Elevated concentrations of circulating testosterone are present in hyperreactio luteinalis (HL), a pregnancy-specific, self-limited condition. HL is associated with maternal virilization in about 30% of cases. The correlation between testosterone levels and maternal virilization has not yet been quantified. Our aim was to identify a testosterone cut-off level which may allow to predict maternal virilization.
A literature research was performed. Publications were chosen if serum testosterone concentrations and presence or absence of maternal virilization was mentioned. Additionally, we report serial levels of steroids analyzed by Liquid chromatography–tandem mass spectrometry (LC–MS/MS) in one case of HL managed at our institution.
In all, 31 cases fulfilled the search criteria. We found significant overlap between testosterone levels in asymptomatic women and women with signs of virilization (range 6.2–37.3 nmol/l and 13.7–197.5 nmol/l, respectively). The method applied for testosterone analysis was mentioned in three reports only. Peak serum testosterone concentration in our case was 120.3 nmol/l.
From the available data, maternal virilization in HL cannot be predicted by the level of circulating testosterone. However, comparability of results is hampered by the analytical methods applied. LC–MS/MS should preferably be used for reporting concentrations of circulating testosterone.
Maternal virilization during pregnancy or the puerperium is an extremely rare event. It may occur by coincidence, caused by congenital adrenal hyperplasia or adrenal tumors. However, elevated levels of circulating androgens usually result in anovulatory cycles or amenorrhea in affected women. In the vast majority of cases, hyperandrogenemia with or without maternal virilization constitutes a pregnancy-associated condition. Here, excess amounts of testosterone arise from luteomas or hyperreactio luteinalis (HL), where androgens are synthesized in the affected ovaries, and deficiency of placental aromatase, resulting in insufficient metabolization of circulating androgens. Further details are reviewed in Kanova et al. and Hakim et al. [1, 2].
Both, luteomas and HL are benign, self-limited conditions. Luteomas arise from proliferating luteinized cells; some 50% of cases occur unilaterally . Hyperreactio luteinalis can be found in pregnancies with elevated levels of circulating human chorionic gonadotrophin (beta-HCG). Conditions associated with increased HCG production include multiple pregnancy, placental or fetal hydrops, and gestational trophoblastic diseases . However, HL can even occur in pregnancies where beta-HCG concentrations are within normal range. ‘Hypersensitivity’ towards circulating beta-HCG, caused by polymorphisms of the protein or its receptor may result in a gain-of function and may underlie the manifestation of HL with normal beta-HCG levels . Characterized by bilateral multicystic, grossly enlarged ovaries (‘spoke wheel’), symptoms other than maternal virilization may include abdominal discomfort or acute-onset pain caused by ovarian torsion or cyst rupture with the development of hemoperitoneum . The majority of cases, however, either go unnoticed or are a chance finding at the time of obstetric ultrasound or during cesarean delivery . In both conditions, luteoma and HL, hormone levels return to normal range and the ovarian enlargement regresses spontaneously after delivery. A conservative management approach with close follow-up is, therefore, appropriate, reserving surgical intervention for acute complications .
Pregnancy-associated virilization may occur in the mother or the fetus, or both. Maternal symptoms include hirsutism in androgen-dependent areas, acne, temporal balding, hypertrophy of the clitoris, and deepening voice, the latter being irreversible in the majority of cases . Female pseudohermaphroditism in the context of pregnancy-associated hyperandrogenemia is exceedingly rare and has been described in less than ten cases; gestational age at exposure may play a role .
Maternal virilization in HL occurs in some 30% of cases . The association of circulating testosterone levels with maternal virilization has not been systematically analyzed and predictors are not available. In particular, there is no threshold level of circulating testosterone which could help in risk assessment. This may be due to the following reasons: First, number of reported cases: less than 100 cases are published, the majority being case reports. Second, the analytical method used for measurement.
Unfortunately, there is no standardization for testosterone analysis. Testosterone and its metabolites are usually measured by radio-immunoassays (RIA), immunoassays, or Liquid chromatography–tandem mass spectrometry (LC–MS/MS). Strengths and shortcomings of immunoassays include rapid results and high throughput, but no standardization regarding antibodies used is available . Furthermore, specific laboratory requirements are mandatory for RIAs as radioactive tests. LC–MS allows to analyze multiple steroids within one run with low sample volume; additionally, in contrast to immunoassays, no cross-reactions with other metabolites are observed.
The aim of our study was to correlate maternal testosterone levels in HL with clinical symptoms of maternal virilization. For this, we performed a literature research and compiled the published data. Additionally, we report the clinical course and serial hormonal analyses using LC–MS/MS in one patient under our care who developed maternal virilization during the second trimester of pregnancy.
A 34-year-old Asian primigravida was referred to our department in the 21st week of gestation. She complained of tachycardia, palpitations, sleeplessness and loss of weight since 4 weeks and had been started on metoprolol (47.5 mg/day) for symptom control. Thyroid function tests had been within normal range on several occasions (see Table 1). She had conceived after ovarian stimulation and insemination. Gestational diabetes had been diagnosed at 17 weeks of gestation. Ultrasound examination revealed an appropriate for gestational age singleton male fetus, and bilaterally enlarged polycystic ovaries (see Fig. 1). Tumor markers (CA125, Inhibin B) and beta-HCG were within normal range. After extensive discussion, a conservative management with close monitoring was agreed upon, and symptoms of ovarian torsion were explained.
At 32 weeks of gestation, symptoms of hyperthyroidism had subsided, and metoprolol had been tapered. Ovarian size was stable. She now complained of deepening and hoarseness of her voice; when asked, she confirmed worsening seborrhea and hypertrophy of the clitoris. Hormone analysis revealed excessively elevated androgen levels (see Table 1). A provisional diagnosis of HL was established. The patient was reluctant to undergo any further diagnostic tests. A conservative approach with close clinical and laboratory controls was agreed upon. Further complications did not occur. Induction of labor was performed at 37 weeks of gestation with the aim to reduce prolonged maternal and fetal exposure to high circulating testosterone levels. She delivered a healthy male baby without signs of hyperandrogenism (weight 2920 g, Apgar 9/10/10, umbilical artery pH 7.22).
Breastfeeding was initiated successfully without delay. Two days after delivery, magnetic resonance imaging (MRI) was performed to rule out adrenal or ovarian malignancy (see Fig. 2). During the subsequent weeks, serial ultrasound examinations showed a progressive decline of the ovarian volume; serum androgen levels normalized within 20 days. The vocal changes improved gradually; however, 1 year after delivery, a full recovery has not occurred.
Venous blood samples were drawn and hormone analyses were performed at the central laboratory of the University Hospital Bonn. Serum steroids (androstenedione, progesterone, cortisol, dehydroepiandrosterone sulfate (DHEAS), estradiol, 17-OH-progesterone and testosterone) were analyzed with LC–MS/MS analysis by MassChrom® assay (Chromsystems, Graefelfing, Germany) according to the manufacturer’s protocol. A Xevo® TQ-S mass spectrometer (Waters, Eschborn, Germany) with electrospray ionization was applied for LC–MS/MS analysis. Adrenocorticotropic hormone (ACTH), luteinizing hormone (LH), follicle-stimulating hormone (FSH), human chorionic gonadotropin (HCG and beta-HCG) and sex hormone-binding globulin (SHBG) were measured by electrochemiluminescence immunoassays (cobas® e801, Roche Diagnostics, Mannheim, Germany).
A literature research was performed covering the time between 1975 and 2018, using PubMed database. The search was restricted to English language. Publications were chosen for the review if androgen concentrations and the presence or absence of maternal virilization were mentioned. Additionally, references of publications were handsearched for further reports.
Our literature research revealed 40 cases with hyperandrogenemia due to HL [1, 5,6,7,8,9,10, 13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41]. In 29 cases, testosterone levels were measured. Only four publications [10, 13, 15, 18] mentioned the method of testosterone measurement. One case was excluded due to implausibly high values . For further analysis, we focused on the 29 cases where a comparison between testosterone levels was possible, see Fig. 3. In 16 cases, clinical symptoms of maternal virilization occurred. In the other cases, no symptoms were described or maternal virilization was not mentioned in the publication. In clinically asymptomatic women, testosterone concentration varied between 6.2 and 37.3 nmol/l; in cases with maternal virilization, the testosterone concentrations ranged between 13.7 and 197.5 nmol/l, see Fig. 3.
Our analysis of published cases of maternal virilization in hyperreactio luteinalis shows that there is no testosterone cut-off level to predict the development of maternal virilization in HL.
One bias might be the technique used for testosterone measurements. In most of the reported cases, authors did not mention the method used for determination. As immunoassays and LC–MS/MS results are not comparable, a direct comparison between the published cases is difficult. We recommend that future reports on hyperandrogenism should include statements of the laboratory techniques and assays used. Preferably, testosterone values should be obtained with state-of-the-art LC–MS technique due to the absence of cross-reactions. In addition, in contrast to immunoassays, LC–MS offers a large linearity. This is another advantage particularly for cases with high testosterone levels.
Genetic polymorphisms of testosterone receptors may also be a reason for the varying clinical manifestations with the same levels of testosterone . The androgen receptor gene is located on the X chromosome and contains a highly polymorphic region with a variable number of CAG repeats, which normally ranges between 9 and 38 (mean range between 21 and 23) and varies with ethnicity. Epidemiological studies have associated shorter CAG repeats with hyperandrogenic syndromes, and in vitro studies have shown a negative association between androgen receptor activation and CAG repeats length .
In all published cases, hormone levels and ovarian size normalized after delivery, obviating the need for surgical interventions during pregnancy, and reserving them for the management of acute complications . Some publications mention delayed lactation due to elevated testosterone levels [24, 26]. In our case, breastfeeding was successfully initiated without delay despite elevated testosterone concentrations (testosterone 95.0 nmol/l on day two after delivery).
Once the testosterone levels return to normal range, most of the clinical symptoms of maternal virilization resolve. However, vocal changes may be irreversible, and without options for medical treatments.
In conclusion, maternal virilization in the context of HL is a rare, self-limited condition. Interventions during pregnancy are usually not necessary unless surgical complications, ovarian torsion in particular, do occur. Virilization cannot be predicted by the level of circulating testosterone. For the sake of comparability, reports on hyperandrogenemia should state the analytical method used.
Free androgen index
Human chorionic gonadotropin
Sex hormone-binding globulin
Kanova N, Bicikova M (2011) Hyperandrogenic states in pregnancy. Physiol Res 60(2):243–252
Hakim C, Padmanabhan V, Vyas AK (2017) Gestational hyperandrogenism in developmental programming. Endocrinology 158(2):199–212
Masarie K, Katz V, Balderston K (2010) Pregnancy luteomas: clinical presentations and management strategies. Obstet Gynecol Surv 65(9):575–582
Masuyama H et al (2009) Hyperreactio luteinalis with both markedly elevated human chorionic gonadotropin levels and an imbalance of angiogenic factors subsequently developed severe early-onset preeclampsia. Fertil Steril 92(1):393-e1
Amoah C et al (2011) Hyperreactio luteinalis in pregnancy. Fertil Steril 95(7):2429-e1
Li Q, Li X, Zhang P (2015) Ovarian torsion caused by hyperreactio luteinalis in the third trimester of pregnancy: a case report. Int J Clin Exp Med 8(10):19612–19615
Sarmento Goncalves I et al (2015) Hyperreactio luteinalis encountered during caesarean delivery of an uncomplicated spontaneous singleton pregnancy. BMJ Case Rep 2015:bcr2015211349
Cavoretto P et al (2014) Hyperreactio luteinalis: timely diagnosis minimizes the risk of oophorectomy and alerts clinicians to the associated risk of placental insufficiency. Eur J Obstet Gynecol Reprod Biol 176:10–16
Deknuydt M et al (2018) Recurrent maternal virilization during pregnancy in patients with PCOS: two clinical cases. Reprod Biol Endocrinol 16(1):107
Hensleigh PA, Carter RP, Grotjan HE Jr (1975) Fetal protection against masculinization with hyperreactio luteinalis and virilization. J Clin Endocrinol Metab 40(5):816–823
Malinowski AK, Sen J, Sermer M (2015) Hyperreactio luteinalis: maternal and fetal effects. J Obstet Gynaecol Can 37(8):715–723
Rosner W et al (2007) Position statement: utility, limitations, and pitfalls in measuring testosterone: an Endocrine Society position statement. J Clin Endocrinol Metab 92(2):405–413
Bradshaw KD et al (1986) Endocrine studies in a pregnancy complicated by ovarian theca lutein cysts and hyperreactio luteinalis. Obstet Gynecol 67(3 Suppl):66S–69S
Ito M et al (1987) Theca lutein cysts with maternal virilization and elevated serum testosterone in pregnancy. Acta Obstet Gynecol Scand 66(6):565–566
Muechler EK, Fichter J, Zongrone J (1987) Human chorionic gonadotropin, estriol, and testosterone changes in two pregnancies with hyperreactio luteinalis. Am J Obstet Gynecol 157(5):1126–1128
Okadome M et al (1989) Hyperreactio luteinalis in normal singleton pregnancy. Int J Gynaecol Obstet 29(4):365–371
Bakri YN, Bakhashwain M, Hugosson C (1994) Massive theca-lutein cysts, virilization, and hypothyroidism associated with normal pregnancy. Acta Obstet Gynecol Scand 73(2):153–155
Reubinoff BE et al (1994) Hyperreactio luteinalis associated with non-immune hydrops fetalis—the role of pituitary hormones. Eur J Obstet Gynecol Reprod Biol 53(2):144–146
Schnorr JA Jr et al (1996) Hyperreactio luteinalis associated with pregnancy: a case report and review of the literature. Am J Perinatol 13(2):95–97
Foulk RA et al (1997) Hyperreactio luteinalis differentiated from severe ovarian hyperstimulation syndrome in a spontaneously conceived pregnancy. Am J Obstet Gynecol 176(6):1300–1302 (discussion 1302-4)
Csapo Z et al (1999) Hyperreactio luteinalis in a normal singleton pregnancy. A case report. J Reprod Med 44(1):53–56
Check JH, Choe JK, Nazari A (2000) Hyperreactio luteinalis despite the absence of a corpus luteum and suppressed serum follicle stimulating concentrations in a triplet pregnancy. Hum Reprod 15(5):1043–1045
Tanaka Y et al (2001) Naturally conceived twin pregnancy with hyperreactio luteinalis, causing hyperandrogenism and maternal virilization. Acta Obstet Gynecol Scand 80(3):277–278
Hoover KL, Barbalinardo LH, Platia MP (2002) Delayed lactogenesis II secondary to gestational ovarian theca lutein cysts in two normal singleton pregnancies. J Hum Lact 18(3):264–268
Gherman RB et al (2003) Intractable hyperemesis gravidarum, transient hyperthyroidism and intrauterine growth restriction associated with hyperreactio luteinalis. A case report. J Reprod Med 48(7):553–556
Betzold CM, Hoover KL, Snyder CL (2004) Delayed lactogenesis II: a comparison of four cases. J Midwifery Womens Health 49(2):132–137
Haimov-Kochman R et al (2004) Spontaneous ovarian hyperstimulation syndrome and hyperreactio luteinalis are entities in continuum. Ultrasound Obstet Gynecol 24(6):675–678
Saisto T et al (2004) Clinical cure of severe, early onset preeclampsia with low molecular weight heparin therapy in primigravida with hyperreactio luteinalis and thrombophilia. Hum Reprod 19(3):725–728
Sherer DM et al (2006) Hyperreactio luteinalis presenting with marked hyperglycemia and bilateral multicystic adnexal masses at 21 weeks gestation. Am J Perinatol 23(2):85–88
Angioni S et al (2007) Hirsutism and hyperandrogenism associated with hyperreactio luteinalis in a singleton pregnancy: a case report. Gynecol Endocrinol 23(5):248–251
Morken NH, Friberg-Otterstad UM, Kahn JA (2007) Hyperreactio luteinalis, presented as an acute abdomen. Acta Obstet Gynecol Scand 86(1):104–106
Binder H et al (2008) Association of FSH receptor and CYP19A1 gene variations with sterility and ovarian hyperstimulation syndrome. Reproduction 135(1):107–116
Onodera N et al (2008) A case of recurrent hyperreactio luteinalis. Am J Obstet Gynecol 198(5):e9–e10
Page SM, O’Brien J (2009) Hyperreactio luteinalis with clitoromegaly in a twin pregnancy. Rev Obstet Gynecol 2(1):3–4
Abe T et al (2011) Conservative management of hyperreactio luteinalis: a case report. J Nippon Med Sch 78(4):241–245
Conway F, Jarvis S, Thornton M (2012) Virilisation during pregnancy in a patient with metastatic colorectal cancer. Case Rep Surg 2012:704039
Kalelioglu IH, Karamustafaoglu B, Has R (2012) A case of hyperreactio luteinalis associated with fetal cystic hygroma. J Obstet Gynaecol 32(3):303–304
Simsek Y et al (2012) Severe preeclampsia and fetal virilization in a spontaneous singleton pregnancy complicated by hyperreactio luteinalis. Eur Rev Med Pharmacol Sci 16(1):118–121
Das G et al (2013) Late presentation of hyperandrogenism in pregnancy: clinical features and differential diagnosis. Endocrinol Diabetes Metab Case Rep 2013:130048
Gonzalez Aguilera B et al (2015) Hyperreactio luteinalis, a rare cause of hyperthyroidism in pregnancy. Endocrinol Nutr 62(3):146–147
Sakae C et al (2015) Ultrasound-guided percutaneous aspiration of hyperreactio luteinalis avoids laparoscopic untwisting of ovarian torsion. Ultrasound Obstet Gynecol 46(2):243–246
Ibanez L et al (2003) Androgen receptor gene CAG repeat polymorphism in the development of ovarian hyperandrogenism. J Clin Endocrinol Metab 88(7):3333–3338
Weintrob N et al (2018) The effect of CAG repeats length on differences in hirsutism among healthy Israeli women of different ethnicities. PLoS ONE 13(3):e0195046
This research did not receive and funding. Open Access funding provided by Projekt DEAL.
Conflict of interest
RDP received honoraria from Siemens Healthineers and Roche Diagnostics.
The University Bonn Institutional Review Board does not require formal approval for case reports. Written consent to participate was obtained from the patient.
Written consent for publication was obtained from the patient.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Condic, M., Merz, W.M., Gembruch, U. et al. Testosterone serum levels are not predictive of maternal virilization in hyperreactio luteinalis. Arch Gynecol Obstet 303, 129–134 (2021). https://doi.org/10.1007/s00404-020-05745-6
- Hyperreactio luteinalis
- Maternal virilization
- Pregnancy-associated virilization
- Liquid chromatography–tandem mass spectrometry
- Electrochemiluminescence immunoassay