Kathrine Agergård Kaspersen

Combined oral contraceptives and increased risk of venous thromboembolism and arterial thrombosis: results from the Danish Blood Donor Study (DBDS)

Publikation: KonferencebidragKonferenceabstrakt til konferenceForskningpeer review

  • Khoa Manh Dinh
  • Kathrine Agergård Kaspersen
  • Susan Mikkelsen
  • Ole Birger Pedersen, Department of Clinical Immunology, Næstved Sygehus, Næstved, Denmark., Danmark
  • Erik Sørensen, Department of Clinical Immunology, Copenhagen University Hospital., Danmark
  • Lise Wegner Thørner, Department of Clinical Immunology, Copenhagen University Hospital., Danmark
  • Henrik Hjalgrim, Epidemiology Research, Statens Serum Institut, Copenhagen, Denmark., Department of Hematology, Rigshospitalet, Copenhagen
  • ,
  • Henrik Ullum, Københavns Universitet, Department of Clinical Immunology, Copenhagen University Hospital., Danmark
  • Christian Erikstrup
BACKGROUND:
Combined oral contraceptives (OC) are the most commonly used method of contraception among women aged 15-49 years in the Nordic countries. OC offer convenient, effective, and reversible fertility regulation.
Epidemiological studies in 2011 demonstrated that OC are associated with increased risk of venous thromboembolism (VTE), thrombotic stroke, and myocardial infarction. This resulted in revised OC prescription practices, demanding renewed follow-up studies of the risk of thrombotic complications linked to OC. Indeed, unpublished data from our group reveal that OC users have a higher blood platelet level (PLT) but the association between PLT and thrombotic events remains unknown.

AIMS:
The aim of this study was, first, to examine the association between PLT and venous or arterial thrombotic events and, second, to investigate the risk of VTE, thrombotic stroke, and myocardial infarction associated with use of OC among healthy female blood donors.

METHODS:
We included 28,950 women aged 18-49 years from The Danish Blood Donor Study. Participants completed a standard questionnaire on lifestyle factors and stated their use of contraception. All participants were followed in national registers between March 2010 and December 2015: OC exposure period and type of OC were identified in the Danish National Prescription Register and venous or arterial thrombosis endpoints were identified by ICD-10 diagnosis codes in the Danish National Patient Register. PLT from each donation were measured by automated haematology analysers. The association between OC and thrombosis endpoints was explored by multivariable Cox proportional hazards regression analysis with age as underlying timescale. Adjustment for age, body mass index, smoking, education level, and duration of OC use was performed. Groups were compared by unpaired t-test or analysis of variance. Results are presented as hazard ratios (HR) with 95% confidence intervals (CI) and p-values.

RESULTS:
Within 209,164 person-years of observation, 135 first ever events occurred. 118 (87%) were VTE and 17 (13%) were arterial thrombotic events. Mean PLT was 262.4 (CI: 248.3 – 276.6) and 254.7 (254.0 – 255.4) in the group with thrombotic event and without event, respectively. No statistically difference in mean PLT was observed between OC users having an event compared with OC users without event (p = 0.92).
The crude HR of overall thrombotic event in users of OC was 3.14 (CI: 2.10 – 4.78) compared with non-users. The adjusted HR was 3.70 (2.23 – 6.13). The adjusted HR of VTE for OC users of 2nd generation, 3rd generation, and 4th generation was 2.51 (1.21 – 5.21), 4.76 (2.68 – 8.45), and 4.41 (1.93 – 10.0), respectively, all compared with non-users of OC. Similarly, the risk of arterial thrombosis was increased among OC users compared with non-users, however not statistically significant.

CONCLUSION:
No difference in blood platelet levels in participants with thrombotic event compared with controls was observed. OC users had higher risk of VTE compared with non-users and the risk was highest among users of 3rd and 4th generation OC. Our findings underscores that blood donor cohorts are feasible for the study of general health risks in otherwise healthy persons.
OriginalsprogEngelsk
Udgivelsesår6 sep. 2018
StatusUdgivet - 6 sep. 2018
Begivenhed3rd European Conference on Donor Health and Management - Copenhagen, Danmark
Varighed: 5 sep. 20187 sep. 2018
http://www.ecdhm.org

Konference

Konference3rd European Conference on Donor Health and Management
LandDanmark
ByCopenhagen
Periode05/09/201807/09/2018
Internetadresse

Se relationer på Aarhus Universitet Citationsformater

ID: 148060605