Naturally conceived twin pregnancy with hyperreactio luteinalis, causing hyperandrogenism and maternal virilization

Abstract

Hyperreactio luteinalis (HL) is a marked cystic enlargement of the ovary due to multiple benign theca lutein cysts (1). It is an uncommon condition, and the cause of this disorder is not well known. HL is believed to be induced by elevated serum levels of human chorionic gonadotropins (hCG) (1). It occurs commonly in pregnant patients with trophoblastic disease, occasionally in multiple pregnancies, and rarely in normal pregnancy (2). Maternal virilization occurs in approximately 15% of cases with HL (3). We herein report a case of HL with maternal virilization in spontaneously conceived twin pregnancy. We also report two- and three-dimensional ultrasonographic and MRI features of HL, and discuss its management. A 25-year old Japanese woman, gravida 1, para 0, was referred to our highrisk pregnancy clinic at 13 week’s gestation because of twin pregnancy with bilateral multicystic ovarian enlargements. Three-dimensional ultrasonography showed an interrelationship of two fetuses in utero. Bilateral enlarged multicystic ovaries (left, 13×14×7 cm; right, 13×14×7 cm) without solid components were evident. There were no ascites and pleural effusion. MRI depicted same findings of both ovaries as those by ultrasonography (Fig. 1A). A. T2-weighted MR image of both ovaries. B. Macroscopic finding of both ovaries at exploratory laparotomy. At 18 weeks of gestation, she noticed facial acne and hoarseness. At that time, markedly enlarged clitoris was noted. The serum testosterone concentration (24.4 ng/ml) was significantly high. The size of both ovaries was unchanged during pregnancy. At 34 weeks of gestation, the patient was admitted to our hospital because of irritable uterine contractions. At 38 weeks of gestation, a 2602 g male baby with an Apgar score of 8 at 1 min and 9 at 5 min and a 2460 g male baby with an Apgar score of 6 at 1 min and 7 at 5 min were born vaginally. There were no abnormal findings in either neonate. After delivery, serum testosterone levels still remained high. At 3 weeks later, exploratory laparotomy was performed because of unchanged size of both ovaries and persistent high testosterone levels. Bilateral ovaries were multicystic and markedly enlarged (Fig. 1B). Partial resection of both ovaries and puncture of cyst were performed. The pathological diagnosis was HL with no malignancy. Her postoperative course has been uneventful. At discharge, the serum testosterone level (0.5 mg/ml) was within normal range. Three months later, virilization disappeared, and the both ovarian sizes became normal. Both babies were also doing well without any changes, although the karyotype of these two boys was not checked. HL is found in 10–22% of trophoblastic diseases (4). It has also been associated with fetal hydrops secondary to erythroblastosis fetalis, twin pregnancy, or a small number of apparently normal pregnancies (5). In 15% of HL, maternal virilization occurs (3). Luteoma of pregnancy and HL associated with theca lutein cysts are two most common benign causes of maternal virilization in pregnancy. Our case is a naturally conceived twin pregnancy, and maternal virilization was clearlly evident. HL mimics a neoplasm in clinical practice, and it sometimes leads to unnecessary operation (3). To avoid an unnecessary operation, especially oophorectomy, making a definite diagnosis is very important. Most cases with HL regress in size after delivery (1). Therefore, conservative management is the best way for HL. Surgical intervention is sometimes performed only to remove infarcted tissue, control hemorrhage or reduce the ovarian size in patients with marked virilization (6). With respect to three-dimensional ultrasonographic findings, smooth internal surface and no solid component in the cyst suggested benign disorders (7). MRI also showed no solid part in the enlarged ovaries. Noninvasive diagnostic modalities, such as two- and three-dimensional ultrasonography, and MRI may be useful to differentiate benign from malignant ovarian tumors during pregnancy. After delivery, we at first selected conservative management for enlarged bilateral ovaries. However, these symptoms did not change, and maternal virilization was still evident. Serum testosterone levels also remained high. Therefore, bilateral partial resection and puncture of cyst were performed 3 weeks later after delivery. Consequently, the serum testosterone level decreased quickly, and both ovarian enlargements decreased within normal range. All symptoms of maternal virilization disappeared, and normal menstruation occurred 3 months later. The pathogenesis of HL is still unknown. It is most commonly associated with pregnancies with excessive hCG secretion. There have been a few cases of HL in normal singleton pregnancies without ovulation induction (4, 8–10). These findings suggest that ovarian sensitivity to hCG might be increased in patients with HL or that another unknown factor may be involved (1). In some ways HL mimics ovarian hyperstimulation syndrome, which is an iatrogenic complication of ovulation induction that is initially seen in a similar fashion, although usually with a greater severity of symptoms (10, 11). In a clinical and pathogenic sense the two are related phenomena. However, ovarian hyperstimulation, because it is iatrogenic, tends to be watched for and identified early, and it is not likely to be confused with other entities (10). On the other hand, ovarian hyperstimulation is more frequent in patients with polycystic ovarian syndrome (PCOS) (12, 13). Muechler et al. (14) reported a case of HL in normal singleton pregnancy with ovulation induction due to polycystic ovarian syndrome (PCO). Ben-Chetrit et al. (15) also reported a case of recurrent maternal virilization during pregnancy in a patient with PCO. In our case, there may be complications with PCO, because of oligomanorrhea and infertility for 5 years before pregnancy. LH stimulates excess androgen production in nonpregnant patients, and hCG stimulates testosterone synthesis in pregnant women (16). PCO might be one of the risk factors for HL. However, the true pathogenesis of HL is currently unknown. Further study is needed to clarify the main pathogenesis of HL during pregnancy.