Introduction

Mirror syndrome (MS) also known as “triple edema” or Ballantyne syndrome, is a rare disorder affecting pregnant women [1]. It describes the unusual association of fetal hydrops and placental edema with diverse maternal edematous manifestations [1,2,3]. As most of the published studies in the literature on MS consist of sporadic case reports that mainly focused on severe presentations, its incidence and pathophysiology are remain unclear [2, 3]. In fact, MS is often underdiagnosed or misdiagnosed due to unnoticed maternal symptoms of an abnormal fetus, or unawareness of MS and sometimes its preeclampsia-like manifestations [4]. Some researchers have hypothesized that MS shares a similar pathogenesis to preeclampsia with the same pattern of increased inflammatory markers or placental factors, which could be used as a diagnostic tool for MS; however, there is insufficient evidence to support this theory [5,6,7,8].

It has been reported that maternal morbidity usually increases in MS, and fetal mortality has been reported to be as high as 67.26% [3]. Most patients require induction of labor or termination of pregnancy because of irreversible fetal hydrops and poor prognosis. However, until now, the characteristics of MS have not been fully elucidated, and the risk factors for MS among hydropic fetuses are still unknown due to paucity of cases. In this study, we retrospectively compared MS cases with non-MS fetal hydrops controls to analyze the clinical characteristics and risk factors of MS.

Methods

Study design and subjects

A total of 59 pregnant women with fetal hydrops were retrospectively reviewed from 58, 428 deliveries at the Third Affiliated Hospital of Sun Yat-Sen University between January 2012 and December 2020. Twenty-two patients were excluded since they terminated their pregnancies in the first trimester and had no maternal complications. The remaining 37 patients who were in the second and third trimesters were divided into the MS and the non-MS groups according to the presence or absence of maternal mirroring symptoms. This study was approved by the Institutional Review Board of the Third Affiliated Hospital of Sun Yat-Sen University and written informed consent was obtained from each participant.

Data collection

Maternal data included age, primiparity, body weight gain (kg), blood pressure (mmHg), hemoglobin (g/L), hematocrit (%), platelets (× 109/L), albumin (g/L), uric acid (μmol/L), creatinine (μmol/L) and urinary protein. Fetal features included the time of onset of fetal hydrops (gestational weeks), site of edema, structural anomalies, volume of amniotic fluid, and placental thickness. Maternal and fetal outcomes included gestational weeks at time of delivery, mode of delivery, complications, transfusion of blood products, intensive care unit (ICU) admission, length and cost of hospital stay, neonatal death or survival, birth weight, placental weight, and pathologic examination.

Diagnostic criteria

The diagnostic criteria were as follows: a) Fetal hydrops was defined as the pathological fluid accumulation in ≥2 different fetal compartments, including skin edema (skin thickness > 5 mm), ascites, pleural effusion, pericardial effusion, and other sonographic findings such as placental thickening or polyhydramnios (atypical fetal hydrops manifests as simple skin edema or body cavity effusion) [9]; b) placental thickening, indicated by ultrasound-detected placental thickness of ≥40 or ≥ 60 mm in the second and third trimesters, respectively [9]; c) placental edema confirmed by pathology of the placenta after delivery; and d) for MS, maternal edematous manifestations characterized by accumulation of excessive fluid within subcutaneous and body cavities, weight gain, and preeclampsia-like symptoms, which were associated with fetal hydrops and placental edema.

Statistical analysis

Data are presented as median (interquartile range) for continuous variables and number (percentage) for categorical variables. Differences in continuous variables were compared between groups using Mann-Whitney U test. Fisher’s exact test was performed to analyze the categorical variables. The risk factors for MS were analyzed using binary logistic regression model. Data analysis was performed using SPSS statistical software version 27.0 (IBM; Armonk, NY, USA). A P-value of < 0.05 was considered statistically significant.

Results

Among the 37 pregnant women with fetal hydrops, 14 (37.8%) patients who developed MS were enrolled in the MS group and the remaining 23 cases in the non-MS group. The overall incidence of MS in this population was 0.024% (14/58, 428). Genetic analysis for fetal hydrops was conducted in 27 cases, and 17 fetuses were diagnosed with alpha thalassemia major. The other known associated etiologies included twin-to-twin transfusion syndrome (TTTS) (3/37), structural cardiac abnormalities (2/37), congenital cystic adenomatoid malformation (1/37), fetal anemia (1/37), and placental chorioangioma (1/37). All 23 patients in the non-MS group and 12 patients in the MS group opted to terminate their pregnancies.

Clinical features and outcomes of MS

Among the14 cases, the time of diagnosis of MS ranged from 22.6 to 34 gestational weeks. The etiology was identified in 11 cases and included alpha thalassemia major (7/14), structural cardiac abnormalities (2/14), TTTS (1/14), and fetal anemia (1/14). Prenatal sonographic findings indicated fetal hydrops (14/14), placental thickening (12/14), polyhydramnios (6/14), and oligohydramnios (3/14). Maternal clinical manifestations included edema (14/14: lower limb edema [10/14], ascites [6/14], pleural effusion, [5/14]), anemia (14/14), hypoalbuminemia (12/14), proteinuria (7/14) and hypertension (5/14). After delivery, placental edema was confirmed in all 14 cases by both gross and histologic placental examinations (Table S1).

Of the 12 pregnant women who opted for termination, seven patients had uneventful vaginal deliveries, and five underwent a cesarean section for various indications, which included maternal complications and failed labor induction. The other two patients for expectant management eventually underwent emergency cesarean section due to critical maternal conditions; two very preterm twin infants survived and one preterm neonate died. Maternal complications occurred in 12 cases, including postpartum hemorrhage (7/14), placenta accreta (5/14), acute left heart failure (3/14), renal dysfunction (3/14), pulmonary edema (2/14), HELLP (hemolysis, elevated liver enzymes, low platelet count) syndrome (2/14), liver dysfunction (1/14), placental abruption (1/14), and disseminated intravascular coagulation (DIC) (1/14). The ICU admission rate was 42.9% (6/14), and the median time for maternal symptoms to resolve was 7.5 days (range: 2–10 days) after delivery.

Differences between MS and non-MS groups

Compared to the non-MS group, the median gestational age at fetal hydrops onset was significantly later (27.8 vs. 23.0 weeks, P = 0.003) and the proportion of placental thickening was higher (85.7% vs. 34.8%, P = 0.006) in the MS group. Regarding maternal symptoms, pregnant women in the MS group had more weight gain (9.0 vs. 5.5 kg, P = 0.003) and higher rates of hypertension (35.7% vs. 0%, P = 0.005), anemia (100% vs. 65.2%, P = 0.015), and proteinuria (64.3% vs. 4.3%, P < 0.001). Moreover, their hemoglobin levels (88 vs. 105 g/L, P < 0.001), platelet counts (155 vs. 237 × 109/L, P = 0.005), and serum albumin levels (25.8 vs. 35.0 g/L, P < 0.001) were significantly lower. In contrast, the levels of uric acid (546 vs. 290 μmol/L, P = 0.042) and creatinine (59 vs. 44 μmol/L, P = 0.002) were higher in the MS group. As for outcomes, the rates of cesarean section, postpartum hemorrhage, blood product transfusion, and ICU admission were significantly higher in the MS group than in the non-MS group (50% vs. 0%, P < 0.001; 50% vs. 4.3%, P = 0.002; 64.3% vs. 4.3%, P < 0.001; and 42.9% vs. 0%, P = 0.001; respectively). Meanwhile, the median length of hospital stay was longer and the costs were significantly higher in the MS group (Table 1).

Table 1 Comparison of clinical data between mirror syndrome group and non-mirror syndrome group
Full size table

Risk factors for MS

MS is characterized as “triple edema”, and fetal and/or placental edema are precursors of maternal edematic manifestations. Therefore, the fetal and placental parameters associated with the edematous manifestations were chosen as risk factors of MS among cases with fetal hydrops. Binary logistic regression analysis showed that at time of fetal hydrops onset ≥24 weeks (odd ration [OR] 15.83, 95% confidence interval [CI] 1.56–160.10) and placental thickening (OR 8.63, 95% CI 1.29–57.72) were risk factors related to the MS (Table 2).

Table 2 Logistic regression analysis of risk factors for mirror syndrome
Full size table

Discussion

Although MS is a rare obstetric disease, the present case-control study found that it was relatively prevalent among fetal hydrops cases with an incidence of 37.8% (14/37) in the late second and third trimesters. The most common etiology of hydropic fetuses in southern China was alpha thalassemia major, and the fetal prognosis was very poor in both MS and non-MS groups. However, in the MS group, maternal morbidity increased with higher rates of postpartum hemorrhage, blood products transfusions, and ICU admissions; the key maternal features of MS included more weight gain, hemodilution and hypertension. In addition, our findings indicated that the time of onset of fetal hydrops at ≥24 weeks and prenatal placental thickening were risk factors for MS.

Since maternal edema is related to fetal hydrops and placental edema, any of the conditions associated with fetal hydrops or placental edema may cause MS. In southern China, alpha thalassemia major, which accounts for 28.4–55.1% of non-immune hydrops fetalis [10, 11], is also the common cause of MS. In our study, nearly half (17/37, 45.9%) of hydropic fetuses in the second and third trimesters were caused by alpha thalassemia major, and seven cases (7/17, 41.2%) developed MS. The high proportion of alpha thalassemia major in our study was inconsistent with frequently reported Rh-isoimmunization and viral infections in Western countries, which could be explained by differences in regions and populations. The other known etiologies included TTTS, fetal congenital structural anomalies, and placental chorioangioma, which were also common in previous studies [2, 3].

Although the maternal features of MS varied widely, all patients in the MS group manifested consistent symptoms of edema and hemodilution, characterized by lower extremity pitting edema, ascites, pleural effusion, anemia and hypoalbuminemia. Four patients presented with hypertension and proteinuria, and two of them developed HELLP syndrome before giving birth. Some studies have found that the changes in multiple placental factors associated with MS have the same pattern as preeclampsia [5,6,7,8]. This overlap between MS and preeclampsia can easily lead to misdiagnosis. However, consistent with previously reported cases of MS with maternal hemodilution [2, 3, 12, 13], our laboratory findings showed that the levels of hemoglobin, hematocrit, platelets and serum albumin were significantly lower in the MS group than in the non-MS group. Regardless of the fetal and placental causes, the hemodilution observed in cases of MS compared with typical hemoconcentration of preeclampsia may be considered in the differential diagnosis of these two diseases.

The prognosis of hydropic fetuses is generally poor and depends on the etiology and antenatal therapy [2, 3, 13]. In our study, the causes of hydrops were identified prenatally in 67.6% (25/37) of patients, and the most common etiology was alpha thalassemia major (17/37). All the parents in the non-MS group and 85.7% (12/14) in the MS group opted to terminate their pregnancies due to unsatisfactory guarantee for good fetal prognosis or worsening maternal conditions. Similar to previous reports [2, 3], no maternal mortality occurred in the MS group. However, maternal morbidity was as high as 85.7%, which was significantly higher than 8.7% in the non-MS group. Postpartum hemorrhage occurred in half of the patients in the MS group and the potential causes included placenta accreta, placentomegaly and maternal hemodilution, which are also features of MS. Moreover, nearly two-thirds of the patients in our study needed blood product transfusions and nearly half underwent ICU admission due to acute maternal heart failure, pulmonary edema, or DIC. However, the median time for resolution of maternal manifestations was 7.5 days (range: 2–10 days) after delivery, which was similar to the 5.5 and 8.9 days reported in other literatures [2, 3]. These adverse maternal outcomes and the subsequent resolution following delivery highlight the importance of early diagnosis and timely treatment of MS.

In this study, 37.8% (14/37) of fetal hydrops cases progressed to MS, implying that MS was relatively common among fetal hydrops cases, even though the actual incidence of MS at our hospital was as low as 0.024%. Thus, investigating the potential risk factors of MS is important for the early detection and close surveillance of high-risk pregnant women. However, to the best of our knowledge, there are no published studies on the risk factors for MS due to a paucity of cases or lack of awareness of this problem. MS mainly occurs during the late second and third trimesters [2, 3]. Among our MS cases, the median gestational age of fetal hydrops onset was 27.8 weeks, which was close to a previous report of 26.4 weeks [3] and significantly later than the median of 23 weeks in the non-MS group. Multivariate analysis showed that the time of fetal hydrops onset at ≥24 gestational weeks was an independent risk factor for MS; therefore, more attention should be paid to this high-risk population. Although the pathogenesis of MS is still unclear, some studies have suggested that placental abnormalities may be the inciting etiology [14, 15]. Our results showed that placental thickening was also a risk factor for MS. This finding highlights the importance of evaluating the placenta in hydropic fetalis cases.

Some limitations of our study should be discussed. First, despite MS being a rare obstetric condition, this retrospective study from a single center with a small sample may lead to selection bias. Second, nearly all patients in this study opted to terminate their pregnancies following diagnosis of fetal hydrops, and the actual rate of fetal survival and the efficacy of intrauterine interventions could not be assessed. Lastly, clinical data did not consider any pathophysiologic markers, and the mechanism of MS could not be discussed herein.

Conclusion

MS is a common obstetric disease among fetal hydrops cases in late second and third trimesters. The main cause of fetal hydrops and also MS is alpha thalassemia major in southern China. With diverse maternal manifestations, the key features include more weight gain, hemodilution, and hypertension. The high maternal morbidity is mainly attributed to postpartum hemorrhage. In addition, this study demonstrated that the time of onset of fetal hydrops at ≥24 weeks and prenatal placental thickening are both risk factors for MS; thus, more attention should be paid to high-risk populations that have these characteristics. Additional prospective studies with larger study populations of fetal hydrops and MS cases are needed to further assess these risk factors.