Article Text
Abstract
Pregnancy luteoma is a benign ovarian tumour that presents during pregnancy and regresses spontaneously post partum. The mass may secrete androgenic hormones which can result in virilisation of mother and female foetus. Clinical presentation of pregnancy luteoma is varied from asymptomatic to significant virilisation. We present two cases of ovarian luteomas identified incidentally at caesarean section. The first case had marked hirsutism, while the second case had mild acne. Both foetuses were male and appeared unaffected at birth. Where maternal virilisation is identified during pregnancy, differential diagnoses including pregnancy luteoma should be considered.
- Pregnancy
- Gynecological cancer
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
Background
Luteoma of pregnancy is a benign mass of theca lutein ovarian cells that secrete excess androgens in the presence of human chorionic gonadotropin, which can result in virilisation of the pregnant person and any female sex foetus.1 This case series describes two patients with luteoma of pregnancy, each diagnosed at the time of caesarean section. Early application of the differential diagnosis of luteoma of pregnancy can assist in management of patients in the antenatal, intrapartum and postpartum periods, as will be described.
Case presentation
Case 1
A gravida 2 para 0 pregnant woman in her 30s underwent induction of labour for pre-eclampsia and macrosomia at 39 weeks and 3 days gestation. She required an emergency caesarean section at 3 cm dilatation for foetal distress and obstructed labour. Yellow intraperitoneal fluid was noted on entering the abdomen, which could have been attributed to obstructed labour. Intraoperatively, an 80-mm right ovarian mass was identified. It had a smooth brown surface which split open with manipulation of the ovary to reveal solid, friable, brown, haemorrhagic tissue inside (figure 1). There was no evidence of overt bowel, omental or peritoneal disease. The left ovary and both fallopian tubes appeared normal. With verbal patient consent, two samples were biopsied with advanced bipolar energy and sent in formalin for histopathology. As it was unplanned and ‘after hours’, intraoperative frozen section histological examination was not available.
Histological examination showed features of a pregnancy luteoma with large eosinophilic epithelioid cells with eosinophilic cytoplasm (figure 2). There were areas of acute haemorrhage and no intact ovarian stroma. Using immunohistochemistry, the luteinised cells were reactive with inhibin A and melan-A, and weakly to Wilms tumour 1 (WT1), but showed no reactivity with oestrogen receptor (ER) or the broad spectrum cytokeratin AE1/AE3.
On review of antenatal documentation, no abdominal masses were noted on examination or on her antenatal ultrasounds at 24 and 32 weeks gestation, although the ultrasounds were targeted towards growth and the ovaries were not included in the formal assessment. The patient retrospectively reported virilising symptoms, including hirsutism on the chin, abdomen and thighs; acne on the face, chest and arms; and a deepened voice. These symptoms started during the third trimester and did not regress completely until 4 months post partum.
While awaiting the tissue diagnosis of the solid lesion, empiric tumour marker serum tests were ordered with the inadvertent absence of inhibin A. Beta human chorionic gonadotropin and alpha fetoprotein were not requested due to pregnancy. The results were normal apart from the cancer antigen-125 (CA-125), which was elevated, potentially due to pregnancy or recent abdominal peritoneal surgery. 3 weeks post partum, following the diagnosis of luteoma, serum androgens were requested with the goal to find an indirect marker for tumour surveillance. The serum testosterone and dehydroepiandrosterone sulphate levels were normal (table 1).
Case 2
A gravida 1 para 0 pregnant woman in her 20s underwent induction of labour for thrombocytopenia and elevated body mass index (BMI) at 39 weeks and 3 days gestation. She required an emergency caesarean section for foetal distress at 5 cm dilatation. A left ovarian cyst measuring 40 mm was identified. Macroscopically, it resembled a simple ovarian cyst with a smooth, shiny exterior containing serous fluid. There was no evidence of bowel, omental or peritoneal disease, and the right ovary and both fallopian tubes were normal. With verbal patient consent, the cyst was removed intact and completely.
Histological examination revealed clusters of cells with abundant eosinophilic and focally granular cytoplasm which was set within a highly oedematous ovarian stroma (figure 3). Immunohistochemistry showed a patchy reactivity to inhibin A and melan-A and low ER reactivity in approximately 50% of cells. There was no reactivity to AE1/AE3 or WT1. A reticulin stain was variable showing focal staining surrounding both small clusters of cells as well as individual cells. Overall, the features were best in keeping with a sex cord stromal lesion favoured to represent a pregnancy luteoma with degenerative change.
Antenatally, no abdominal masses on examination or on ultrasound were reported. Retrospectively, the patient did report acne on the cheeks and chin during the third trimester; however, this was in keeping with occasional acne in her medical history. She denied any other virilism or hirsutism symptoms. Tumour markers were not sent as the intraoperative suspicion for malignancy was low. On diagnosis of the luteoma of pregnancy, the patient was 8 weeks post partum and asymptomatic. Serum testosterone was therefore not requested.
Differential diagnosis
When faced with an unexpected solid ovarian mass at caesarean section, the differentials are carcinoma (such as epithelial, germ cell or sex cord stromal tumour), benign tumours (including endometrioma, pregnancy luteoma or fibroma) or tumours of borderline malignant potential.2 3 Hyperreactio luteinalis (HL) also causes maternal virilisation; however, this is typically associated with bilateral cystic ovaries and hypertrophied theca interna cells.4 The large, solid and friable characteristics of the mass described in case 1 were suspicious for a malignant neoplasm, while the simple appearance of the ovarian cyst in case 2 favoured a benign ovarian cyst. In both cases, the definitive diagnosis was confirmed via histopathology as a pregnancy luteoma.
Treatment
Luteomas naturally regress in the absence of human chorionic gonadotrophin.5 Surgical management includes cystectomy-style excision of the lesion (such as in case 2) or oophorectomy. Case 1, who had incomplete removal of her lesion at the time of her caesarean section, required no further treatment. Given the natural course of pregnancy luteoma and the resolution of her symptoms, surveillance ultrasound was not performed. As previously described, case 2 had complete excision of her luteoma intraoperatively.
Outcome and follow-up
Case 1 was followed up 3 and 4 months post partum. She reported the hirsute features had resolved by 4 months post partum, and the acne and deep voice were the last symptoms to regress. Case 2 was followed up 2 months post partum at which point she did not report any virilisation or hirsutism.
Discussion
This report describes two cases of pregnancy luteoma with different macroscopic, microscopic and clinical manifestations. The first case experienced maternal hirsutism and virilisation that persisted for 4 months post partum, while the second case reported only mild acne that resolved within 1 month post partum. Histopathology from both cases showed eosinophilic epithelioid cells with eosinophilic cytoplasm, consistent with previous reports of luteoma histopathology.1 6 Both luteomas were immunoreactive to inhibin A and melan-A in keeping with cells of sex cord stromal origin.7 The first luteoma was weakly WT1 positive but ER negative, while the second luteoma was weakly ER positive and WT1 negative (table 2).8 WT1 is a marker expressed in ovarian serous epithelial tumours and sex cord stromal tumours.9
Pregnancy luteoma was first described in 1966 by Dr William Sternberg as an ovarian tumour that emerges during pregnancy and regresses spontaneously post partum.10 There have been many cases of pregnancy luteoma reported in the literature, each describing a range of clinical presentations.11 The true incidence is not well documented, and there are likely many undetected cases. One study identified eight luteomas among a screening programme of 64 580 pregnancies where abdominal ultrasounds were performed in the first, second and third trimesters.12 It is usually during caesarean section, routine antenatal imaging or from targeted imaging for acute abdominal pain where a luteoma is found incidentally.5 11
Where maternal hyperandrogenism is associated with virilising symptoms, there is an 80% chance the female fetus will also be virilised. The remaining 20% of infants are likely protected by the aromatisation of excess maternal androgen by the placenta.13 14 If there is no maternal virilisation, then it is accepted that the foetus will not be virilised.15 The concern for foetal virilisation is generally reserved for female foetuses; however, elevated prenatal testosterone may also cause reproductive dysfunction in the male foetus.16 Studies in sheep have demonstrated that excess prenatal testosterone leads to reduced sperm count and motility, alternations in Sertoli and germ cell number and expression of testicular follicle-stimulating hormone.17 18 Both foetuses in this report were male. Interestingly, studies have identified an association between pre-eclampsia and serum testosterone, with elevated maternal androgen levels considered an important risk factor for pre-eclampsia. This is noteworthy given case 1 developed pre-eclampsia in her pregnancy.19
Luteomas are bilateral in 25%–50% of cases, and maternal virilisation or hirsutism is reported in approximately 25% to 65% of cases.6 11 12 20 There are reports of recurrent luteomas in subsequent pregnancies, although this is rare and there are no data available on recurrence rate.1 21 High serum androgen concentrations are not always associated with maternal virilisation. One case report described two women with pregnancy luteomas who had very high serum testosterone levels but no clinical virilisation.22 This is in contrast to case 1 in this report, where the patient was virilised but had normal testosterone concentrations. The limitation of delayed testosterone testing (measured 3 weeks post partum) is mitigated by the fact that her virilising symptoms persisted at this time.
Luteomas are distinct from other causes of gestational hyperandrogenism such as HL, Sertoli-Leydig and Krukenberg tumours. Like luteomas, HL is a benign, self-limiting, pregnancy-associated ovarian mass that can cause virilisation via excess androgen production. Unlike luteomas, HL is characterised by bilateral cystic ovaries with hypertrophied theca interna cells. The significant cystic ovarian enlargement gives rise to a distinct sonographic ‘spoke-wheel’ appearance of HL which may help to differentiate these tumours from luteomas.4 HL is more commonly associated with pregnancies complicated by trophoblastic disease.20
Sertoli-Leydig cell tumours (SLCT) and Krukenberg tumours should also be considered among differentials of androgen-secreting tumours. Krukenberg tumours are ovarian metastatic adenocarcinomas that most commonly arise from the stomach; however, it can also originate in the appendix, colon, small bowel, gallbladder, pancreas or breast. They are often bilateral and can be solid, cystic or a combination. Compared with primary ovarian tumours, Krukenberg tumours are more commonly seen in younger women. Histologically, Krukenberg tumours are characterised by signet ring cell mucinous features.23 Occasionally, Krukenberg tumours produce androgens by luteinisation of the tumorous stroma, which can cause virilisation. Studies suggest that Kurekberg tumours are more likely to increase androgen production during pregnancy due to the elevated levels of human chorionic gonadotropin.24 In relation to case 1, the clinical picture of virilisation, hirsutism and solid ovarian mass make Krukenberg tumour a pertinent differential diagnosis, which could only be ruled out after the histopathological analysis confirmed a pregnancy luteoma.
SLCT is a rare, primary ovarian neoplasm of sex cord stromal origin. It is more common in younger women, usually unilateral, and has a high incidence of virilisation (40%–80%), making it an important clinical differential in case 1 of this series.25 26 They are very rarely associated with pregnancy; however, there have been documented cases in the literature.27 28 Ultrasound characteristics of SLCT are nonspecific and variable ranging from solid to multilocular masses.25
There are no guidelines for the optimal management of luteoma in pregnancy.11 Maternal virilisation in pregnancy should be investigated promptly. Abdominal ultrasound is the best modality of identifying ovarian masses antenatally; however, adequate visualisation can be difficult at an advanced gestation. The 80-mm luteoma in case 1 was not identified despite obstetric ultrasounds performed at 24 and 32 weeks (this may have been due to the luteoma possibly being smaller at this earlier gestation, difficult views of the ovaries due to the gravid uterus or the patient’s high BMI). Early identification of a solid ovarian mass during pregnancy is important for diagnostic workup, clinical differentials, delivery planning and biopsy including intraoperative frozen section. Luteoma confirmed with tissue diagnosis is appropriate for conservative management as these lesions regress spontaneously.5
Hyperandrogenism is a complaint that distresses patients and confuses clinicians. Hirsutism, voice deepening and acne are socially impactful and, in our cases, medically important. Increased clinician awareness of these symptoms, targeted antenatal imaging and serum androgen testing may help with antenatal diagnosis of pregnancy luteoma, noting that tissue diagnosis is required to differentiate from other causes of maternal virilisation.
Patient’s perspective
Case 1
I first noticed the symptoms in my third trimester. My symptoms were not only the hair growth on the chin, inner thigh and stomach but also the deep voice and sharp pain in my lower right abdomen. I told my midwife about it and she referred me for an ultrasound which didn’t show anything, so I concluded it could have been the position of the baby that was giving me the sharp pain. Little did I know that it could have been a growth on my ovary. As to how the symptoms made me feel, I was going to work and did not quite feel awkward, rather I made jokes about my facial hair.
Case 2
When I was pregnant with my son, I did not give much thought to how little things like acne could be possibly related to a pregnancy luteoma. I developed acne during the third trimester, predominantly on my cheeks and chin. At the time I thought it was odd, but not something of concern considering I’ve always suffered with acne. It began to go away within the first month post partum. Moving forward 4 months, the acne has not reappeared as intensely as it was during my pregnancy. I cannot confirm whether the acne is related to a luteoma and I did not report my symptoms to a midwife or doctor knowing it was common for me to have it. Nonetheless, I did not experience any other symptoms relating to a luteoma, but I will be conscious in future pregnancies.
Learning points
Luteoma of pregnancy can cause virilisation of the female sex foetus, especially when associated with elevated maternal serum androgens.
Investigation of maternal hirsutism and virilisation is important to identify pregnancy luteoma or malignant hyperandrogenic neoplasms.
If a pregnancy luteoma is suspected, biopsy should be taken to exclude malignancy.
Ethics statements
Patient consent for publication
Acknowledgments
The authors would like to acknowledge the help and expertise of Dr Diana Moir.
References
Footnotes
Contributors The following authors were responsible for drafting of the text, sourcing and editing of clinical images, investigation results, drawing original diagrams and algorithms and critical revision for important intellectual content: IS and ET. The following authors gave the final approval of the manuscript: IS and ET.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Case reports provide a valuable learning resource for the scientific community and can indicate areas of interest for future research. They should not be used in isolation to guide treatment choices or public health policy.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.