Abstract 3367


Menorrhagia, or excessive menstrual bleeding (EMB), is defined as menstrual blood loss of greater than or equal to 80 ml per menstrual cycle1,2. EMB is an important public health problem affecting as many as 10–15% of pre-menopausal women3 and is a common presenting symptom in women referred to a gynecologist4,5. Although there are many etiologies of EMB, inherited bleeding disorders may be identified in as many as 10–20% of women with EMB6,7. The levonorgestrel-releasing intrauterine system (LNG-IUS) is an intrauterine device that secretes 20 mcg of levonorgestrel per day and is indicated for contraception and for the treatment of heavy menstrual bleeding8. There have been limited data for the use of the LNG-IUS among women with inherited bleeding disorders and expulsion and malposition rates are not well known. Expulsion occurs in approximately 5–10% of women without an inherited bleeding disorder9,10. The purpose of this study is to examine the proportion of levonorgestrel-releasing intrauterine systems resulting in expulsion or malposition among women with inherited bleeding disorders.


Women with an inherited bleeding disorder were retrospectively identified from the records of the Women and Bleeding Disorders clinic at a tertiary care hospital in Kingston, Canada. Women were included if they were seen for EMB and treated with an LNG-IUS between May 2005 and June 2012. All LNG-IUS insertions during this time were performed by one of 2 gynecologists experienced in LNG-IUS insertion. The primary outcome was a combined endpoint of expulsion and/or malposition. Expulsion or malposition was suspected if LNG-IUS strings were not apparent on direct visualization, if the LNG-IUS was visible at the cervix os, or if the ongoing patient-related symptoms or concerns suggested possible malposition. Suspected expulsion and malposition were confirmed by endovaginal ultrasound.


108 women were seen in the Kingston Women and Bleeding Disorders clinic between May 2005 and June 2012. During this time, there were 23 LNG-IUS inserted in 20 women for the management of EMB. Follow-up was complete for 19/23 LNG-IUS insertions. The median age was 31 years (range 18 to 43). The most common diagnosis in the study population was type 1 Von Willebrand disease (VWD) (13/20, 65%). While there was no pre-medication used in the large majority of cases, 2 patients received treatment at the time of insertion: 1 with tranexamic acid (type 1 VWD) and 1 with recombinant factor VIII (hemophilia A carrier). During this study period, there were 5 episodes of LNG-IUS expulsion or device malposition resulting in removal [5/19 (26.3%), 95% CI 11.8% to 48.8%]. An additional 5 women had their device removed for other reasons: 1 due to pain, and 4 due to failure of the LNG-IUS to satisfactorily reduce menstrual bleeding. The proportion of devices requiring removal in this population was 10/19 (52.6%, 95% CI 31.7% to 71.7%). The median time to device removal was 218 days (range 30 to 636). Among women whose LNG-IUS remained in proper position there was a significant increase in hemoglobin concentration associated with LNG-IUS insertion [122 g/L (SD 11.9) vs.129.5 g/L (SD 8.5), p=0.04, paired sample t-test].


In this retrospective study, over half (52.6%) of women with an inherited bleeding disorder had an LGN-IUS removed due to poor patient satisfaction, malposition, or expulsion. In this group of women with bleeding disorders (and no premedication at the time of the insertion), the proportion of LNG-IUS resulting in expulsion or malposition was 26.3%, which is higher than that reported among women without a bleeding disorder. Women with inherited bleeding disorders may be at increased risk of expulsion due to heavy menstrual blood flow and/or uterine contractions. Further study into the causes of higher complication rates and interventions such as premedication or prolonged treatment with anti-fibrinolytic agents targeted at improving outcomes in this population is required.


James:CSL-Behring, Baxter, Bayer: Honoraria, Research Funding.

Author notes


Asterisk with author names denotes non-ASH members.