Proposal of a novel index in assessing perinatal mortality in prenatal diagnosis of Sacrococcygeal teratoma

Introduction

Sacrococcygeal teratoma (SCT) is the most frequent fetal germ cell tumor [1] generally benign and often highly vascularized. The reported incidence is 1 in 35,000 to 40,000 newborns [2]. Prenatal incidence is believed to be as high as 1 in 28,500, due to the risk of intrauterine death of 30–50 % [3], 4]. Currently, prenatal ultrasound can detect up to 90 % of SCTs [3]. Moreover, magnetic resonance imaging (MRI) has been prove useful in differential diagnosis and precising tumor extension and lesion type [5]. Perinatal outcome has been related to tumor size, vascularity and composition of the tumor [6]. A large tumor volume with rapid tumor growth and a predominant solid component associated with high vascularization is the characteristic of a tumor with poor prognosis [7], [8], [9], [10]. Cardiac failure can occur due to high output status ending in hydrops and fetal death [9].

High tumor volume to fetal weight ratio (TFR) calculated by 24 weeks of gestation has been used to provide prognostic information [11]. A higher value of 0.12 has shown to be predicted of an adverse perinatal outcome [8], 12].

Management includes preterm delivery with postnatal tumor resection. A few fetal interventions have been attempted to prevent death [13]. However, the effectiveness of these interventions is not proved, and may pose at higher fetal risk for preterm delivery, and maternal risks such as blood loss and need for an urgent caesarean section [6].

We describe the perinatal characteristics and outcomes of fetuses with SCT diagnosed in a tertiary teaching hospital over the last 10 years. Additionally, we describe an index based on the head circumference and tumor volume ratio plus polyhydramnios with postnatal mortality.

Materials and methods

We performed a retrospective study of patients with prenatal diagnosed of SCT at Instituto Nacional Materno Perinatal of Lima – Peru between 2013 and 2023. Maternal and neonatal clinical records were reviewed. A previous published case was included [14]. All ultrasound examination were performed by our team of fetal medicine specialist. Data collected included demographic and obstetric characteristics gestational age at diagnosis, tumor dimensions, estimated fetal weight, TFR, tumor volume to head circumference ratio (THC), presence or absence of hydrops, presence or absence of polyhydramnios, Mirror syndrome, preterm premature rupture of membranes (PPROM), preterm birth, delivery mode, intrapartum hemorrhage, gestational age at birth. Perinatal outcomes included: livebirth, termination of pregnancy, intrauterine fetal death, postnatal death. Intrauterine fetal death was defined as an infant delivered without signs of life after 24 weeks of gestation [10]. Hydrops was defined as accumulation of fluid in two or more body cavities and polyhydramnios as an index amniotic fluid greater than 25 cm.

Hadlock formula was used to estimate fetal weight and tumor volume was calculated by the ultrasound software based on its three dimensions [15]. TFR was calculated as previously described, by dividing the calculated tumor volume (cc³) by the estimated fetal weight (grams) [12].

Since fetal head has a constant and predicted growth throughout gestation, here we propose the THC ratio which was calculated dividing the tumor volume (cc³) by the head circumference. Additionally, since polyhydramnios is commonly assessed in fetuses with tumors as a proxy of an increased cardiac output [16], here we propose a novel index to assess postnatal mortality based on the THC ratio and the addition of the presence of polyhydramnios.

Statistical analysis was performed using Stata statistical software package version 14.0. We performed a descriptive analysis, and the distribution of the absolute and relative frequencies of categorical variables was calculated. For numerical variables, summary measures were applied as median and range. Patients were grouped according to the main outcome: survival at six months of age and a receiver operating characteristic (ROC) curve analysis was used to determine the optimal TRF, and THC plus polyhydramnios cutoff value based on optimal sensitivity and specificity. p-values < 0.05 were considered significant. This study was approved by the local Ethical Institutional Board (reference number: 031-2021-CIEI/INMP).

Results

From 2013 to February 2023, 15 fetuses were diagnosed with SCT. Three cases were excluded because they had incomplete medical data. Table 1 details the characteristics and outcomes of fetuses with SCT. The median maternal age was 27 (range, 17–33) years and the median gestational age at prenatal diagnosis was 30.9 (range, 17.9–38.7) weeks. Seven patients developed polyhydramnios. Five patients underwent an additional study with fetal magnetic resonance imaging (MRI). Five patients received tocolysis as treatment to threatened preterm delivery. Two fetuses underwent intrauterine blood transfusion. One of them was complicated with a subsequent Mirror syndrome and the pregnancy was ended at 23 weeks. An emergency cesarean section due to preterm labor was reported in eight patients (72.7 %). Three women patient was complicated with intrapartum hemorrhage (27.3 %), and two of them required intraoperative blood transfusion.

The median gestational age at birth was 35.8 weeks (range, 32.8–39.7). There was no intrauterine death. There were four deaths after birth (4 out of 11, 36.4 %). Two fetuses died in the first hours of life due to heavily bleeding at delivery. A third fetus died at 12 days after birth due to severe pulmonary hypertension. The fourth fetus died after 3 months following removal surgery due to sepsis. Among those underwent removal surgery, five tumors (62.5 %) were reported as mature cystic teratoma, and two as immature cystic teratomas.

Figure 1 describes the ROC curves illustrating sensitivity and specificity of both TFR, and THC plus polyhydramnios. A cutoff value of THC plus polyhydramnios higher than 3.0 was predictive of postnatal mortality showing a sensitivity of 100 % and specificity of 86 %. The area under the curve was 0.96. However, there is not significant difference comparing TFR with THC plus polyhydramnios (p=0.631) associated with postnatal mortality. Besides, the evaluation THC independently was 0.89 (area under curve) ROC, compared TFR 0.93, and THC + Polyhydramnios 0.96.

Figure 1:

ROC curves using ratios with tumor measurements in the assessment of postnatal mortality.

Discussion

This retrospective study from a tertiary hospital reports a high postnatal mortality of 36 % among fetuses with diagnosis of Sacrococcygeal Teratoma. Additionally, we propose a potential useful index associated with postnatal mortality based on the THC and the addition of polyhydramnios with a sensitivity of 100 %.

Currently, the prenatal diagnosis of a large Sacrococcygeal Teratoma is straightforward during the second or third trimester routine ultrasound scan. However, the management is still challenging due to the high perinatal mortality. Recently a systematic review has found that fetal cardiomegaly, placentomegaly, hydrops, tumor hypervascularity, solid tumor morphology and TFR >0.12 before 24 weeks are significant markers of poor outcomes [17].

The high postnatal mortality could be explained mainly by the size of the tumor and the polyhydramnios leading overdistension of the uterus causing contraction and eventually early preterm delivery. Additionally, a fetus with such a high vascular tumor size could steal blood flow from the placenta and fetus, causing severe anemia and compensatory hypovolemia ending in a cardiac failure and hydrops [17].

The postnatal mortality is variable according to the authors, which could be explained to a biological differences in size, invasiveness in the pelvic region, type of tumor and the local resource settings [10]. Interestingly, two out of four immediate postnatal death presented bleeding at delivery probably caused by the difficult extraction during the emergency cesarean section.

Recent studies have reported a decrease in the postnatal mortality as low as approximately 13 %, probably related to the more anticipated and comprehensive multidisciplinary perinatal approach in developed countries [18].

However, we still show a higher postnatal mortality than other studies reported by other major referral centers in developed countries, which can reflect the differences in resources and the lack of a standardized multidisciplinary protocol to address the delivery of a fetus with Sacrococcygeal Teratoma.

Surprisingly, in our series we only had one case of fetal hydrops with very early presentation that could not be reversed with intrauterine blood transfusion and resulting in Mirror syndrome, which is a rare association reported occasionally [19]. Other studies have reported a variable prevalence of hydrops (16–70 %) [9], 20]. Similarly, we did not have any case of intrauterine death.

There have been several reports describing predictors of adverse perinatal outcomes [8], 17]. Rodriguez et al. in 2011, described the relationship between tumor volume to fetal weight ratio with perinatal outcome and propose a cut-off >0.12 prior to 24 weeks gestational age associated with poor perinatal outcome [11]. In assessing fetal tumors, a more common and simple approach is to compare the tumor volume with the fetal head circumference, since fetal head has a constant growth throughout gestation and using the abdominal circumference in estimate fetal weight in a fetus with hydrops or intrapelvic tumor could overestimate the actual fetal weight altering the TFR. We considered adding polyhydramnios, since has been reported as a predictor of poor outcome in fetuses with SCT [21], 22]. Although, van Heurn et al. in a systematic review suggested that there are not statistical association between polyhydramnios and adverse outcomes, despite a close margin OR: 1.98 CI (1.01–3.90, p=0.05). Thus, the association of THC and polyhydramnios could be more sensitive to determine postnatal mortality. Here, we explored a simple approach to assess postnatal mortality based on the THC and the addition of the presence of polyhydramnios reporting a high sensitivity equivalent to the TRF as showed in Figure 1 [7].

There are some limitations in this study. First, the study design was retrospective with a short number of cases, however, SCT is a rare pathology, and the current available information is based in these types of studies. Second, in our institution there is no a standardized multidisciplinary protocol for follow-up and perinatal management. Third, a potential referral bias with a delayed gestational age at referral, however, our institution is one of the major centers for fetal diagnosis and fetal therapy.

In summary, in our setting, we still have a high postnatal mortality and we propose a novel index based on THC plus polyhydramnios with a sensitivity of 100 % to assess postnatal mortality equivalente to the classical TRF ratio, but avoiding the bias of the estimated fetal weight. We agree that there is need of studies exploring long-term outcomes.