2011/7

Y1 - 2011/7

N2 - Background:The association between the use of non-selective non-steroidal anti-inflammatory drugs (NSAIDs) or cyclooxygenase-2-selective inhibitors (COX2Is) and the risk of venous thromboembolism (VTE) remains unclear. Objectives:To examine this association. Patients/Methods:We conducted a population-based case-control study in northern Denmark (population of 1.7million). Using the National Patient Registry, we identified patients with a first hospital VTE diagnosis during 1999-2006 (n=8368) and their comorbidities. For each case, we selected 10 controls (n=82218) matched by age and sex. From the prescription database, we ascertained the use of NSAIDs at the time of diagnosis (current use) or before (recent use), and comedications. Current use was further classified as new use (first-ever prescription redemption within 60 days before diagnosis date) or long-term use. We used odds ratios from a logistic regression model to estimate incidence rate ratios (IRRs) with 95% confidence intervals (CIs). Results:As compared with no use, the adjusted IRR associating current non-selective NSAID use with VTE was 2.51 (95% CI 2.29-2.76), and that for current COX2I use was 2.19 (95% CI 1.99-2.41). Recent users had substantially smaller increases than current users. The adjusted IRRs among long-term users were 2.06 for non-selective NSAIDs (95% CI 1.85-2.29) and 1.92 for COX2Is (95% CI 1.72-2.15). Similarly increased risks were found for unprovoked VTE (occurrence in the absence of pregnancy, cancer, major trauma, fracture or surgery within 3 months preceding the VTE), deep vein thrombosis, pulmonary embolism, and individual NSAIDs. Conclusions:The use of non-selective NSAIDs or COX2Is was associated with a two-fold or more increased risk of VTE.

AB - Background:The association between the use of non-selective non-steroidal anti-inflammatory drugs (NSAIDs) or cyclooxygenase-2-selective inhibitors (COX2Is) and the risk of venous thromboembolism (VTE) remains unclear. Objectives:To examine this association. Patients/Methods:We conducted a population-based case-control study in northern Denmark (population of 1.7million). Using the National Patient Registry, we identified patients with a first hospital VTE diagnosis during 1999-2006 (n=8368) and their comorbidities. For each case, we selected 10 controls (n=82218) matched by age and sex. From the prescription database, we ascertained the use of NSAIDs at the time of diagnosis (current use) or before (recent use), and comedications. Current use was further classified as new use (first-ever prescription redemption within 60 days before diagnosis date) or long-term use. We used odds ratios from a logistic regression model to estimate incidence rate ratios (IRRs) with 95% confidence intervals (CIs). Results:As compared with no use, the adjusted IRR associating current non-selective NSAID use with VTE was 2.51 (95% CI 2.29-2.76), and that for current COX2I use was 2.19 (95% CI 1.99-2.41). Recent users had substantially smaller increases than current users. The adjusted IRRs among long-term users were 2.06 for non-selective NSAIDs (95% CI 1.85-2.29) and 1.92 for COX2Is (95% CI 1.72-2.15). Similarly increased risks were found for unprovoked VTE (occurrence in the absence of pregnancy, cancer, major trauma, fracture or surgery within 3 months preceding the VTE), deep vein thrombosis, pulmonary embolism, and individual NSAIDs. Conclusions:The use of non-selective NSAIDs or COX2Is was associated with a two-fold or more increased risk of VTE.

KW - Cardiovascular disease

KW - Case-control study

KW - Cyclooxygenase-2 inhibitors

KW - Epidemiology

KW - Non-steroidal anti-inflammatory drugs

KW - Venous thromboembolism

UR - http://www.scopus.com/inward/record.url?scp=79959817996&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79959817996&partnerID=8YFLogxK

U2 - 10.1111/j.1538-7836.2011.04354.x

DO - 10.1111/j.1538-7836.2011.04354.x

M3 - Article

C2 - 21592304

AN - SCOPUS:79959817996

VL - 9

SP - 1326

EP - 1333

JO - Journal of Thrombosis and Haemostasis

JF - Journal of Thrombosis and Haemostasis

SN - 1538-7933

IS - 7

ER -