Intrauterine devices are long-acting reversible methods for contraception that provide high rates of contraceptive efficacy and rates of satisfaction compared to other contraceptive methods [1,2,3,4]. Copper-containing intrauterine devices (Cu-IUDs) such as the Paragard T380A, the only copper-containing and 100% hormone free IUD approved in the U.S., have shown excellent contraceptive effectiveness, despite the lack of hormones, and few contraindications [3, 5,6,7,8]. A 2006 systematic review of over 50,000 women compared various Cu-IUDs and found that those with 380 mm2 surface area of exposed copper were the most effective for contraception [9]. Cu-IUDs have been recommended as the first line contraceptive methods offered in women with contraindications to estrogen and/or progesterone and for patients with certain cancers, such as breast and cervical cancer [10]. Cu-IUD is also the most effective option for emergency contraception with a 92–99% effectiveness [9,10,11].

MRI is a common imaging modality for multiple clinical problems ranging from musculoskeletal injuries, to evaluating for and staging cancers of the bone, breast, abdomen and pelvis, chest, and nervous system. Multiple in-vitro and in-vivo studies have shown that 1.5 T MRI can be safely performed on patients with Cu-IUDs when measuring device heating, deflection, and torque [12,13,14]. In addition, previous in-vitro studies have shown that there are no contraindications or safety concerns for performing 3 T MRI on patients with Cu-IUDs when testing for device deflection, torque, heating and artifact [15, 16]. However, in the only in-vivo study to date of Cu-IUDs at 3 T, one of 18 included women had partial perforation and dislocation of the IUD into the fundal myometrium [17]. Although the authors acknowledged that the perforation may have occurred prior to the MRI examination, they concluded that women with Cu-IUDs should consult their healthcare professionals after an MRI at 3 T to check the correct positioning of the IUD to exclude any complications. Currently, Cu-IUDs are classified as “MR conditional”, meaning patients can be safely scanned in an MRI system when meeting specified conditions and specific absorption rate (SAR) limitations. For Paragard T380A, the only approved Cu-IUD in the United States, these conditions are a static magnetic field of 3 T or 1.5 T, maximum spatial field gradients of 4000 gauss/cm, and maximum MR system reported whole body average SAR of 2 W/kg [1].

Given the increased use of 3 T MRI as well as IUDs, it is important to further investigate the safety profile of Cu-IUDs at 3 T [18, 19]. The goal of this study is to prospectively evaluate whether patients with Paragard T380A, a Cu-IUD, could safely undergo 3 T MRI.

This study was approved by the institutional review board at Naval Medical Center San Diego and was compliant with the Health Insurance Portability and Accountability Act. All patients provided written informed consent. This prospective single-site study enrolled patients who presented for 3 T MRI for any indication and had a Paragard T380A Cu-IUD between 12 January 2018 and 19 December 2019. Eligible participants were women ≥ 18 years old with a confirmed Cu-IUD who underwent 3 T MRI for any reason. Women who did not consent or were under the age of 18 years old were excluded from the study.

On the day of the MRI examination, all consented patients underwent an ultrasound prior to the MRI to determine the position of the IUD (Fig. 1). All pelvic ultrasounds were performed transabdominally using Philips EPIQ 7G (Philips Healthcare, Best, Netherlands) by dedicated study sonographers. Transabdominal ultrasounds were reviewed by two study radiologists to confirm proper placement, as defined as IUD crossbar in the fundal portion of the endometrial cavity using 3D imaging (Fig. 2). A malpositioned IUD was defined as IUD crossbars not in the fundal portion of the endometrial cavity (Fig. 3), a visualized tip in the mid or lower uterus, any part of the device located in the cervical canal or outside of the endometrial canal, a fractured device, or a non-visualized IUD [20]. Per protocol, if the transabdominal ultrasound could not determine the location or position of the IUD, a transvaginal ultrasound was performed. During the study, transvaginal ultrasound was performed on one participant prior to the MRI. This additional step was performed as the IUD was found to be appropriately positioned on the transabdominal ultrasound and the sonographer performed a transvaginal ultrasound as per clinical practice, and not per protocol. If the IUD was in the proper position, then the patient was taken to the MRI suite for their clinical MRI (Fig. 1). If the IUD was malpositioned, the patient received their clinical MRI as scheduled but were removed from the study (Fig. 1). In addition, these patients were counseled to see their provider for repositioning of the IUD and to use another form of contraception in the meantime.

Study schema. Intrauterine device (IUD). Copper-containing IUD (Cu-IUD). Magnetic Resonance Imaging (MRI)

Proper IUD placement. 2D grayscale transabdominal ultrasound images of the uterus in long (A) and short axis (B) demonstrating an IUD stem (arrow) centered within the endometrial cavity and both arms (arrowheads) of the device extending laterally at the uterine fundus. 3D grayscale (C) and gynecologic tissue rendered (D) ultrasound images of the same patient confirming appropriate positioning of the stem and arms of the IUD

Malpositioned IUD. 2D grayscale transvaginal ultrasound images of the uterus in long (A) and short axis (B) demonstrating an IUD stem (arrow) centered within the endometrial cavity but arms not at the uterine fundus. 3D grayscale (C) and gynecologic tissue rendered (D) ultrasound images of the same patient demonstrating a flipped device with the IUD arms (arrowheads) extending outward within the lower uterus near the internal cervical os

All clinical MRI examinations were performed on Philips 3.0 T Ingenia (Philips Healthcare, Best, Netherlands). All MRI examinations used a “low SAR” setting to comply with the conditions for imaging patients with a Cu-IUD at 3 T [21]. The type of MRI by body part was recorded for each patient.

After the MRI examination and while in the MRI suite, all patients completed a questionnaire asking about any symptoms they had during their MRI examination (Appendix 1). Patients were then sent to the ultrasound suite for a post-MRI ultrasound examination to evaluate IUD placement. IUD displacement criteria included IUD crossbars no longer positioned within the fundal portion of the endometrial cavity compared to the pre-MRI ultrasound using 3D ultrasound. Two study radiologists confirmed the location of the IUD after the MRI. Patients that demonstrated displacement of the IUD were referred to their referring physician for replacement of the IUD and counseled to use another form of contraception until discussion with their referring physician (Fig. 1).

The data collected included the patient age, body part being imaged by MRI, and the presence or absence of pelvic pain before or after the study. The primary data point was the absence or presence of IUD displacement by post-MRI pelvic ultrasound analysis, as compared to pre-MRI pelvic ultrasound. The secondary data point was new or worsening pelvic pain, or any other new non-pelvic pain after the MRI examination.

As the only prior in-vivo study to date of Cu-IUDs at 3 T included 18 women, our study aimed to enroll between 50 and 100 patients to have over 2–5 times the number of patients in the first study [17]. The upper 95% confidence limit for p when no events were observed was calculated using the 3/n approximation as described by Hanley and Lippman-Hand [22].