Intrauterine polyps—a cause of unscheduled bleeding in women using the levonorgestrel intrauterine system: Case Report

S. Brechin1,4, S.T. Cameron1, A.M. Paterson2, A.R. Williams3 and H.O.D. Critchley1

1 Obstetrics and Gynaecology, Reproductive and Developmental Sciences, Centre for Reproductive Biology, 37 Chalmers Street, Edinburgh, EH3 9EW, 2 Obstetrics and Gynaecology, Dumfries and Galloway Royal Infirmary, Dumfries and 3 Department of Pathology, Medical School, Teviot Place, Edinburgh, EH8 9AG, UK


    Abstract
 Top
 Abstract
 Introduction
 Case reports
 Discussion
 References
 
The levonorgestrel intrauterine releasing system is a contraceptive that has been shown to reduce menstrual blood loss dramatically. Breakthrough bleeding, however, is a relatively common occurrence as with all methods of progestogen-only contraception and this limits its acceptability for women. Amenorrhoea can be achieved in the majority of women within 12 months of insertion. Any new pattern of bleeding after amenorrhoea or a persistence of heavy bleeding may be due to co-existing intrauterine pathology such as endometrial polyps. The use of out-patient techniques such as hysteroscopy and saline infusion sonography are indicated in these instances to exclude other intrauterine pathology.

Key words: intrauterine polyps/hysteroscopy/levonorgestrel/saline infusion sonography


    Introduction
 Top
 Abstract
 Introduction
 Case reports
 Discussion
 References
 
The levonorgestrel intrauterine system (LNG-IUS, Mirena, Leiras, Oy, Finland) has been licensed in the UK since 1995. It is a safe and effective method of contraception comparable to female sterilization (Sturridge and Guillebaud, 1997Go). A dose of 20 µg of LNG is released every 24 h to the endometrial cavity giving intrauterine concentrations up to 1000 times greater than that seen with systemically administered progestogens, such as norplant (Pekonen et al., 1992). This method avoids first-pass metabolism of progestogens in the liver and by its local administration there is little systemic absorption, equivalent to a quarter of that seen after taking oral LNG 30 µg, and therefore fewer unwanted systemic side-effects. One such side-effect, however, which is exploited is that of its beneficial effect of reducing menstrual blood loss. Menstrual blood losses are reduced by up to 82% after 3 months and 96% after 12 months (Andersson and Rybo, 1990Go; Milsom et al., 1991Go). Amenorrhoea or hypomenorrhoea will occur in up to 65% of women using LNG-IUS (Crosignani et al., 1997Go). It is such an effective treatment that 64.3% of women in a multicentre randomized trial cancelled their planned hysterectomies after treatment with this system (Lahteenmaki et al., 1999Go). Breakthrough bleeding, however, can occur in up to 53% of women after 3 months using the LNG-IUS (Irvine et al., 1998Go). The mechanisms underlying this pattern of bleeding, as seen with other progestogen regimes, are not clear. However, women who continue to have heavy bleeding after insertion or who develop bleeding after a period of amenorrhoea may have intrauterine pathology such as endometrial polyps and, as illustrated in the following three cases, a hysteroscopy or ultrasonography may be indicated in these women before considering removal of the LNG-IUS.


    Case reports
 Top
 Abstract
 Introduction
 Case reports
 Discussion
 References
 
Case 1
A 45-year-old parous patient, attended in 1996 with a 3-year history of regular heavy menstruation. At that time she had a normal vaginal examination and an ultrasound scan showing a small 2x2.3 cm submucous fibroid. Medications previously tried, without improvement in symptoms, included mefenamic acid, ethamsylate, norethisterone, and conjugated oestrogen with cyclical progestogens. Despite subjective menorrhagia she had a normal haemoglobin. A Pipelle biopsy showed normal secretory endometrium and after discussion, and in view of the need for concurrent contraception, an LNG-IUS was inserted. Within 6 months of insertion she had become amenorrhoeic and remained so for the next 12 months. However, she then presented with a recurrence of heavy and erratic bleeding. On clinical examination, the intrauterine system threads were visible and there was no evidence that the system had expelled within the endocervical canal. Hysteroscopy was performed and showed a large vascular intrauterine polyp, which was removed (Figure 1Go). Histology showed typical features of an endometrial polyp, in which endometrial glands were distributed throughout a vascular stroma, forming a rounded, smooth-surfaced mass. Glands showed some cystic dilatation and architectural distortion, but no evidence of hyperplasia or atypia. The stroma contained thick-walled vessels and increased collagen formation. The background endometrium showed inactive appearances. The LNG-IUS was removed at the time of polypectomy and not reinserted. Bleeding continued to be a problem over the next 2 months so a repeat hysteroscopy was performed. This again revealed a large polyp, which was removed completely, and the LNG-IUS reinserted. Curettings again confirmed normal endometrium. No further episodes of breakthrough bleeding have occurred.



View larger version (105K):
[in this window]
[in a new window]
 
Figure 1. Hysteroscopy showing vascular polyp alongside the levonorgestrel intrauterine system (LNG-IUS, Mirena) in case 1.

 
Case 2
A 42-year-old parous patient presented with a recent onset of heavy erratic bleeding after 12 months of amenorrhoea whilst using the LNG-IUS. Hysteroscopy revealed a large vascular intrauterine polyp. Histology showed very similar features to the polyp described for case 1. There was no evidence of hyperplasia or atypia. Although the polyp was removed and the LNG-IUS reinserted, bleeding continued to be a problem. A repeat hysteroscopy 4 months later revealed the recurrence of a large vascular intrauterine polyp. The LNG-IUS was not reinserted and the patient subsequently went on to have a laparoscopically assisted vaginal hysterectomy.

Case 3
A 67-year-old parous patient presented with recurrent postmenopausal bleeding, having had a menopause at the age of 42. She had a hysteroscopy, which showed a normal uterine cavity, and endometrial biopsies that consistently showed simple cystic hyperplasia. A consensus view from local gynaecological oncologists and pathologists advised high-dose progestogen therapy in order to avoid surgical management, if possible, in view of the patient's extensive past surgical history, which included a Hartmann's procedure. An LNG-IUS was inserted but bleeding did not settle despite endometrial biopsy showing pseudo-decidualization. In view of this, 3 months after insertion of the LNG-IUS a transvaginal ultrasound scan was performed, which demonstrated what appeared to be a thickened endometrium (22 mm) with a cystic appearance (Figure 2Go). In order to clarify the nature of the endometrial abnormality, saline infusion sonography (SIS) was performed. This technique involved the instillation of a small amount (5 ml) of sterile saline into the uterine cavity through a 5-Foley paediatric catheter (or hysterosalpingography or insemination catheter), followed by transvaginal ultrasound. As seen in the scan photographs (Figure 3Go), the anechoic saline distended the uterine cavity and clearly delineated an intracavity polyp, thus demonstrating that the endometrial lining was normal and that the apparent thickened endometrium, which had been observed with plain vaginal ultrasound, was the result of the polyp. Admission was arranged for in-patient hysteroscopy and polypectomy. Again pathology showed simple cystic hyperplasia within the endometrial polyp with surrounding endometrium showing pseudo-decidualization of the stroma and glandular atrophy. (Figure 4Go) She remained bleed free after the removal of this polyp and an outpatient hysteroscopy 6 months later showed an atrophic but otherwise normal uterine cavity with the LNG-IUS in situ. Currently the patient is bleed free, and biopsy is to be conducted only if bleeding recurs.



View larger version (100K):
[in this window]
[in a new window]
 
Figure 2. Transvaginal ultrasound scan showing thickened cystic endometrium in case 3.

 


View larger version (114K):
[in this window]
[in a new window]
 
Figure 3. Saline infusion sonograph delineating an intrauterine polyp in case 3: alongside the LNG-IUS (longitudinal), left panel, and in transverse, right panel.

 


View larger version (138K):
[in this window]
[in a new window]
 
Figure 4. Histology of endometrial polyp showing simple cystic hyperplasia with dilated glands in loose stroma. Original magnification x40.

 

    Discussion
 Top
 Abstract
 Introduction
 Case reports
 Discussion
 References
 
Breakthrough bleeding is a well-recognized problem with progestogen-only methods of contraception, the aetiology of which is unknown. The progestational effects seen as a result of the LNG-IUS are seen throughout the endometrial cavity and not only in the immediate vicinity of the intrauterine system.

Mechanisms responsible for the development of an endometrial polyp on the background of endometrial gland atrophy are unclear. Micropolyps have been described within the endometrium after a mean duration of use of 48 months (Silverberg et al., 1986Go). This finding, however, has not been confirmed by other groups who looked at endometrium up to 12 months after insertion of the intrauterine system (Critchley et al., 1998aGo). Changes within the endometrium, which resemble first-trimester decidua, with LNG-IUS use, were demonstrated. There was elevation of granulocyte–macrophage colony-stimulating factor (GM-CSF) and an increase in CD56+ large granular lymphocytes (LGL) seen with the pseudo-decidualization associated with the LNG-IUS as is seen in normal first-trimester decidua. However, bleeding seen with progestogen-only contraceptives is not commonly encountered in the first trimester of a healthy continuing pregnancy. Progesterone receptors (PR), mainly the PRB subtype, are down-regulated in the stroma of LNG-IUS users (Critchley et al., 1998bGo). A down-regulation of PR may result in the endometrium being less responsive to progestin-mediated events. Other factors may also be involved in breakthrough bleeding which may explain some of the underlying mechanisms. Metalloproteinase-9 (MMP-9) is an enzyme capable of degrading basement membrane. MMP-9 positive cells, neutrophils and eosinophils have been shown to be significantly increased in endometrial biopsies taken from users of the progestogen only implant, displaying a shedding morphology and in normal controls at menstruation (Vincent et al., 1999Go). This would suggest that MMP-9 may be involved in breakthrough bleeding in women.

Although not commonly used as a treatment for endometrial hyperplasia, the LNG-IUS used in this case produced a pseudo-decidualized endometrium with glandular atrophy, although hyperplasia persisted within the polyp itself. The use of an LNG-IUS releasing only 3 µg of levonorgestrel each day, has been used in women with histologically confirmed endometrial hyperplasia of all types (Perino et al., 1987Go). In 93% of cases the typical decidual reaction seen with LNG-IUS was evident without hyperplasia. In one case, of adenomatous hyperplasia with atypia, there was decidualization of the stroma but with persistent hyperplasia. The use of the 20 µg-releasing LNG-IUS was found to be associated with regression of hyperplasia, regardless of pattern, to an endometrium showing glandular atrophy and pseudo-decidualization (Scarselli et al., 1988Go)

Studies using SIS have established that it has an accuracy for detecting uterine cavity abnormalities at least equivalent to that of hysteroscopy (Goldstein, 1994Go; O'Connell et al., 1998Go). The advantages of SIS include the widely available inexpensive equipment, detection of uterine, adnexal and other pelvic abnormalities, and good performance in bleeding uteri.

Out-patient hysteroscopy and saline infusion scanning are warranted investigations in women who develop new symptoms of bleeding after a period of amenorrhoea, or who continue to complain of persistent heavy bleeding with an LNG-IUS in situ. Co-existing intrauterine pathology, such as endometrial polyps, can be excluded before considering removing the levonorgestrel intrauterine releasing system.


    Acknowledgments
 
We acknowledge Dr Sarah Chambers, Consultant Radiologist, for assistance with the saline infusion sonography.


    Notes
 
4 To whom correspondence should be addressed Back


    References
 Top
 Abstract
 Introduction
 Case reports
 Discussion
 References
 
Andersson, K. and Rybo, G. (1990) Levonorgestrel-releasing intrauterine device in the treatment of menorrhagia. Br. J. Obstet. Gynaecol., 97, 690–694.[ISI][Medline]

Crosignani, P.G., Vercellini, P., Mosconi, P. et al. (1997) Levonorgestrel-releasing intrauterine device versus hysteroscopic endometrial resection in the treatment of dysfunctional uterine bleeding. Obstet. Gynecol., 90, 257–263.[Abstract/Free Full Text]

Critchley, H.O.D., Wang, H., Kelly, R.W. et al. (1998a) Progestin receptor isoforms and prostaglandin dehydrogenase in the endometrium of women using a levonorgestrel-releasing intrauterine system. Hum. Reprod., 13, 1210–1217.[Abstract]

Critchley, H.O.D., Wang, H., Jones, R.L. et al. (1998b) Morphological and functional features of endometrial decidualization following long-term intrauterine levonorgestrel delivery. Hum. Reprod., 13, 1218–1224[Abstract]

Goldstein, S.R. (1994) Use of ultrasonography for triage of perimenopausal patients with unexplained uterine bleeding. Am J. Obstet. Gynecol., 170, 565–579[ISI][Medline]

Irvine, G.A., Campbell-Brown, M.B., Lumsden, M.A. et al. (1998) Randomised comparative trial of the levonorgestrel intrauterine system and norethisterone for the treatment of idiopathic menorrhagia. Br. J. Obstet. Gynaecol., 105, 592–598[ISI][Medline]

Lahteenmaki, P., Haukkamaa, M., Puolakka, P. et al. (1999) Open randomised study of the use of levonorgestrel releasing intrauterine system as an alternative to hysterectomy. Br. Med. J., 316, 1122–1126[Abstract/Free Full Text]

Milsom, I., Andersson, K., Andersch, B. et al. (1991) A comparison of flubroprofen, tranexamic acid and a levonorgestrel-releasing intrauterine contraceptive device in the treatment of idiopathic menorrhagia. Am. J. Obstet. Gynecol., 164, 879–883[ISI][Medline]

O'Connell, L.P., Fries, M.H., Zeringue, E. et al. (1998) Triage of abnormal postmenopausal bleeding: A comparison of endometrial biopsy and transvaginal sonohysterography versus fractional curettage with hysteroscopy. Am. J. Obstet. Gynecol., 178, 956–961[ISI][Medline]

Pekonnen, F., Nyman, T., Lahteenmaki, P. et al. (1992) Intrauterine progestin induced continuous insulin-like growth factor-binding-protein-1 production in the human endometrium. J. Clin. Endocrinol. Metab., 75, 660–664[Abstract]

Perino, A., Quartararo, P., Catinella, E. et al. (1987) Treatment of endometrial hyperplasia with levonorgestrel releasing intrauterine devices. Acta. Eur. Fertil., 18, 137–140[Medline]

Scarselli, G., Tantini, C., Colafranceschi, M. et al. (1988) Levo-norgestrel-nova-T and precancerous lesions of the endometrium. Eur. J. Gynaecol. Oncol., 9, 284–286.[Medline]

Silverberg, S.G., Haukkama, M., Arko, H. et al. (1986) Endometrial morphology during long-term use of levonorgestrel-releasing intrauterine device. Int. J. Gynecol. Pathol., 5, 235–241[ISI][Medline]

Sturridge, F. and Guillebaud, J. (1997) Gynaecological aspects of the levonorgestrel-releasing intrauterine system. Br. J. Obstet. Gynaecol., 104, 285–289.[ISI][Medline]

Vincent, A.J., Malakooti, N., Zhang, J. et al. (1999) Endometrial breakdown in women using Norplant is associated with migratory cells expressing matrix metalloproteinase-9 (gelatinase B) Hum. Reprod., 14, 807–815.[Abstract/Free Full Text]

Submitted on September 24, 1999; accepted on November 29, 1999.