Imaging features, clinical characteristics and neonatal outcomes of pregnancy luteoma: A case series and literature review

Abstract

Introduction

This study aimed to investigate the imaging features, clinical characteristics and neonatal outcomes of pregnancy luteoma.

Material and methods

We retrospectively analyzed patients with pregnancy luteoma admitted to the First Affiliated Hospital of Sun Yat‐sen University between January 2003 and December 2022. We recorded their imaging features, clinical characteristics and neonatal outcomes. Additionally, we reviewed relevant studies in the field.

Results

In total, 127 cases were identified, including eight from our hospital and 119 from the literature. Most patients (93/127, 73.23%) were of reproductive age, 20–40 years old, and 66% were parous. Maternal hirsutism or virilization (such as deepening voice, acne, facial hair growth and clitoromegaly) was observed in 29.92% (38/127), whereas 59.06% of patients (75/127) were asymptomatic. Abdominal pain was reported in 13 patients due to compression, torsion or combined ectopic pregnancy. The pregnancy luteomas, primarily discovered during the third trimester (79/106, 74.53%), varied in size ranging from 10 mm to 20 cm in diameter. Seventy‐five cases were incidentally detected during cesarean section or postpartum tubal ligation, and 39 were identified through imaging or physical examination during pregnancy. Approximately 26.61% of patients had bilateral lesions. The majority of pregnancy luteomas were solid and well‐defined (94/107, 87.85%), with 43.06% (31/72) displaying multiple solid and well‐circumscribed nodules. Elevated serum androgen levels (reaching values between 1.24 and 1529 times greater than normal values for term gestation) were observed in patients with hirsutism or virilization, with a larger lesion diameter (P < 0.001) and a higher prevalence of bilateral lesions (P < 0.001). Among the female infants born to masculinized mothers, 68.18% (15/22) were virilized. Information of imaging features was complete in 22 cases. Ultrasonography revealed well‐demarcated hypoechoic solid masses with rich blood supply in 12 of 19 cases (63.16%). Nine patients underwent magnetic resonance imaging (MRI) or computed tomography (CT), and six exhibited solid masses, including three with multi‐nodular solid masses.

Conclusions

Pregnancy luteomas mainly manifest as well‐defined, hypoechoic and hypervascular solid masses. MRI and CT are superior to ultrasonography in displaying the imaging features of multiple nodules. Maternal masculinization and solid masses with multiple nodules on imaging may help diagnose this rare disease.

Distinguishing pregnancy luteoma from malignant ovarian tumors may be challenging. However, features such as maternal masculinization and the presence of solid masses with multiple nodules on imaging may provide valuable diagnostic clues.

Abbreviations

ADNEX

assessment of different neoplasias in the adnexa

AFP

alpha‐fetoprotein

CA‐125

carcinoma antigen‐125

CT

computed tomography

MRI

magnetic resonance imaging

O‐RADS

ovarian‐adnexal reporting and data system

PL

pregnancy luteoma

SR

Simple Rules

SR‐Risk

Simple Rules Risk

1. INTRODUCTION

Pregnancy luteoma (PL) is a rare tumor‐like lesion of the ovary that occurs during pregnancy and regresses spontaneously after delivery, with or without virilization in pregnant women and fetuses. Fewer than 200 cases have been reported in the literature. ¹ , ² , ³ , ⁴ , ⁵ , ⁶ , ⁷ , ⁸ , ⁹ , ¹⁰ , ¹¹ , ¹² , ¹³ , ¹⁴ , ¹⁵ , ¹⁶ , ¹⁷ , ¹⁸ , ¹⁹ , ²⁰ , ²¹ , ²² , ²³ , ²⁴ , ²⁵ Previous studies have indicated that distinguishing PL from other types of ovarian tumors, such as malignancies, may be difficult, especially when accompanied by massive ascites and an elevated CA125 level. ²⁶ , ²⁷ , ²⁸ Hence, a good appreciation of the clinical and imaging features of PL can avoid unnecessary exploratory laparotomies or termination of pregnancy. Current evidence on sonographic appearance is predominantly limited to case reports. ²⁹ , ³⁰ , ³¹ , ³² , ³³ , ³⁴ , ³⁵ , ³⁶ , ³⁷ , ³⁸ , ³⁹ , ⁴⁰ , ⁴¹ , ⁴² Herein, we summarize the clinical characteristics and neonatal outcomes of patients with PLs to improve our understanding of the natural history of the disease and help clinicians manage patients. We also conducted a literature review and explored the existing literature on the imaging findings associated with this condition.

2. MATERIAL AND METHODS

2.1. Study scheme

The clinical data of eight patients with PLs who were admitted to the First Affiliated Hospital of Sun Yat‐sen University between January 2003 and December 2022 were retrospectively analyzed. During this period, 64 580 gravidas were admitted to our hospital for delivery. All pregnant women underwent routine obstetric examinations and ultrasound screening (including observation of bilateral adnexa) during the first, second and third trimesters. All cases of PL were confirmed by pathologic examination. Table 1 summarizes the clinical characteristics and neonatal outcomes of the patients.

TABLE 1.

Cases with pregnancy luteoma in the First Affiliated Hospital of Sun Yat‐sen University.

Case No.	Age, years	Clinical manifestations	Maternal or female fetal masculinization	Mode of delivery	Cesarean indications	Newborns gender/weight (kg)	Time and method of lesion detection	Location of the lesions	Lesion size (mm)	Tumor characteristic	Tumor color score	Ascites	Cut surface
1	30	Abdominal pain	Yes	Vaginal delivery	/	Female/2.27	36+1 weeks sonography	Bilateral	92; 97	Solid	Abundant	Yes	Brownish‐yellow, soft
2	27	Proteinuria, edema	No	Cesarean section	Adnexal mass	Female/2.90	37+1 weeks sonography	Right	60	Solid	Moderate	No	Grayish‐yellow, soft
3	20	Abdominal pain, vaginal bleeding	No	Cesarean section	Ectopic pregnancy	/	5+3 weeks sonography	Left	73	Unilocular (Anechoic)	Minimal	No	Smooth inner and outer walls; clear yellow cyst fluid
4	25	Asymptomatic	No	Cesarean section	Complete placenta previa	Female/3.32	12+4 weeks sonography	Left	71	Unilocular (Ground glass)	Minimal	No	Smooth outer walls, chocolate‐like liquid, and many small papillary projections could be seen inside.
5	37	Asymptomatic	No	Cesarean section	Scarred uterus	Male/3.45	38+5 weeks cesarean section	Right	10	Solid	/	No	Medium texture; information about the cut surface of the mass was unknown
6	29	Asymptomatic	No	Cesarean section	Cephalopelvic disproportion	Female/3.65	40 weeks cesarean section	Right	40	Solid	/	No	Brownish‐yellow, soft
7	26	Asymptomatic	No	Cesarean section	Fetal distress	Male/3.30	38+6 weeks cesarean section	Left	15	Solid	/	No	Brownish‐yellow, soft
8	29	Asymptomatic	No	Cesarean section	Fetal distress	Female/3.05	38+6 weeks cesarean section	Left	25	Solid	/	No	Hard texture; information about the cut surface of the mass was unknown

All original ultrasound images and video clips of cases 1–4 were collected at a central reading site, where they were independently reviewed by another experienced ultrasound examiner who was blinded to the medical information and pathological results of patients. The reviewer independently categorized the ultrasound images of each mass according to the ovarian‐adnexal reporting and data system (O‐RADS) published by the American College of Radiology (ACR), ⁴³ IOTA Simple Rules (SR), Simple Rules Risk (SR‐Risk) assessment, and assessment of different neoplasias in the adnexa (ADNEX) model without CA125. We calculated the risk of malignancy using the IOTA SR‐Risk and ADNEX online algorithms (http://www.iotagroup.org; Table S1).

A literature search of the PubMed, Web of Science and Embase databases was conducted, using the Medical Subject Headings terms “pregnancy luteoma” and “luteoma of pregnancy”. Only the articles published in English between January 1965 and December 2022 were included. To determine study inclusion, two authors screened all titles and abstracts, and reviewed relevant articles. Disagreements were resolved either by consensus or by consultation with a third reviewer. Data on clinical characteristics, imaging features and neonatal outcomes were extracted from our hospital and published case reports.

2.2. Statistical analyses

The chi‐square test was used to analyze the relationship between maternal virilization and clinicopathologic characteristics. The clinical and sonographic parameters of the PLs were summarized using absolute frequencies (%) as categorical variables and medians (ranges) as continuous variables. Statistical analyses were performed using SPSS (version 22.0; SPSS). A two‐sided value of P < 0.05 was considered statistically significant.

2.3. Ethics statement

This study was approved by the Ethics Committee of the First Affiliated Hospital of Sun Yat‐sen University on March 21, 2023 (Approval Number: [2022]064).

3. RESULTS

3.1. Case series

A total of eight patients were enrolled. The median age of patients was 28 (range 20–37) years. Four patients were primiparous. Two patients had gestational diabetes mellitus and maintained reasonable glycemic control without pharmacological intervention. Symptoms that led to the discovery of the masses included abdominal pain in two patients (one after ectopic pregnancy and associated with abnormal vaginal bleeding), proteinuria/edema in one preeclampsia patient, and incidental findings in five others. Case 1 is the only patient who complained of masculine changes (clitoral hypertrophy at 24 weeks of gestation). Unfortunately, hormonal testing was not performed on any patient. Only the preoperative serum levels of alpha‐fetoprotein (AFP) and carcinoma antigen (CA)‐125 were recorded, significantly elevated in Case 1 (AFP, 235.09 μg/L; CA‐1252356.8 U/mL).

Adnexal masses were detected on sonography in four cases. Cases 1 and 2 were detected during the third trimester, and ultrasonography showed adnexal heterogeneous hypoechoic solid masses with clear boundaries and abundant (Figure 1A,B) or moderate (Figure 2A,B) color Doppler signals, both of which were misdiagnosed as ovarian tumors. Contrast‐enhanced computed tomography (CT) showed a solid mass with multiple nodules (Figure 1C), and massive hydrothorax and ascites were detected in Case 1. Cases 3 and 4 were detected during early pregnancy and ultrasonography revealed a unilocular cystic mass with a color score of 1 or 2 (Figure 2C,D). The cyst content was anechoic in one case and ground‐glass opacity (because of hemorrhage) in the other. We calculated the risk of malignancy in PL (Cases 1–4) using four ultrasound‐based diagnostic models, of which approximately half were diagnosed with malignant tumors, especially those with solid masses (Table S1).

FIGURE 1.

A 30‐year‐old woman presented with pregnancy luteoma at 36+1 weeks of gestation (Case 1). (A) Abdominal transverse images show a 97 × 75 mm heterogeneous hypoechoic solid mass with clear boundaries (right side). (B) Color Doppler image demonstrated prominent vascularity within the mass. (C) Contrast‐enhanced CT showed a solid mass with multiple nodules and massive ascites. (D) The nodular appearance of the enlarged ovary. (E) The cut surface showed multiple brownish‐yellow solid nodules with a soft texture. (F) External genitalia of the female infant 23 months after delivery. CT, computed tomography.

FIGURE 2.

The gray‐scale and Doppler ultrasound images of pregnancy luteoma. (A,B) Transabdominal image identified a well‐circumscribed 60 × 55 mm heterogeneous hypoechoic solid mass with moderate signal intensity on color Doppler (Case 2). Ultrasonography showed a unilocular cystic mass (Cases 3 and 4). Cyst content was anechoic with a color score of 2 in one case (C), and ground glass opacity (because of hemorrhage) with a color score of 1 in the other (D).

All the patients underwent surgery: seven patients underwent lesion resection, and Case 1 underwent exploratory laparotomy and intraoperative frozen biopsy of bilateral ovaries. The maximum diameters of the masses were significantly different, ranging from 10 mm to 97 mm. Overall, seven cases were unilateral, and one had bilateral ovarian involvement. Six patients presented with solid masses with clear borders. Moreover, in these cases, the cut surface showed single or multiple brownish‐yellow/reddish‐brown solid nodules (Figure 1D,E), four of which had a soft texture. The remaining two cases were uncommon with regard to the apparent cystic appearance of the lesion.

The gestational age at delivery was 37–40 weeks. The seven newborns included five girls (5/7, 71.4%) and two boys (2/7, 28.6%). The Apgar scores were to 10–10 after 1–5 minutes. The newborns weighed between 2.27 and 3.65 kg. Physical examinations of the four female neonates were normal, and virilization was detected in one girl (clitoral hypertrophy in Case 1). We conducted the telephone follow‐up in December 2022, and all mothers recovered normal menstruation 1–4 months after stopping breastfeeding or undergoing surgical treatment for ectopic pregnancy. However, whether the surgery impairs ovarian function has not yet been explicitly evaluated. In Case 1, the serum levels of tumor markers (CA125 and AFP) gradually decreased to normal 1 month after delivery. Two months postpartum, ultrasonography revealed normalized bilateral ovaries. However, the clitoris was still enlarged at 23 months postpartum in both mother and daughter (Figure 1F).

3.2. Literature review

A total of 46 articles were retrieved from the literature, which reported 119 PL cases. ¹ , ² , ³ , ⁴ , ⁵ , ⁶ , ⁷ , ⁸ , ⁹ , ¹⁰ , ¹¹ , ¹² , 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 , ⁴² , ⁴⁴ , ⁴⁵ In comparison, 19 articles were excluded for the following reasons: (1) did not have an English version (n = 7); (2) the full text was not available (n = 11); and (3) duplicates were removed (n = 1). Therefore, 127 cases (including eight cases from our hospital) of PLs were identified and summarized (Figure 3; Table 2).

FIGURE 3.

Flow chart summarizing inclusion in the systematic review of studies on pregnancy luteomas. CT, computed tomography; MRI, magnetic resonance imaging; PL, pregnancy luteoma.

TABLE 2.

Clinicopathologic characteristics and neonatal outcomes of pregnancy luteomas.

Variable	n	%
Age at diagnosis, (years)
≦20	10	7.87
>20 and ≤30	49	38.58
>30 and ≤40	44	34.65
>40 and ≤50	1	0.79
Unknown	23	18.11
Reproductive history
Primigravida	34	26.77
Multipara	66	51.97
Unknown	27	21.26
Clinical manifestations
Asymptomatic	75	59.06
Maternal hirsutism or virilization	38	29.92
Abdominal pain	13	10.24
Other	7	5.51
Maternal hirsutism or virilization
Yes	38	29.92
No	83	65.36
Unknown	6	4.72
Newborn gender
Early abortion	2	1.56
Ectopic pregnancy	3	2.34
Male	29	22.66
Female	51 a	39.84
Unknown	43	33.60
Female fetal virilization
Yes	19	37.25
No	32	62.75
Time of discovery mass
First trimester	8	6.30
Second trimester	9	7.09
Third trimester	79	62.20
After delivery	10	7.87
Unknown	21	16.54
Method of discovery mass
Imaging or physical examination	39	30.71
Incidental finding	75	59.06
Unknown	13	10.23
Tumor size (mm)
≤50	40	31.50
>50 and ≤100	38	29.92
>100 and ≤150	12	9.44
>150	1	0.79
Unknown	36	28.35
Location of the mass
Unilateral	80	62.99
Bilateral	29	22.84
Unknown	18	14.17
Mass complexity
Solid mass	94	74.02
Complex mass	6	4.72
Cystic mass	7	5.51
Unknown	20	15.75
Solid mass with multiple nodules
Yes	31	32.98
No	41	43.62
Unknown	22	23.40
Multilocular cystic mass
Yes	4	57.14
No	3	42.86

^a One case was a dichorionic diamniotic twin pregnancy.

3.2.1. Clinical characteristics

The age at diagnosis mostly ranged between 20 and 40 years (93/127, 73.23%). Two‐thirds of patients (66/100, 66%) were parous. Most patients (75/127, 59.06%) were asymptomatic. In 29.92% (38/127) of cases, luteomas were hormonally active, with the secretion of androgens, which can result in maternal hirsutism and virilization (such as deepening voice, acne, facial hair growth and clitoromegaly). In 13 cases, PLs caused abdominal pain due to symptoms associated with compression, torsions or combined ectopic pregnancy. Among the 127 patients, four had hydrothorax and/or ascites, and three had elevated serum CA125 levels. The PLs, primarily discovered during the third trimester (79/106, 74.53%), varied in size from 10 mm to 20 cm in diameter. Seventy‐five cases were incidentally detected during cesarean section or postpartum tubal ligation, and 39 were found on imaging or physical examination during pregnancy. Approximately one‐quarter of the patients (29/109, 26.61%) had bilateral lesions. The PLs were mainly solid and well‐defined (94/107, 87.85%), and 43.06% (31/72) showed multiple solid and well‐circumscribed nodules.

Serum levels of testosterone and other androgens are often markedly elevated (reaching values between 1.24 and 1529 times greater than normal values for term gestation) in patients with hirsutism and virilization. Virilization of the female fetus was found in 68.18% of patients (15/22) with maternal hirsutism, which resulted in clitoral enlargement and ambiguous genitalia (Table S2). Table 3 presents the correlation between maternal virilization and the clinicopathologic features of PLs. Compared with the non‐maternally masculinized group, patients with maternal masculinization had a larger maximum lesion diameter (P < 0.001) and a higher prevalence of bilateral lesions (P < 0.001). However, the two groups showed no significant differences in age at diagnosis, reproductive history, time of mass discovery or mass complexity.

TABLE 3.

Correlation between maternal virilization and clinicopathologic features of pregnancy luteoma.

Variable	Maternal virilization		P‐value
	No (n)	Yes (n)
Age at diagnosis, (years)
≤30	28	18	0.361
>30	28	11
Reproductive history
Primigravida	17	14	0.087
Multipara	36	12
Time of discovery mass
First and second trimesters	11	6	1.000
Third trimester and after delivery	31	20
Method of discovery mass
Imaging or physical examination	12	17	0.005
Accidental discovery	30	9
Tumor size (mm)
≤50	31	3 a	<0.001
>50	23	24
Location of the mass
Unilateral	50	14	<0.001
Bilateral	6	16
Mass complexity
Solid	47	21	0.207
Other types	6	6

^a Fisher's exact test.

3.2.2. Imaging features

Information of imaging features was complete in 22 cases of PLs (Table 4). Ultrasonography revealed bilateral or unilateral well‐demarcated hypoechoic solid masses with rich blood supply in 12 cases (12/19, 63.16%), complex masses in four cases, and cystic masses in three cases. Nine patients underwent magnetic resonance imaging (MRI) or computed tomography (CT) examination, and six had solid masses, including three patients with multinodular solid masses. The other two cases had multiple cystic masses.

TABLE 4.

Review of the literature on imaging findings of pregnancy luteoma.

First author [ref.], year	Number of cases	Maternal or female fetal masculinization	Location of the lesions	Lesion size (mm)	Time of finding mass	Sonographic findings	Color score	CT or MRI findings
Rodriguez [29], 1999	1	1	Right	125	Postpartum	Complex right adnexal mass with focal areas of increased chogenicity. Ascites and a small right pleural effusion	Increased venous vascularity	Unknown
Choi [30], 2000	Case 1	Yes	Right	57	28 weeks	Heterogeneous solid mass	Increased vascularity	Unknown
Choi [30], 2000	Case‐2	No	Left	55	21 weeks	Complex mass	Color score 4	Unknown
Warmann [31], 2000	1	Yes	Bilateral	Unknown	Postpartum	Enlargement of both ovaries with cystic alterations	Unknown	Unknown
Mazza [32], 2002	1	Yes	Bilateral	80; 80	5 weeks	An enlarged ovary containing small hypoechoic and hyperechoic areas	Unknown	Unknown
Wang [33], 2003	1	No	Bilateral	120; 100	27 weeks	Unknown	Unknown	Multilocular cystic masses with thickened internal septa
Kao [34], 2005	1	Yes	Bilateral	80; 74	35 weeks	Heterogeneous echogenicity mass	Hypervascularity	Multinodular mass; T1‐weighted images: high signal intensity; T2‐weighted images: low signal intensity
Wang [35], 2005	1	Yes	Bilateral	90; 70	Unknown	Unknown	Unknown	Solid ovarian mass with multiple nodules
Spitzer [36], 2007	1	Yes	Right	86	6.5 weeks	Apparent fibroid	Unknown	A heterogeneous, predominantly solid mass
Tannus [27], 2009	1	No	Right	83	22 weeks	Hypoechoic predominantly solid mass	Color score 4; low‐resistance arterial flow	T1‐weighted images: heterogeneous low signal intensity; T2‐weighted images: predominantly high signal intensity
Masarie [37], 2010	1	Yes	Right	115	33 weeks	Solid ovarian mass	Unknown	Solid adnexal mass without associated hemorrhage or ascites
Dasari [38], 2013	1	Yes	Bilateral	60; 100	8 weeks	Enlarged and homogeneously isoechoic to hypoechoic ovaries; moderate ascites	Unknown	Unknown
Wadzinski [39], 2014	1	Yes	Bilateral	58; 58	15 days after birth	Unknown	Unknown	Each ovary contains numerous small follicles (2–3 mm) and no dominant ovarian mass or cyst
Wang [26], 2015	1	No	Right	60	15 weeks	Solid tumor; massive ascites	Color score 4	Unknown
Rapisarda [40], 2016	1	Yes	Right	120	28 weeks	Adnexal mass with necrotic areas	Color score 2	Unknown
Khurana [41], 2017	1	No	Left	81	19 weeks	Heterogenous solid component and a 4‐cm cystic component	Internal vascularity and arterial blood flow	Heterogenous mass with both cystic and soft tissue components
Brar [42], 2017	1	No	Right	20	8 weeks	Predominantly solid right ovarian mass containing multiple small cystic areas and a small solid focus	Unknown	Unknown
Dayal [28], 2022	1	No	Right	30	First trimester	Adnexal solid mass	Color score 4	Unknown
Shang, this study, 2023	Case 1	Yes	Bilateral	97; 97	36+1 weeks	Adnexal heterogeneous hypoechoic solid masses with clear boundaries	Color score 4	Solid mass with multiple nodules; massive hydrothorax and ascites
Shang, this study, 2023	Case‐2	No	Right	60	37+1 weeks	Heterogeneous hypoechoic solid masses with clear boundaries	Color score 3	Unknown
Shang, this study, 2023	Case‐3	No	Left	73	5+3 weeks	Unilocular cystic mass; cyst content was anechoic	Color score 2	Unknown
Shang, this study, 2023	Case‐4	No	Left	71	12+4 weeks	Unilocular cystic mass; cyst content was ground glass	Color score 1	Unknown

Abbreviations: CT, computed tomography; MRI, magnetic resonance imaging.

3.2.3. Follow‐up

Serum testosterone levels decreased rapidly after delivery and returned to normal within 2 weeks postpartum. Maternal and infant virilizing signs (hirsutism and acne) had regressed 6 months postpartum, except for the patient's deepened voice and clitoromegaly. PLs typically undergo spontaneous postpartum regression, usually within 3 months of delivery.

4. DISCUSSION

The clinical and imaging characteristics of 127 patients with PL (including eight cases from our hospital and 119 published case reports in the literature) were retrospectively analyzed. The incidence of PLs in our hospital was 8/64 580. However, the actual incidence of PLs has not been fully elucidated. The clinical characteristics of our patients were generally similar to those reported in the literature. ² , ²⁰ They were usually asymptomatic and were incidentally detected during cesarean section or postpartum tubal ligation. PLs varied in size, and three‐quarters of the lesions were discovered in the late stages of pregnancy. Most PLs were solid lesions with clear borders, and 43.06% appeared as multiple, solid well‐circumscribed nodules. According to a 1993 literature report, ¹² one‐third of patients with PLs had bilateral masses. Subsequently, several studies have reported that PLs almost always occurred unilaterally. ²³ , ²⁵ Our review found that lesions were bilateral in one‐quarter of the cases.

Thirty percent of the lesions were hormonally active, resulting in maternal or fetal hirsutism and virilization. Sixty‐eight percent of female infants born to masculinized mothers were also virilized, consistent with the literature. ² , ⁶ , ⁹ , ¹⁰ , ¹¹ , ²⁰ , ²⁵ Serum levels of testosterone and other androgens were often markedly elevated (reaching values between 1.24 and 1529 times greater than normal values for term gestation) in patients with hirsutism and virilization. ¹² During a normal pregnancy, the maternal circulating testosterone may rise up to seven times the non‐pregnant levels, which does not cause virilization. Tissue analysis of PLs revealed similar quantities of androgen hormones in cases with and without hirsutism. ³ Generally, masculine changes occur in cases with a larger maximum diameter of bilateral lesions which can increase serum testosterone levels. Therefore, we speculated that masculinization may be related to lesion size and bilateral lesions. Maternal masculinization may be a valuable clinical sign to aid in diagnosis, although it can also be caused by sex cord tumors such as Sertoli–Leydig cell tumors.

Most PL masses are solid with clear borders, brownish‐yellow or reddish‐brown cut surfaces, and a soft texture. ¹² PL must be distinguished from other solid ovarian neoplasms, such as granulosa cell tumors, steroid cell tumors, germ cell tumors, Krukenberg tumors and Sertoli–Leydig cell tumors, to avoid oophorectomy in pregnant women, especially when accompanied by massive ascites and an elevated CA125 levels. ²⁶ , ²⁷ , ²⁸ , ⁴⁶ , ⁴⁷ The gross morphology and histopathologic features of PL can provide clues for a differential diagnosis. PLs originate from luteinized cells in a “follicle‐like” structure, particularly theca‐lutein cells. The proliferation of the theca‐lutein cells can obliterate the follicular cavity and form solid nodules. ¹² , ²⁵ To date, the imaging features of 22 cases (including four cases from our hospital) have been reported in the literature. ²⁹ , ³⁰ , ³¹ , ³² , ³³ , ³⁴ , ³⁵ , ³⁶ , ³⁷ , ³⁸ , ³⁹ , ⁴⁰ , ⁴¹ , ⁴² The ultrasound findings were mostly bilateral or unilateral hypoechoic solid masses with clear borders and rich blood supply (63.16%), ²⁶ , ²⁷ , ²⁸ , ³⁰ , ³⁴ which correlated well with the imaging findings in Cases 1 and 2 in our hospital. Recently, the ESGO/ISUOG/IOTA/ESGE demonstrated that the IOTA ADNEX and SR‐Risk models were the best ultrasound‐based diagnostic models for the characterization of ovarian masses. ⁴⁸ However, these models revealed some difficulties in differentiating PLs from solid masses using the suggested thresholds in our study.

MRI and CT showed that approximately half of the solid lesions had multiple nodules. MRI and CT are superior to ultrasonography in displaying the imaging features of multiple nodules. In addition, some PL lesions can have multilocular cystic masses with thickened internal septa or a complex cystic appearance due to regions of hemorrhage. ²⁹ , ³⁰ , ³³ , ³⁹ , ⁴⁰ Interestingly, ultrasonography demonstrated unilocular cystic masses with well‐defined margins in our Cases 3 and 4. Based on the current described cases, PLs cannot be differentiated from other ovarian tumors using ultrasonography. Once PL is suspected, the patient should undergo detailed MRI to visualize the multiple nodules of the lesions, and hormonal assays for testosterone and its derivatives.

The main strengths of our study are that it is the first to summarize imaging features of PLs and compare the clinicopathologic features of PLs between patients with and without maternal masculinization in the same cohort. Our study has several limitations. First, the ultrasound information was retrospectively extracted from ultrasound reports and images, and therefore our conclusions on ultrasound features of PLs may be biased and should be interpreted with caution. Secondly, complete information on serum levels of testosterone and other androgens was only available in some cases.

5. CONCLUSION

PLs primarily manifested as bilateral or unilateral, well‐defined, hypoechoic and hypervascular solid masses. MRI and CT are superior to ultrasonography in displaying the imaging features of multiple nodules. Maternal masculinization and the presence of solid masses with multiple nodules on imaging may help in the diagnosis of this rare disease.

AUTHOR CONTRIBUTIONS

JS and HX conceived and designed the study; JS and QZ screened and selected the articles. JS, CH, JF and KH analyzed the data. JS drafted the manuscript. QZ and HX revised the paper for important content. All authors read and approved the paper.

FUNDING INFORMATION

This study was supported by research grants from the National Scientific Foundation Committee of China (82171938, 81801705 and 82202156).

CONFLICT OF INTEREST STATEMENT

The authors have stated explicitly that there are no conflicts of interest in connection with this article.

Supporting information

Table S1

Table S2

Shang J‐H, Huang C‐X, Zheng Q, Feng J‐L, He K, Xie H‐N. Imaging features, clinical characteristics and neonatal outcomes of pregnancy luteoma: A case series and literature review. Acta Obstet Gynecol Scand. 2024;103:740‐750. doi: 10.1111/aogs.14672