The levonorgestrel intrauterine system (Mirena) for the treatment of menstrual problems in adolescents with medical disorders, or physical or learning disabilities


Dr M Pillai, Consultant Obstetrician Gynaecologist, Department of Obstetrics & Gynaecology, Gloucestershire Hospitals NHS Foundation Trust, Cheltenham General Hospital, Sandford Road, Cheltenham, GL53 7AN, UK. Email


Please cite this paper as: Pillai M, O’Brien K, Hill E. The levonorgestrel IUS (Mirena) for the treatment of menstrual problems in adolescents with medical disorders or physical or learning disabilities. BJOG 2010;117:216–221.

Menstrual problems are common in adolescence. Management can be very problematic in girls with disability or medical disorders, as their condition and other medications may mean they are less likely to achieve satisfactory control with standard medical treatments. Those with disability tend to tolerate menses poorly. We report experience with the levonorgestrel intrauterine system (LNG IUS) for treatment of menstrual problems in adolescents with a range of medical problems or with severe learning disability. It provided of significant therapeutic benefit for 12 of 14 participants, with the majority opting to continue for the 5-year duration of the device. The rates of expulsion, bleeding and amenorrhoea appear similar to those reported in older women, where contraception is the primary indication.


Being responsive to the very different healthcare needs of young people with disabilities is challenging. For girls with severe physical or learning disability, reaching the menarche can herald the onset of significant disruption to their lives. Issues around menstruation and contraception are often of major concern to caregivers. Caregivers frequently approach medical professionals seeking menstrual suppression. Fears about difficulties with hygiene, vulnerability to exploitation and pregnancy, and problems with the management of behaviour at the time of menstruation are common.

There is often a feeling that neither the girl nor the carer should have to deal with the inconvenience. It is not uncommon for families to request hysterectomy or permanent sterilisation. This raises complex legal and ethical issues that neither carers nor doctors are at liberty to decide upon in the UK.

It is estimated that 15% of adolescent girls seek medical help at least once for period problems, but probably many more do not present because of embarrassment. Menstrual irregularity has been reported in 55–82% of adolescents in the first 2 years post menarche, decreasing to 20% after around 4–5 years.1 In the majority this is physiological and reflects the fact that regular ovulatory cycles may take several years to establish after the menarche, because of the relative immaturity of the hypothalamic–pituitary–ovarian axis. The lack of luteal phase progesterone may then lead to unstable endometrium, resulting in heavy, prolonged bleeding. Other causes of anovulation can be pathologic, including hyperandrogenic states (like polycystic ovary syndrome, PCOS), hypothalamic dysfunction secondary to anorexia or excessive exercise (which usually results in amenorrhoea) and endocrinopathies. Within the population of teenagers with disability, medication and endocrinopathies may make a relatively more common contribution to menstrual problems. Many anticonvulsants increase the level of sex hormone binding globulins and decrease the levels of free sex hormones. Many frequently used medications, such as neuroleptics, can cause hyperprolactinaemia, which leads to irregular bleeding. Women with trisomy 21 have a high incidence of hypothyroidism, which can lead to irregular and heavy bleeding.

Materials and methods

The participants were either referred by a consultant paediatrician or by their general practitioner in the years 1998–2008. The ages of participants ranged from 11 to 21 years. All had received other medical treatments for menstrual problems that were unsatisfactory prior to using the levonorgestrel intrauterine system (LNG IUS). Acquired and congenital bleeding disorders were considered and tested for, where indicated. The treatment principles used for older women were followed, with anti-fibrinolytic and non-steroidal anti-inflammatory drugs being first line. In seven cases combined oral contraception (COC) was considered contraindicated because of immobility, liver disease, obesity with diabetes and/or hypertension, or a history of stroke. In the remaining seven cases COC was tried, and where swallowing or compliance was problematic the pill patch (Evra) was tried. In these seven cases COC was rejected in the longer term because of a lack of acceptable control or because of side effects. High-dose progestogen (usually norethisterione 15–45 mg, in divided doses) was used to control bleeding in refractory cases, but control could not be sustained with a lower dose regime or with a progestogen-only pill (POP). Depo-provera was not a good option for four participants because of difficulties in obtaining their acceptance of an intramuscular injection, and for two other participants because of obesity. For one participant other hormonal methods apart from the IUS were contraindicated because of significant liver dysfunction. In seven cases Depo-provera had been tried, but had resulted in protracted bleeding. For those without contraindications concurrent use of the pill was offered, but either failed to give a satisfactory bleeding pattern or was rejected for other concerns.

Many of the young women were unable to give consent for treatment. For children under age of 16, where possible, the child’s assent to the treatment was sought, as well as parental consent. For those who had attained the age of 18 but were not competent to give consent, treatment was given on the basis of best interests. However, in all cases there was agreement with the family or carer that this was serving the individual’s best interests. There were also cases not included here where the family disagreed with trying the IUS. They either persevered with other options or were referred for another opinion. Where the family persistently rejected alternatives and wanted hysterectomy, referral for another opinion and explanation that they would need to seek a court ruling was offered.

All but one participant required day-case admission for a brief general anaesthetic for the fitting.

The paediatric ward accommodated those with long-term disability up to the age of 21, as these girls were familiar with the ward setting. The one participant that did not require admission was a 15-year-old with congenital hemiplegia, who was both competent and already sexually active: advice on sexually transmitted infections (STIs) and screening was undertaken prospectively, and she was fitted with the IUS in a routine outpatient setting using instillagel. No difficulty was encountered.

In most cases a pelvic scan was performed prior to the admission to ensure the uterus would be appropriate for the IUS.


Table 1 illustrates the ages, medical conditions, menstrual problems and outcomes of treatment with an LNG IUS. The menstrual pathologies were one or more of dysmenorrhoea, polymenorrhoea and menorrhagia, which had failed to respond to standard medical therapies, or where co-morbidities precluded their use.

Table 1.   Summary of cases treated with LNG IUS
 Age/date IUS fittedIndication and summary of medical problemsOutcomeComments
  1. AXR, abdominal X-ray; CP, cerebral palsy; NET, norethisterone.

  2. *In case 3 a replacement IUS was requested 6 months after removal following marked deterioration in behaviour with return of periods.

1Fitted 1998
Age 16
Replaced 2003
Removed 2009
Difficulty with menstrual hygiene. CP & severe learning disability due to congenital CMV. Multiple medications for epilepsy. Not competent and vulnerable.Amenorrhoeic
No adverse effects
After 10 years with Mirena the family requested removal rather than replacement
2Fitted 2003
Age 19
Removed 2008
Difficulties with menstrual hygiene. Severe learning disability. Not competent and vulnerable. Behavioural problems.AmenorrhoeaFamily requested removal without replacement after 5 years. Reasons unclear.
3Fitted 2004
Age 20
Removed 2009
Menstrual distress and hygiene problems. Severe learning disability. Not competent and vulnerable.5 years amenorrhoeaAt 5 years family requested removal rather than replacement. No reason given.*
4Fitted 2004
Age 11
Removed 2009
Chaotic bleeding exacerbating behavioural problems. Failed medical treatments. Competent.Amenorrhoea with return of light periods during years 4/5.Requested removal at age 15 to become pregnant. Deferred and removed at 16 years.
5Fitted 2005
Age 15
Menorrhagia and contraceptive difficulties. Congenital stroke. Thrombophilia disorder. Competent and sexually active – IUS fitted routinelyAmenorrhoeaLast seen 2006
6Fitted 2005
Age 12
Replaced in 2009 for bleeding
Polymenorrhagia. Agenesis of corpus callosum. Not competent with multiple severe disabilities. Poorly controlled epilepsy – multiple medications.Prolonged bleeding managed with NET. Improved for 2–3 years then bleeding heavily againCurrently managing with new IUS plus cerazette.
7Fitted 2005
Age 17
Removed after 7 months
Multiple treatments for menorrhagia and dysmenorrhoea. Unhappy with every treatment tried. Quadriplegic CP but mentally competent.No improvement reported.Continuing referrals with pain/bleeding. Normal laparoscopy. Requested IUS removed
8Fitted 2005
Age 21
Severe dysmenorrhoea. Physical disability (myotonic dystrophy). Unable to swallow tablets. Mentally competent.Marked improvement – hypomenorrhoea and resolution of painContinuing with IUS
9Fitted 2006
Age 14
Menorrhagia/severe dysmenorrhoea. Failed medical treatments. CompetentHypomenorrhoeaLast seen 2007. Continuing with IUS
10Fitted 2006
Age 15
MenorrhagiaDiabetes (Type I & II) and obesity.Did not attend FU, but GP asked to refer back if any problemsSeen with abdo pain 2008. Scan and AXR confirmed IUS had been expelled
11Fitted 2006
Age 15
Chaotic bleeding Type II Diabetes, Obesity & Hypertension CompetentMarked improvement but takes additional NET for light bleedingContinuing with IUS. Seen age 17 when also sexually active
12Fitted 2007
Age 15
Severe dysmenorrhoea. Fatty liver with portal fibrosis. Obesity. Impaired glucose tolerance. Alpha-1 antitrypsin deficiency.Shorter less painful bleedsHappy and continuing with IUS
13Fitted 2007
Age 18
Difficulty coping with menses. Severe myotonic dystrophy. Not competent and vulnerable.AmenorrhoeaContinuing with IUS
14Fitted 2008
Age 17
Severe dysmenorrhoea CompetentResolution of symptoms
Continuing with IUS

Of the 14 young women, one (with quadriplegia and normal intellectual function, complaining of pain and protracted bleeding) was dissatisfied with the LNG IUS. She requested its removal after only 7 months. A laparoscopy was performed at the time of removal and was normal. All established medical treatments for the control of bleeding and chronic pelvic pain, including the COC, but not including long-term gonadotrophin-releasing hormone (GnRH) analogues, had been tried previously. It has not been possible to provide her with any satisfactory treatment. One other participant was confirmed to have expelled the IUS 2 years later, but had not been re-referred with menstrual problems. Two participants have required longer-term progestogen in addition to the IUS for continuing bleeding.

Three participants with a range of medical problems and who lacked competence appeared to have been very successfully managed, in that they became amenorrhoeic in the long term after having the IUS fitted. However, when the IUS was due to be replaced after 5 years (two participants), or re-replaced (10 years for one participant), their families requested to not have the IUS replaced.

For two participants the repeat day-case procedure was arranged through the GP, and the request to not have the IUS replaced was sent by a carer only on the day of admission. This left inadequate opportunity to establish the reasoning behind their decisions. In one of these a further IUS was requested within six months owing to return of difficulties with the resumption of her periods. In the case of the participant that had been amenorrhoeic for 10 years with the IUS, her mother thought that her difficult behaviour might be related to the ‘coil’, despite the explanation that this would be unlikely.

One participant (participant 5) has been lost to longer term follow up, but at the time of writing the other participants are all continuing with the IUS.

Participants 8 and 13 are related, and suffer from the same inherited neuromuscular disorder: participant 8 was competent, but participant 13 was much more severely disabled and incompetent. Participant 8 was 21-years-old by the time the IUS was fitted, but was still very ‘adolescent’ in that puberty and menarche had occurred late. Her mother was initially not receptive to the idea of a ‘coil’ when she was first referred at 19 years of age. With the IUS she experienced a complete resolution of menstrual symptoms and a significant improvement in her quality of life, so that the same option was readily requested when the younger more severely affected relative developed menstrual problems.

The contraceptive cover provided by the IUS is an additional benefit. Participant 4 had initially been referred by a paediatrician at 11 years of age with menorrhagia and dysmenorrhoea that had failed to respond to medical treatment. Her mother had a very positive view of the potential contraceptive protection offered by the IUS, since a 19-year-old sister already had four children. An IUS was fitted in 2005, which resulted in amenorrhoea. Following 4 years of satisfactory menstrual control, the girl, then aged 15, presented to her GP requesting the removal of the IUS in order to become pregnant.

Three doctors and the girl’s mother expressed a view that the removal of the LNG IUS would be against her best interests at this stage, which raised ethical issues about non-consent to treatment in under 16 years of age. She attended the clinic just after her 16th birthday, and came alone. She remained adamant that she wanted the IUS removed to become pregnant. At this clinic appointment we were able to screen her for STIs and offer advice about preconception care, as well as discuss the benefits of deferring pregnancy. The IUS was removed as requested. It is not unlikely that this device may have prevented an earlier pregnancy, and had allowed us to offer care and advice that would otherwise not have occurred.


Menstrual distress is frequently reported in adolescents with special healthcare needs, and often has a negative impact on behaviour and other health problems. There has been reluctance in the past amongst clinicians to use coils in adolescents. Some doctors do not offer coils at all to nulliparous young women. Both WHO and UK medical eligibility criteria consider those under the age of 20 as category 2, whereas women over 20 are considered as category 1 for intrauterine contraception, despite a lack of research-based evidence showing that teenagers are any more at risk of complications. This seems illogical against the background of an unacceptably high teenage pregnancy rate, and the fact that adolescents have significantly better compliance and continuation rates with coils than with pills.2

The National Institute for Clinical Excellence (NICE) guidelines advise that IUS use is not contraindicated in nulliparous women of any age.3 The LNG IUS has been shown to reduce menstrual loss by 85% within 3 months.4 The Cochrane 2005 meta-analysis of randomised controlled trials comparing the use of LNG IUS for the control of heavy menstrual bleeding against the use of oral cyclical norethisterone or balloon ablation demonstrated LNG IUS to be significantly more effective at reducing menstrual blood loss.5

There is little data on the IUS from USA and Canada as it has only recently been licenced. A Canadian study reported that Depo-provera was the most commonly prescribed and accepted method of menstrual suppression used in teenagers with disability.6 The association between long-term Depo-provera and reduced bone density is well established. The IUS has not been associated with any detrimental effect on the skeleton.

Endometriosis was thought to be rare until later reproductive years, but is now recognised to occur in adolescents, and may present as early as the menarche. The IUS has the added benefit of being therapeutic for endometriosis.

Quint7 has observed that ‘Patients with developmental disabilities are often treated lifelong once hormonal manipulation is started in the teenage years, because the hormones will get renewed year after year by different caregivers and the need for the hormones is not always periodically assessed.’ It may be a benefit that the IUS requires review every 5 years, in that the management options are then likely to be revisited. It was a surprise that in three women who had been very successfully treated, the carers requested not to continue with a second or third IUS. It is speculative that they may have lost sight of the difficulties for which the treatment was originally sought, or possibly changes in the challenging behaviour of these young women over time may have led to an idea that a hormone change might help.

There is no published research on the IUS in adolescence, but a reported series from New Zealand has shown very encouraging results with use in 179 adolescents between the ages of 11 and 19 years.8 The indications for treatment included contraception and menstrual problems. They demonstrated an 85% continuation rate with the coil after 1 year. The subgroup of adolescents affected with intellectual disability appeared to be over-represented (19% versus an expected 1%), which supports other reported experience that girls with disability may be more likely to suffer menstrual problems, as well as being less able to tolerate menstruation. The only other case series was 22 Australian women under the age of 20 reported in a meeting abstract.9 The indications included menstrual difficulties with intellectual disability, and not primarily contraception. There was one failed insertion, two expulsions and two removals. Symptom resolution was reported in 16 of the remaining 17 participants.

A practical consideration is that the LNG IUS requires a minimum uterine cavity length of 5–6 cm. As many of these participants will require anaesthesia for insertion it is highly desirable to do a pre-placement assessment with ultrasound. Although all had sufficient cavities, the lead author has experience of a minority of adolescent girls without a cavity length appropriate for this size of device. This may be more common in those closer to their menarche and in those who have recently finished long term Depo-provera. It is probably unlikely in those presenting with a history of protracted heavy bleeds. Newer smaller devices containing LNG (Femilis/Femilis slim) may be more appropriate for some cases when they become available in the UK.

Disadvantages of the IUS in this patient group include the possible need for a general anaesthetic for fitting and removal, and that additional medical management may be needed to control prolonged bleeding. The inconvenience and relative risk of a brief anaesthetic needed for fitting may be relatively insignificant when weighed against the potential benefits of 5 years of menstrual suppression, reliable protection from pregnancy, few risks and side effects, and removal of the long term need to take prescribed medication for menstrual problems. Additional progestogen or the COC was used in the short term in several participants. This was not acceptable in one participant, and was only partially effective in two others who continued progestogen in the long term. It is possible that other treatments that influence endometrial bleeding, like doxycycline, might be more effective in refractory cases. However, it was not tried in these cases, and as yet there is little data to guide its place, duration or dose for this purpose.

The potential for hormonal side effects should also be considered, although studies show that the rates of discontinuation as a result of side effects (such as acne and headache) are not significantly different from those using the copper intrauterine device.10


All of the limited data reported so far on adolescents with disability supports the use of the LNG IUS for this group.

Adolescents with physical and learning disability seem to have a high frequency of reported pain, heavy and prolonged bleeding, and hygiene problems during their menses. The LNG IUS provided good therapeutic benefit to a high proportion of adolescents in this category. This resulted in an improved quality of life for the young women, and in some situations for their carers. The additional contraceptive protection is an added bonus where contraception or vulnerability is a long-term concern. There were no adverse events, although there was one therapeutic failure and one case of expulsion among the 14 participants. The remaining 12 of 14 participants with medical problems derived some or significant benefit, where other treatments had failed.

Disclosure of interests


Contribution to authorship

MP was the clinician responsible for the management of these participants and is the main author of the paper. KO and EH assisted with the data collection and collation, literature review, and the drafting and editing process.

Details of ethics approval

None sought as this is reporting on the clinical outcome of response to an established treatment in a patient group rarely considered for this management.