The potential value of magnetic resonance imaging in infertility

Clinical Radiology (1995) 50, 75-79

The Potential Value of Magnetic Resonance Imaging in Infertility N. M. de SOUZA, J. J. BROSENS*, J. E. SCHWIESO, T. PARASCHOS* and R. M. L. WINSTON*

The Robert Steiner Magnetic Resonance Unit, Royal Postgraduate Medical School, and *Department of Obstetrics and Gynaecology, Hammersmith Hospital, London The potential value of magnetic resonance imaging as a diagnostic tool in infertility was investigated. Twenty-six women with primary or secondary infertility and symptoms of dysmenorrhoea or menorrhagia were studied prospectively using conventional T1- and T 2weighted spin-echo techniques. Positive diagnoses were obtained in 20 of 26 (76.9%) patients. Of these, 18 (69.2%) had lesions likely to be significantly contributing to infertility. Adenomyosis was detected in 14 patients (53.8%) with 11 showing the diffuse pattern while three had discrete adenomyomas. Cystic lesions typical of endometriosis were detected in seven patients (26.9%), four of these also had evidence of adenomyosis. The endometriotic lesions were also seen at laparoscopy in each case. Five patients (19.2%) had leiomyomas, one in a patient with adenomyosis and endometriosis and one in a patient with endometriosis alone. Only one patient had submucous leiomyomas causing significant distortion of the endometrial mucosa likely to affect fertility. Magnetic resonance imaging is valuable in the investigation of unexplained infertility where it provides a high diagnostic yield particularly if uterine pathology is suspected, de Souza, N.M., Brosens, J.J., Schwieso, J.E., Paraschos, T. & Winston, R.M.L. (1995). Clinical Radiology 50, 75-79. The Potential Value of Magnetic Resonance Imaging in Infertility

Accepted for Publication 23 September 1994

The investigation of infertility can be lengthy but often no definite cause is found. Multiple factors frequently contribute to the problem: uterine lesions such as adenomyosis and leiomyomas may be suspected if the patient presents with dysmenorrhoea or menorrhagia. Despite the fact that these lesions can be diagnosed on magnetic resonance (MR) imaging with a high degree of accuracy [1,2], MR imaging remains underused in the investigation of infertility. The purpose of this prospective study was to evaluate the potential role of M R in the assessment of uterine pathology in patients with infertility and symptoms of dysmenorrhoea or menorrhagia. PATIENTS AND METHODS Twenty-six patients, aged 26-41 years (mean 34.3 years), presenting with dysmenorrhoea and menorrhagia to the infertility clinic were recruited for MR scans. All had been trying to conceive for at least 18 months. Seventeen had primary infertility, five secondary infertility and four had recurrent miscarriages. All had dysmenorrhoea of moderate or severe degree and/or menstrual loss of moderate or severe degree. Laparoscopy had been performed as part of the routine infertility work-up in all cases. Details of clinical grading of menstrual characteristics and laparoscopic findings are given in Table 1. Two patients had recently commenced a gonadotrophin releasing hormone analogue (GnRH-a, goserelin depot) because of severe pain and evidence of endometriosis. Twenty-three patients were studied in the follicular phase of the menstrual cycle (between day 3 and day 13),

one in the secretory phase. The two patients treated with GnRH-a were amenorrhoeic. Magnetic resonance imaging was performed on a Picker 1.0T HPQ Vista system using a body coil. Sagittal and transverse 6.0 mm slices using conventional intermediate and T2-weighted spin-echo sequences (TR 2500ms, TE 20ms/80ms) with a 192 • 256 matrix, a 25cm FOV and 2 signal excitations were employed. Sagittal Tl-weighted spin-echo scans (TR 500ms, TE 20 ms) were obtained in some patients. Images were analysed for both lesion detection and characterization using previously published MR criteria [1,2] and were interpreted in conference by two radiologists without prior knowledge of the ultrasound findings. The mid-position sagittal slice of the T2-weighted set was used for measurements, and the width of the junctional zone and the outer myometrium were recorded. In patients with focal lesions, the number and size of each lesion was measured on every slice of the sagittal images. The largest area for each lesion was then recorded. Statistical analysis was performed using a separate variance Student's t-test with P < 0.05 considered significant. RESULTS Positive diagnoses were obtained in 20 of 26 patients (76.9%): 18 (69.2%) had lesions very likely to be contributing to infertility and two had incidental leiomyomas. Fourteen patients (53.8%) fulfilled the established M R criteria for adenomyosis. Adenomyosis

Correspondence to: Dr Nandita M. de Souza, The Robert Steiner Magnetic Resonance Unit, Hammersmith Hospital, DuCane Road, London Wl2 0HS.

Eleven of 14 patients (78.6%) showed the diffuse pattern of adenomyosis (Fig. 1) with between one and

76

CLINICAL RADIOLOGY

Table 1 - Patient characteristics, laparoscopic findings and MR diagnosis

Patient

Age (years)

Parity6ravida

Dysmenorrhoea* menorrhagia4f

Laparoscopic findings

MR diagnosis

1 2 3 4 5

30 41 35 36 33

2+s 2 +3 0 +~ 0 +~ 0 +5

I/II II/lI I/III III/II II/lI

Normal Leiomyomas Endometriomas Endometriosis Endometriosis

6 7 8 9 10 11

37 37 35 30 37 36

0 +~ 0 +~ 0 +~ 0+~ 0 +~

III/III I/II III/III III/III II/III II/III

12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

33 39 29 26 32 38 30 35 30 38 30 33 35 32 32

0 +~ 0 +~ 0 +~ 0 +~ 0 +~ 0 +1 0 +~ 0 +~ 0 +~ 3+o 1+0 2 +2 0 +~ 2 +o 0 +~

III/II II/II II/III III/III II/III II/II III/III III/III III/II III/III II/III I/III II/III III/I III/III

Normal Leiomyoma Endometriosis Left cornual block Frozen pelvis Left hydrosalpinx/right cornual block Normal Bilateral cornual block Left endometriomas Endometriomas Endometriomas/adhesions Bulky uterus Endometriosis Leiomyoma/endometriosis Endometriosis Normal Normal Normal Endometriosis Endometriosis Normal

Adenomyosis Leiomyoma Adenomyosis § endometriosis Normal Adenomyosis + endometriosis+ leiomyoma Leiomyoma Adenomyosis Adenomyoma + endometriosis Normal Adenomyoma Normal

0 +2

Normal Adenomyosis Normal Endometriosis Adenomyosis Leiomyoma Endometriosis Adenomyoma + eudometriosis Endometriosis + leiomyoma Adenomyosis Adenomyosis Adenomyosis Adenomyosis Adenomyosis Normal

* Graded: I, minor discomfort; II, pain requiring analgesia; III, requiring analgesia and affecting daily activity. "~Graded: I, no clots; II, clots up to 3 cm; III, clots >3 cm with episodes of flooding.

three ill-defined lesions in the outer myometrium and irregular junctional zone thickening (thickness 1 0 . 0 + 2 . 0 m m ) . The junctional zone to outer myornetrial ratio was increased in this group '(thickness ratio 1.13 4-0.35). This was significantly different from those patients with no evidence of adenomyosis or endometriosis (P~<0.02). Values for zonal thickness measurements are listed in Table 2. Three patients had discrete adenomyomas. The appearances of the adenomyomas which varied in size from 5.81 cm 2 to 43.2 cm 2 were quite characteristic (Fig. 2). At laparoscopy one of these patients had a myometrial biopsy which confirmed the presence of endometrial implants within the myometrium. In contrast to the

Table 2 - Zonal variation in thickness in uteri of patients with adenomyosis and endometriosis

Category

Junctional zone ( JZ) (rnm)

Fig. 1 - Diffuse adenomyosis: T2-weighted spin-echo (SE2500/80) midline sagittal section through the uterus showing thickening and irregularity of the low signal band o f junctional zone (short arrows). Diffuse infiltration of a mixed signal lesion (long arrow) is seen in the anterior wall.

Diffuse adenomyosis n=ll Adenomyoma n=3 Endometriosis (no adenomyosis) n=7 None of the above n=5

Thickness ratio

Thickness Outer myometrium (OM) (mm)

(JZ/OM)

10.0 -4-2.0

9.2 9 1.7

1.13 :k 0.35

4.7 • 1.0

13.1 • 0.7

0.36 • 0.07

5.2 + 2.2

10.0 + 2.5

0.59 -4- 0.36

5.9 t 1.5

10.1 • 4.1

0.66 -4- 0.29

77

MRI IN INFERTILITY

Fig. 2 Focal adenomyoma: T2-weighted spin-echo (SE2500 80) midline sagittal section through the uterus demonstrating a large ill defined mixed signal intensity mass characteristic of an adenomyoma (arrows).

(a)

diffuse group they did not show junctional zone thickening (thickness 4.7 + 1.0 mm) and the junctional zone to outer myometrial ratio was not increased (thickness ration 0.36 4-0.07) compared with those patients without adenomyosis or endometriosis (Table 2).

Endometriosis Seven patients in this series (26.9%) had cysts typical of endometriosis diagnosed on MR imaging (Table 1). Characteristically, multiple o/ten multilocular cysts with fluid levels were seen in the pelvis (Fig. 3). Laparoscopy confirmed the presence of ovarian 'chocolate cysts' in all cases and demonstrated diffuse peritoneal endometriosis in a further five cases. Seven patients had endometriosis with no evidence of adenomyosis: of these, three were diagnosed on M R imaging. Junctional zone thickening was not seen in this group (Table 2). Three of the seven patients with M R evidence of endometriosis had a rectovaginal mass noted on M R imaging. These deposits were solid rather than cystic reflecting their likely fibromuscular nature. One of these had been missed at laparoscopy (Fig. 4); deep pelvic endometriosis with obliteration of the pouch of Douglas was suspected at laparoscopy in the other two.

Leiomyomas Five patients (19.2%) had mural leiomyomas, one of these in a patient with adenomyosis and endometriosis and one in a case of endometriosis alone. Of the other three, only one patient had additional submucous leiomyomas causing significant distortion of the uterine cavity and cornua likely to affect fertility (Fig. 5). On

(h) Fig. 3 - Endometriotic deposits: (a) Ti-weighted spin-echo (SE500/20) and (b) F2-weighted spin-echo (SE2500/80) midline sagittal images at the same level. Multiple cystic lesions with fluid levels are seen. These are high signal on both sequences suggesting a haemorrhagic component.

M R imaging these lesions had the characteristic well defined appearance of leiomyomas but had been misinterpreted on transvaginal US because of poor definition. The average maximal cross-sectional area o f

78

CLINICAL RADIOLOGY

the leiomyomas was 2.8cm2+3.7cm 2 (range 0.3414.33 cm2). Transvaginal US had detected leiomyomas in four of the five cases. DISCUSSION

Fig. 4 Rectovaginal endometriosis: T2-weighted spin-echo (SE2500/ 80) sagittal section demonstrating a rectovaginal mass (arrows) in a patient with laparoscopically proven endometriosis. This was not detected at laparoscopy.

Fig. 5 - Leiomyomas: T2-weighted spin-echo (SE2500/80) sagittal section illustrating large well defined submucous leiomyomas distorting the uterine cavity (short arrows). The junctional zone width is normal (long arrow). The endocervical canal is compressed (arrowhead).

M R imaging prospectively diagnosed a factor likely to be contributing to infertility in 18 out of 26 (76.9%) patients with dysmenorrhoea and menorrhagia with M R features of adenomyosis present in over half of these. In our patient group there was a relatively low incidence of leiomyomas (19.2%). Conventional Tz-weighted spinecho scans provide good differentiation between the low signal junctional zone and the higher signal outer myometrium and allow assessment of myometrial lesions. Adenomyosis has been reported mainly in women in their fourth and fifth decades [3] with no specific studies on its incidence in infertility. This was because until recently its diagnosis has been dependent on the histological examination of uterine specimens, hence its apparent prevalence in multiparous females undergoing hysterectomy. Adenomyosis is usually reported if endometrial glands lie deeper than one-quarter of the endometrial thickness from the endometrial-myometrial junction [4]. In infertility where uterine conservation is paramount, the diagnosis of adenomyosis is often suggested by symptoms of hypermenorrhoea and dysmenorrhoea but similar symptoms are also produced by leiomyomas. Hysterosalpingography may show multiple small tracks of contrast extending into the myometrium but the results are often equivocal [5]. Also the pelvic radiation dose is a consideration in patients trying to conceive. Laparoscopy may reveal a nodular, sometimes injected looking uterine serosa. Myometrial biopsies taken at laparoscopy may be negative in focal disease and, therefore, the diagnosis of adenomyosis may be missed. Twenty-three patients in our group had prior laparoscopy but only one had myometrial biopsy; this was positive for adenomyosis. Ultrasound in skilled hands may be useful [6]: 20 of our patients had transvaginal scanning prior to M R imaging and adenomyosis was suspected in 11 of these. M R imaging confirmed adenomyosis in 10 out of 11 patients while one patient had multiple leiomyomas. With the establishment of MR criteria for diagnosis, adenomyosis may now be diagnosed more commonly in the reproductive years. Adenomyosis is best demonstrated on Tz-weighted images and is recognized in a focal and a diffuse pattern [7,8]. In focal adenomyosis there is a localized ill-defined mixed signal intensity mass (adenomyoma) within the myometrium: this pattern was seen in three of our patients. Diffuse adenomyosis presents with diffuse or irregular thickening of the junctional zone often with underlying high signal foci. Pathologically this has been shown to represent smooth muscle hypertrophy and hyperplasia surrounding a focus of basal endometrium [9]. The smooth muscle hypertrophy consists of closely packed muscle fibres that are more disorientated and less vascular than the smooth muscle of the normal inner myometrium [3]. It is the smooth muscle changes that are easily recognized by M R rather than the foci of heterotopic glandular epithelium. This feature of junctional zone thickening was not seen in

MRI IN INFERTILITY

the three patients with focal adenomyomas. These findings seem to support the hypothesis that focal adenomyomas arise from deep Mullerian rests within the myometrium while diffuse adenomyosis is the result of stromal and glandular invasion of the myometrium with secondary smooth muscle changes [10]. In our experience the thickening of the junctional zone that is so commonly described in adenomyosis [7] is not diagnostic of adenomyosis on its own. Absolute values for junctional zone widths though quoted are unhelpful: ratios to the width of the outer myometrium or the cervical stroma may be more meaningful [1,2,11]. Examination of three premenopausal women with severe menorrhagia prior to hysterectomy revealed junctional zone/outer myometrium ratios of >1.0 (unpublished data). On subsequent pathological examination there was no evidence of adenomyosis in any of the specimens. Whether in these cases there was adenomyosis that was too superficial to meet the histological criteria for definition or whether some totally unrelated cause accounted for this smooth muscle hypertrophy is speculative. The reported prevalence of endometriosis in women of reproductive age varies between 1 and 50% with infertile women seven to 10 times more likely to have endometriosis than their fertile counterparts [12]. A recent study of 57 consecutive laparotomies performed for infertility and pelvic pain reported an incidence of endometriosis as high as 84% [13]. The overall incidence in our study population was 46.2% which probably reflects the selection criteria. The role of MR imaging in the diagnosis of endometriosis depends on the site of the endometriotic implants: deposits on the ovarian cortex are much more readily identified than small peritoneal deposits. The latter are recognized at laparoscopy but may be missed on MR imaging [14]. MR imaging may, however, be valuable in detecting deep enclosed endometriosis typically found in the sacrouterine ligaments and rectovaginal septum. This type is associated with pelvic pain. It consists mainly of fibromuscular rather than endometrial tissue [14] and because of its position often goes undetected on visual inspection of the pelvic cavity. The incidence of leiomyomas in our study group was low and may be related to the age of the population. Although the diagnosis is usually made easily on US, MR imaging has been reported to be superior to US or hysterosalpingography for the diagnosis of leiomyomas [15-17]. In one of our patients the preceding hysterosalpingogram had shown an irregular outline with numerous tracks of contrast into the myometrium: appearances that could be mistaken for adenomyosis. Ultrasound in this patient revealed a bulky posterior uterine wall with high echogenicity. Several large leiomyomas were demonstrated on MR imaging. MR imaging is valuable in the investigation of infertility where it provides a high diagnostic yield in patients with dysmenorrhoea and menorrhagia. It should be part

79

of the investigation of patients with persistent unexplained infertility awaiting costly procedures such as gamete intrafallopian transfer (GIFT) and in vitro fertilization (IVF). A particular use of MR imaging would be in the infertility patient awaiting myomectomy as attempts to perform a myomectomy on a localized adenomyoma often result in extensive uterine damage. It may also have a place in the detection of rectovaginal endometriosis. The magnetic resonance scan may allow diagnosis of any uterine pathology and provides a non-invasive method of assessment of disease regression after treatment, thus optimizing the chances of successful assisted conception. Acknowledgements.We thank Professor R. E. Steiner for his assistance with image analysis and advice in preparing this manuscript. We remain grateful to the Medical Research Council for their financial backing. REFERENCES 1 Mark SA, Hricak H, Heinrichs LW et al. Adenomyosis and leiomyoma: differential diagnosis with MR imaging. American Journal of Roentgenology 1987;163:527-529. 2 Togashi K, Osaza H, Konishi Iet al. Enlarged uterus: differentiation between adenomyosis and leiomyoma on MR imaging. Radiology 1989;17h531-534. 3 Azziz R. Adenomyosis: current perspectives. Obstetric and Gynecologic Clinics in North America 1989;16:221-235. 4 Henrickson MR, Kempson RL. Surgical pathology o f the uterine corpus. Philadelphia: Saunders Co, 1980:452-467. 5 Marshak RH, Eliasoph J. The roentgen findings in adenomyosis. Radiology 1955;64:846-851. 6 Fedele L, Bianchi S, Dorta M e t al. Transvaginal ultrasonography in the diagnosis of diffuse adenomyosis. Fertility and Sterility 1992;58:94-98. 7 Togashi K, Nishimura K, Itoh K et al. Adenomyosis: diagnosis with MR imaging. Radiology 1988;166:111-114. 8 Hricak H, Fincks S, Honda G e t al. MR imaging in the evaluation of benign uterine masses: value of dimeglumine enhanced T1W images. American Journal of Roentgenology 1992;158:1043-1050. 9 Noval E, deLima A. A correlative study of adenomyosis and pelvic endometriosis with special reference to the hormonal reaction of eutopic endometrium. American Journal o f Obstetrics and Gynecology 1948;56:634-644. 10 Cullen TS. Adenomyoma of the uterus. Philadelphia: WB Saunders Co, 1908. 11 Hricak H, Alpers C, Crooks LE et al. Magnetic resonance imaging of the female pelvis: initial experience. American Journal of Roentgenology 1983;141:1119-1128. 12 Rock JA, Markham SM. Pathogenesis of endometriosis. Lancet 1993;340:1264-1267. 13 Koninckx PR, Meuleman C, Demeyre Set al. Suggestive evidence that pelvic endometriosis is a progressive disease, whereas deeply infiltrating endometriosis is associated with pelvic pain. Fertility and Sterility 1991;55:759-765. 14 Brosens IA. Classification of endometriosis revisited. Lancet 1993;341:630. 15 Hricak H, Lacey C, Schiock E et aL Gynaecologic masses: value of magnetic resonance imaging. American Journal of Obstetrics and Gynecology 1985;153:31-37. 16 Hamlin DJ, Petterson H, Fitzsimmons J e t aL MR imaging of uterine leiomyomas and their complications. Journal of Computer Assisted Tomography 1985;9:902-907. 17 Ascher SM, Arnold LL, Patt RH et al. Adenomyosis: prospective comparison of MR imaging and transvaginal sonography. Radiology 1994; 190:803-806.