Efficacy and Safety of Long-Acting Reversible Contraception in Women With Cardiovascular Conditions.

In patients with cardiovascular conditions, there are limited data regarding the safety and efficacy of long-acting reversible contraception (LARC). 1 , 2 This lack of data is evident in the Centers for Disease Control and Prevention (CDC) United States Medical Eligibility Criteria for Contraceptive Use (USMEC) clinical recommendations: “Theoretical concern exists about the effect of levonorgestrel (LNG) on lipids,” and, “No direct evidence exists on the safety of IUDs among women with peripartum cardiomyopathy. Limited indirect evidence from noncomparative studies did not demonstrate any cases of arrhythmia or infective endocarditis in women with cardiac disease who used IUDs”. 1 The present study examined the safety and efficacy of LARC use in women with cardiovascular conditions.

We conducted a retrospective chart review of women previously diagnosed with a cardiac condition who had a LARC device (IUD or contraceptive implant) placed at the University of Washington between January 1, 2007 and March 1, 2012. The University of Washington Institutional Review Board approved the study. Data were both abstracted and extracted from the electronic medical records from the University of Washington Medical Center, Harborview Medical Center, and other University of Washington affiliated clinics using Microsoft Amalga Unified Intelligence System. Study data were collected and managed using REDCap electronic data capture tools hosted at the University of Washington. 3 We identified subjects using ICD-9 codes for a wide range of cardiovascular conditions (see Appendix 1 ). We included the following categories of cardiovascular conditions: cardiac arrhythmias, congenital heart disease, heart failure, valvular heart disease, coronary heart disease, pulmonary heart disease, cardiomyopathy, venous and arterial vascular disease, rheumatic heart disease, and hypertensive heart disease. We analyzed patient characteristics, indications, and complications following LARC placement. All analyses were performed in Stata 13.

There were 470 women with cardiovascular conditions who had a LARC device placed during the study period. Baseline patient characteristics, time of LARC device placement, and type of LARC device chosen are summarized in Table 1 . The most common category of cardiovascular diagnosis was cardiac arrhythmia ( Table 2 ). ICD-9 diagnosis codes included in each category of cardiovascular condition as well as the frequencies for each are listed in Appendix 1 (Supplemental Materials) . Many patients were noted to have more than one cardiovascular diagnosis. There were 328 women (69.79%) who had the LARC device placed for contraception. Medical records for 159 women (33.83%) indicated that the LARC device was placed for treatment of dysmenorrhea or abnormal uterine bleeding, including menorrhagia, metrorrhagia, perimenopausal bleeding, and postmenopausal bleeding. Thirty nine women (8.21%) had LARC placement to treat a gynecologic condition not related to bleeding, including carcinoma of the vulva, polycystic ovary syndrome (PCOS), endometriosis, uterine polyps, ovarian neoplasm, endometrial hyperplasia, and dysplasia of the cervix. Finally, two women chose the LARC method for a reason coded as “Other” in the chart. One experienced migraine headaches related to her menstrual cycle and reported a desire for menstrual suppression. The second opted for an LNG-IUS because she believed the progestin hormone replacement therapy she was previously using was causing urinary incontinence. Complications following LARC insertion are summarized in Table 3 . Four women (0.85%) developed pelvic inflammatory disease (PID) or endometritis after IUD insertion. Notably, two of these had the IUD inserted immediately following an abortion procedure: one underwent suction dilation and curettage (D&C) at 14 5/7 weeks gestation and PID was diagnosed 7 days following the procedure, and the other had a suction D&C at 8 weeks gestation, with PID diagnosed 29 days following the procedure. The third case of PID was diagnosed in a patient who experienced multiple IUD expulsions. One patient (0.21%) developed endometritis after IUD insertion in the setting of a complicated gynecologic history of multiple fibroids and polyps as well as long-standing menorrhagia. This patient's IUD became malpositioned and embedded in the myometrium. There was one case of confirmed endocarditis that was diagnosed 51 days following IUD insertion. According to the patient chart, the infectious disease department found that this patient's endocarditis was likely caused by drug-resistant S. viridans on a vegetation in the left distal pulmonary artery. The patient chart states that the endocarditis was not likely gynecological in origin. Two women (0.43%) had a confirmed pregnancy with IUD in situ. In the first patient, the IUD was visualized on ultrasound in the lower uterine segment. The IUD was subsequently removed. The second patient had a positive home pregnancy test, and the IUD spontaneously expelled 3 days later. No pregnancies were diagnosed in women with the contraceptive implant. Three hundred and ninety seven patients (84.46%) were still using the LARC device at the end of the study period.

Our case series provides evidence that few women with cardiovascular conditions using a LARC method experience side effects and complications, including pregnancy, with rates comparable to the published data in healthy women. 4 - 10 There were only two confirmed pregnancies in our sample of 470 women, which is consistent with the failure rate of <1% in healthy women. 4 The continuation rate was high, and consistent with other known data on these methods. 11 There were several limitations of this study. This is a retrospective case series without a healthy cohort comparison group. The subjects were not randomized to the various methods and were not observed prospectively. Additionally, our patients were selected from one academic medical center, and generalizability may be limited. However, most patients with significant cardiovascular disease typically receive care at a specialized facility. Despite these potential limitations in available data and generalizability, our study represents the largest case series to date on cardiovascular patients and LARC use. Many patients with cardiovascular disease require anticoagulation at various times, which can cause significantly heavier menstrual bleeding. The LNG-IUS decreases menstrual flow by an average of 90% after 6 months of use, which decreases anemia and improves overall health status, and is likely why it was the most commonly chosen LARC method. 12 Ultimately, the superior efficacy of LARC is of critical importance in these patients who need to either avoid pregnancy altogether or maximize their health prior to conception.