SEARCH

SEARCH BY CITATION

Keywords:

  • Fetus;
  • paediatric outcome;
  • ultrasound;
  • unexplained polyhydramnios

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

Our aim was to assess the long-term outcomes of infants born out of a pregnancy complicated by unexplained polyhydramnios. We retrospectively analysed a cohort of 173 singleton pregnancies with polyhydramnios. There were 24 singletons with unexplained polyhydramnios, defined as polyhydramnios where prenatal testing was negative. Infants were examined by a paediatrician at a median age of 12 months (range 12–64 months). Of the 24 infants, 19 (79%) had a normal outcome. West syndrome, polyuria and pulmonary stenosis were diagnosed in three children during follow up, while there were two perinatal deaths. Paediatric follow up of children born out of pregnancies complicated by unexplained polyhydramnios should be carried out systematically.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

Polyhydramnios is a common obstetric complication. Aneuploidy is present in 10% of fetuses with sonographic anomalies and in 1% when the ultrasound examination is considered to be normal.1 In persistent polyhydramnios, the prevalence of aneuploidy is increased (10–20%) compared with polyhydramnios with spontaneous resolution.2 Other major causes of polyhydramnios include maternal diabetes, isoimmunisation disease, congenital abnormalities and multiple gestations.3 Therefore, prenatal ultrasound follow up of polyhydramnios is justified and invasive management may be required. In a large number of pregnancies, the polyhydramnios remain unexplained despite extensive prenatal assessment. Unexplained polyhydramnios has been associated with significantly higher rates of malpresentation, macrosomia, primary caesarean delivery and an increased rate of perinatal mortality.4 However, data are lacking to enable us to determine the outcome of children born following a prenatal diagnosis of unexplained polyhydramnios. This clearly hinders prenatal counselling.

The main objective of this study was to report on the outcome of these children. Ultimately, our aim is to inform clinicians facing unexplained polyhydramnios about the potential health issues related to this condition.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

We performed a retrospective cohort study of 173 singleton pregnancies with polyhydramnios hospitalised between 2001 and 2005. Among them, 24 singletons pregnancies complicated by unexplained polyhydramnios were identified. The medical charts of these pregnancies and children were reviewed. Polyhydramnios was defined as an amniotic index greater than 24 cm by measurement of a four-quadrant amniotic fluid index (AFI).5 Unexplained polyhydramnios was defined as following: (i) normal karyotype, (ii) normal detailed ultrasound at diagnosis, (iii) negative TORCH test (toxoplasma, other [congenital syphilis and viruses], rubella, cytomegalovirus and herpes simplex virus) and (iv) no maternal diabetes mellitus. Furthermore, all women had follow-up ultrasound examinations which showed no abnormalities that could be the aetiology of the polyhydramnios. Amniocentesis was performed at diagnosis to assess fetal chromosomes and TORCH infection. Amnioreduction was performed in cases of polyhydramnios complicated by uterine contractions because of an enlarged uterus and in the presence of cervical shortening, or in cases of maternal discomfort, to reduce the uterine size. Children were examined by a paediatrician within 2 days after birth, at 3 months and every year thereafter until the age of 5 years. The following factors were analysed: the age of the woman, parity, gestational age at the diagnosis, AFI, number of amnioreductions, the amniotic fluid volume retrieved for each procedure, total amniotic fluid volume retrieved, gestational age at the delivery, mode of delivery, birthweight, neonatal outcome and paediatric follow up.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

Twelve thousand seven hundred and eighty-six women delivered in our hospital during the study period. One hundred and seventy-three women were referred to our unit for the management of polyhydramnios. Twenty-four out of 173 (13.9%) were singleton pregnancies and met the criteria to be considered as unexplained. The median maternal age, parity and gestational age at diagnosis were 30 years (range, 17–42 years), 1 (range, 1–5) and 24.3 weeks (range, 22–32.7), respectively. Twenty-three out of 24 women underwent amnioreduction, and one woman did not accept any invasive procedure. The median number of amnioreductions for each woman was 1 (range, 0–12). The median volume retrieved was 1500 ml (range, 0–21 600).

The median gestational age at birth and median birthweight were 39.1 weeks (range, 27.0–41.1 weeks) and 3300 g (range, 1350–3910), respectively. Median follow up was 12 months (range, 12–64).

The outcome of the 24 pregnancies with unexplained polyhydramnios is summarised in Table 1. One case was complicated by in utero fetal death, despite a normal follow-up anomaly scan. The fetopathological examination, performed with the parents’ prior consent, showed an oesophageal atresia with a ventricular septal defect and a horseshoe kidney malformation in keeping with the diagnosis of VACTERL syndrome. An amnioreduction of 1300 ml had been performed because the woman had uterine contractions with cervical length modification at 31.5 weeks. An intrauterine death was diagnosed 2 weeks after amnioreduction. Five of the 24 (20.8%) infants with unexplained polyhydramnios had an abnormal neonatal examination at birth (Table 1). Among these five cases, four were because of neonatal complications related to polyhydramnios. Perinatal anoxia in case 2 was because of a fetal bradycardia that occurred following an amnioreduction at 37 weeks. A caesarean section had to be performed by an obstetrician. Cases 3 and 4 were associated with premature delivery because of polyhydramnios and/or amnioreduction at 27 and 29.5 weeks, respectively. The child born at 27 weeks (case 3) died on day 22 because of prematurity and periventricular leucomalacia. Case 4 had a supraventricular tachycardia at neonatal examination, but follow-up examinations were normal thereafter until 5 years of age. Case 5 had polyuria with a urinary volume of 280–320 ml per day between the fourth and the seventh postnatal day. Hypercalciuria was shown to be responsible for the polyuria, but no definite aetiology had been found at the age of 12 months. In this case, we estimated prenatal urine production using three-dimensional ultrasound with the VOCAL rotational method. This suggested the fetal urine production rate was 200 ml per hour at 36 weeks, five times the normal value for that gestational age. In case 6, a Streptococcal B maternofetal infection occurred. At 4 days of age, a pulmonary stenosis was diagnosed. A neonatal examination performed in 18 out of 23 cases was considered normal. Among them, 17 had a normal neurological development between 12 and 60 months. In case 7, West syndrome was suspected at the age of 3 months, and the diagnosis was confirmed at 12 months. West syndrome is defined by the following triad: infantile spasms, a specific electroencephalogram pattern termed hypsarrhythmia and mental retardation. This case was associated with early, severe and recurrent polyhydramnios diagnosed at 22 weeks. Twelve amnioreductions were required because of threatened premature labour. The total volume of amniotic fluid retrieved was 21 l, and the delivery occurred at 34 weeks of gestation. We estimated the prenatal urine production rate to be 48 ml per hour at 33 weeks, which is in the normal range for the gestational age. The impossibility to swallow correctly may have been the first sign of neurological impairment and, subsequently, the cause of polyhydramnios.

Table 1.  Detailed outcome of the 24 cases of unexplained polyhydramnios
 Number of childrenGestational age at diagnosis (weeks)AFI (cm)Number of amnioreductionsVolume retrieved (ml)Gestational age at delivery (weeks)Mode of deliveryBirthweight (g)Diagnosis at birthFollow up (months)Diagnosis during follow up
  1. CS, caesarean section; IUD, intrauterine death, NA, not applicable; VD, vaginal delivery.

  2. Values represent n or median unless otherwise indicated.

Normal outcome1724.1 (22.0–32.7)25.5 (18.3–28.6)1 (0–6)1300 (0–8650)39.4 (36.0–41.1)3410 (2680–3890)Normal12 (12–63)Normal
Abnormal outcome727.0 (22.0–32.0)30.0 (24.0–37.0)1 (1–12)2000 (1300–21 600)34.0 (27.0–40.1)2500 (1350–3910)16 (12–64)
Case 1131271130034VD1780IUD, VACTERL syndromeNANA
Case 2132361280037CS3300Acute perinatal anoxia22Normal
Case 3127321200027VD1430PrematurityDeceased at day 22
Case 4123241200029VD1350Prematurity and supraventricular tachycardia64Normal
Case 5128373530037VD3000Normal12Polyuric syndrome
Case 6122281150040VD3910Streptococcus B infection12Pulmonary stenosis
Case 7122301221 60034VD2500Normal16West syndrome

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

This is the first report on the outcome of children resulting from pregnancies complicated by unexplained polyhydramnios. Of the 24 children born out of pregnancies complicated by unexplained polyhydramnios, 6 (25%) had abnormalities identified at birth or at follow up. In most cases (5/6), the neonatal examination was abnormal, and in three of these five cases, this was a consequence of the prematurity induced by polyhydramnios and/or the amnioreduction.

Childhood pathologies related to unexplained polyhydramnios during pregnancy were revealed in two cases. Polyhydramnios may be the result of fetal polyuria or fetal impossibility to swallow. In our study, this is illustrated by cases 5 and 7 with neonatal polyuria and West syndrome, respectively. Many studies have evaluated obstetric complications arising from polyhydramnios,5–7 but none of them have reported the paediatric outcome. Data regarding the effect of polyhydramnios on obstetric complications have been conflicting.4,7 Some authors showed an increased rate of perinatal mortality with an increase in intrapartum morbidity, caesarean section and large-for-gestational-age children in women with polyhydramnios, when compared with normal pregnancies.4 However, other authors have not reported an increase in premature delivery, intrapartum complications, neonatal depression, low Apgar scores at 5 minutes, neonatal intensive care unit admissions and perinatal mortality rate in pregnancies complicated by mild, unexplained polyhydramnios in comparison with a control group with a normal fluid volume.7 Idiopathic polyhydramnios was only associated with significantly higher rates of malpresentation, macrosomia and primary caesarean delivery, suggesting that such complications resulted from their underlying cause. None of these studies were limited to idiopathic polyhydramnios, and their results were reflective of outcomes associated with polyhydramnios of all causes.

Glantz et al.2 found that persistent polyhydramnios was associated with increased adverse perinatal outcomes compared with polyhydramnios with spontaneous resolution. This observation is in accordance with our study, as we observed that the worst outcomes have been associated with a severe and persistent polyhydramnios.

The sensitivity for the detection of mild polyhydramnios using AFI has been reported to be low. A recent study has shown that ultrasound measurements of amniotic fluid are poor in identifying oligo and polyhydramnios, even when using the percentile by means of gestational-age-specific values. Magann et al. reported that the probability of correctly identifying polyhydramnios is 33–46%. In this study, AFI was compared with the single deepest pocket in the identification of actual abnormal amniotic fluid volumes. The sensitivity was 29% for a specificity of 97% using an AFI above 20 cm, while a sensitivity of 29% for a specificity of 94% was found when using a single deepest pocket above 8 cm.8 These poor sensitivities suggest that persistence of polyhydramnios with need to perform several amnioreductions could be a more valuable prognostic criterion than measurement of AFI in mild polyhydramnios. In our study, repeated amnioreductions were performed in women with uterine contractions and cervical length modification or maternal discomfort because of polyhydramnios in order to reduce the uterine size. Since the benefits of amnioreduction are short lived as the fluid rapidly reaccumulates, this should be balanced with the risk of prematurely rupturing the membranes and the risk of abruption with removal of amniotic fluid.

The number of cases reported in our study is too limited to allow us to draw conclusions on the outcome of the children born following a prenatal diagnosis of unexplained polyhydramnios. We found that 3 of the 23 children required a close paediatric follow up. Furthermore, the degree of severity of the polyhydramnios, defined by early gestational age at diagnosis, the AFI, the number of amnioreductions and the total volume drained, seems to be important for the prognosis of these children.

Most of the current recommendations for the evaluation of polyhydramnios are reasonable. When polyhydramnios is diagnosed, an evaluation for fetal or maternal factors is indicated, which, at minimum, should include a detailed ultrasound examination, amniocentesis and a diabetes screen. When it remains unexplained, we recommend a specific paediatric follow up. If polyhydramnios is persistent, an ultrasound follow-up examination should be performed, especially when several amnioreductions have been required. In the same way, a close paediatric survey is needed for these children, even with a normal neonatal examination. Additional assessment of long-term outcome of these children needs to be set by obstetricians together with paediatricians in order to achieve appropriate prenatal counselling to parents.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References