Perinatal Outcomes of Pregnancies with Borderline Oligohydramnios at Term

Amniotic fluid volume ranges from about 50 mL at 12 weeks to nearly 1 L at 36–38 weeks, and gradually decreases to 600–800 mL at term and about 200 mL at 43 weeks [1]. Amniotic fluid has crucial roles in protecting the developing fetus. For one, the fluid acts as a protective shock absorber, allowing for fetal growth and movement while preventing umbilical cord compression. Amniotic fluid also provides aseptic and bactericidal protection against infections within the uterine cavity [2].

Amniotic fluid is a vital marker of placental function, and the amniotic fluid index (AFI) is used to assess fetal well-being. An AFI of ≤5 cm signifies oligohydramnios, for which the guidelines recommend labor induction and delivery between 36 0/7 and 37 6/7 weeks. This recommendation holds even in the presence of a non-compromised fetus and the absence of maternal disease [3, 4]. Borderline oligohydramnios is characterized by an AFI ranging from 5.1 to 8 cm [5], and the evidence regarding its relation to pregnancy outcomes is contradictory. Some retrospective studies have shown an association between borderline oligohydramnios in term pregnancies and adverse outcomes [6, 7]. The latter include an increased likelihood of cesarean delivery (CD) due to fetal distress and higher occurrences of low Apgar scores, low birthweight, and higher risk of neonatal intensive care unit (NICU) admission. However, other studies have not found a significant correlation between borderline oligohydramnios and abnormal fetal outcomes [8, 9]. Guidelines have not been established regarding the management and surveillance of pregnancies with AFI falling within the range defined as borderline oligohydramnios.

Considering the limited relevant data available, we aimed to investigate the potential impact of borderline oligohydramnios on adverse pregnancy outcomes. To this end, we compared perinatal outcomes between women with borderline oligohydramnios (AFI 5.1–8 cm), a matched group with normal AFI (8.1–25 cm), and women with oligohydramnios (AFI ≤5 cm). We hypothesized that among pregnancies complicated by borderline oligohydramnios compared to normal AFI, incidences of adverse outcomes would be higher.

In this retrospective study, we examined the medical records of women with singleton term pregnancies that occurred between 37 and 42 weeks of gestation, during January 2018 to December 2022. The study was conducted at the Galilee Medical Center, a tertiary university-affiliated hospital in northern Israel, which serves a diverse population of about 600,000, with an annual birth rate of about 5,000 deliveries. The inclusion criteria were women showing either borderline oligohydramnios (AFI 5.1–8.0 cm) or oligohydramnios (AFI ≤5 cm) when assessed within 1 week of delivery. To compare women with borderline oligohydramnios to those with normal AFI (8.1–25 cm), we matched women with normal AFI measurements one-to-one to those with borderline oligohydramnios. Specifically, one woman with normal AFI, who delivered at the closest time during the same day, was matched as the control for each woman with borderline oligohydramnios. Women with preterm delivery <37 weeks, multiple pregnancy, rupture of membranes, preeclampsia, diabetes, chronic hypertension, fetal anomaly, or severe systemic maternal disease were excluded from the study.

Gestational age (GA) was determined from either the last menstrual period or from ultrasound measurements taken before 12 weeks of gestation. Standardized sonographic evaluations involved assessing placental location, AFI, and fetal biometry. Fetal biometry included measurements of biparietal diameter, head circumference, abdominal circumference, and femur length.

For the three study groups, we gathered data on maternal and pregnancy characteristics, including maternal age, gravidity, parity, rates of nulliparity, ethnicity, a history of previous CD, fetal gender, and AFI measured in centimeters. According to our department protocol, AFI is routinely performed for women referred to the delivery room. Obstetric information collected included the mode of delivery and the rate of cesarean deliveries due to fetal distress. Neonatal information included 1-min and 5-min Apgar scores, umbilical artery pH, the presence of meconium-stained amniotic fluid, birthweight, admission to the NICU, incidences of respiratory distress syndrome and transient tachypnea of the newborn, and the occurrence of small-for-gestational-age (SGA) neonates. The latter was defined as birthweight below the 10th percentile, based on local growth curves [10]. We defined an adverse neonatal outcome that reflects fetal distress. The correlation with fetal distress is under the supposition that borderline oligohydramnios might signal placental insufficiency or be associated with cord compression during labor. We defined a neonatal adverse composite outcome as the presence of at least one of the following: CD due to fetal distress, meconium-stained amniotic fluid, a 5-min Apgar score below 7, umbilical artery pH below 7.15, or NICU admission. Notably, only cesarean deliveries that were due to fetal distress were included in the composite outcome so as to reflect neonatal rather than maternal outcomes.

Continuous variables were compared with the t test and categorical variables with the χ² or Fisher’s exact test when appropriate. Rank sum analysis (Mann-Whitney) was utilized as appropriate. A p value <0.05 was considered significant. Logistic regression analyses were used to assess associations of borderline oligohydramnios with SGA neonates and with CD due to fetal distress.

Our hypothesis posited that borderline AFI is an indicator of placental insufficiency, consequently correlating with a higher rate, compared to normal AFI, of SGA neonates. A previous study [11] reported that term pregnancies characterized by borderline AFI compared to normal AFI exhibited a greater risk of SGA neonates: 16.6% versus 4.3%. Based on these findings, and considering a two-sided alpha of 0.05% and a beta of 0.20, we computed a required sample size of 112 women with borderline oligohydramnios and the same number of matched women with normal AFI.

During the study period, 140 women had borderline oligohydramnios and 345 had oligohydramnios. The control group consisted of 140 women with normal AFI. Table 1 describes the baseline characteristics of the study groups. The proportion whose maternal age was <18 or >35 years was higher in the oligohydramnios group than the control group (10.7% vs. 2.9%, p = 0.008). The proportion with nulliparity was higher in the borderline oligohydramnios and oligohydramnios groups than in the control group (52.9%, 48.1%, 25.0%, respectively, p < 0.001).

Labor induction rates were comparable between women with borderline oligohydramnios (77.4%) and oligohydramnios (70.0%) but significantly higher than in the control group (40.7%, p < 0.001). Among neonates born to women with borderline oligohydramnios and oligohydramnios, the mean birthweights were lower than that of the control group (3,133 ± 484 g, 3,165 ± 406 g, and 3,287 ± 440 g, respectively, p = 0.017). Rates of SGA neonates were comparable in the borderline oligohydramnios (10.0%) and oligohydramnios (10.4%) groups but higher than in the control group (2.9%, p = 0.022). Table 2 shows the outcomes of the study groups. In a multivariate regression analysis that adjusted for maternal age and nulliparity, borderline oligohydramnios was found to be significantly associated with SGA neonates (adjusted odds ratio [aOR] = 3.6, 95% confidence interval [CI] 1.1–11.6, p = 0.034). CD rates due to fetal distress were similar between women with borderline oligohydramnios (15.0%) and oligohydramnios (12.2%) but significantly higher than in the control group (4.3%, p = 0.01). In a multivariate regression analysis that adjusted for maternal age, nulliparity, SGA status, and labor induction, borderline oligohydramnios was found to be significantly associated with CD due to fetal distress (aOR = 3.0, 95% CI 1.1–8.3, p = 0.032). The composite neonatal outcome exhibited comparable rates in the borderline oligohydramnios (20.7%) and oligohydramnios (18.6%) groups but was higher than for the control group (9.3%, p = 0.019) (Fig. 1).

The present study indicates that borderline oligohydramnios in term pregnancies shares similar obstetrical outcomes with oligohydramnios and is associated with increased risks of adverse perinatal outcomes, such as delivering SGA neonates and requiring CD due to fetal distress. Our observation of an increased risk of delivering SGA neonates among women with borderline oligohydramnios aligns with a previous study [8]. There, the SGA rate was 5.4% among 112 term singleton pregnancies with borderline oligohydramnios, compared to 1.6% among women with normal AFI (p = 0.026).

We did not observe increased risks of meconium-stained amniotic fluid and NICU admission among women with borderline oligohydramnios compared to women with normal AFI. This contrasts with studies that indicated an increased likelihood of these outcomes among women with borderline oligohydramnios [11, 12]. However, Gumus et al. [11] reported that women with borderline oligohydramnios were in their third trimester [11], whereas Bank et al. did not provide the GA at delivery for these women [13].

Similar to previous reports [11, 12], we found an association of borderline oligohydramnios with an increased risk of CD due to fetal distress. This contrasts with a study on term pregnancies, in which only 5.4% of women with borderline oligohydramnios underwent CD due to fetal distress, compared to 4.4% of those with normal AFI [8]. Notably, the rate of 5.4% for the borderline oligohydramnios group was lower than ours, 15.0%. This disparity could be attributed to differences in the study population or variations in obstetrical management practices in the delivery room.

Following a review of the literature, Magann et al. [5] concluded that the predictive accuracy of borderline oligohydramnios for adverse pregnancy outcomes is inconclusive, possibly due to variations in study design and the absence of randomized trials. The review identified variations in the inclusion criteria among the limited relevant studies. For instance, some studies have defined borderline AFI as ranging from 5.1 to 8.0 cm, while others have extended it to 5.1–10 cm. Additionally, differences exist in the GA at delivery, with some studies including deliveries in the third trimester and others focusing on those occurring between 37 and 42 weeks. Furthermore, certain studies assess outcomes such as SGA neonates and NICU admission, whereas others do not report on these outcomes. Despite these variations in study design, it was concluded that borderline oligohydramnios might be associated with adverse outcomes; however, evidence is still lacking regarding the necessity of closer antepartum testing or delivery. The authors proposed that sonography might be worth considering for assessing fetal growth, until the establishment of a conclusive link, or the absence of one, between borderline AFI and SGA [5]. Our findings that women with borderline oligohydramnios exhibit outcomes comparable to those with overt oligohydramnios are intriguing. These results support closer antepartum surveillance, especially for identifying abnormal fetal growth of women with borderline oligohydramnios. Notably, a lower risk of developing oligohydramnios within the subsequent 4 days was reported for pregnancies with an AFI >8 cm than for pregnancies with borderline AFI (5–8 cm): 2.3% versus 16.2%, respectively [12].

Our study compared borderline oligohydramnios to two distinct groups: those with normal AFI and those with oligohydramnios; and revealed similarly increased risks of adverse neonatal outcomes in the borderline oligohydramnios as in the oligohydramnios group. Additionally, to the best of our knowledge, our study included the largest reported sample of women with borderline oligohydramnios at term. However, limitations exist. We measured amniotic fluid using AFI, which is associated with a higher rate of diagnosing oligohydramnios than the deepest single pocket method, consequently leading to higher rates of labor induction [14]. Furthermore, we did not include women whose borderline oligohydramnios had resolved during antenatal follow-up, and we only analyzed AFIs measured within 1 week of delivery. Doppler studies, which could have provided further insights into the association between borderline oligohydramnios and adverse perinatal outcomes, were not conducted. Previous research suggests that borderline AFIs, especially when coupled with abnormal Doppler velocimetry, warrant closer antenatal surveillance due to an increased risk of adverse perinatal outcomes [15]. Moreover, our study is limited by the lack of blinding of obstetricians to the AFI results. This could have influenced decisions on CD for presumed fetal compromise, particularly among nulliparous women.

Our findings indicate an association between borderline oligohydramnios and increased adverse perinatal outcomes. Large prospective studies are needed to identify specific subgroups of women with borderline oligohydramnios who require closer antenatal surveillance or earlier labor induction.

The protocol for this study was approved by the Institutional Review Board of the Galilee Medical Center. The need for written informed consent was waived because of the retrospective design and nature of the study (NHR-066-22, date of approval November 3, 2022).

The authors have no relevant financial or non-financial interests.

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

Inshirah Sgayer, Lior Lowenstein, and Marwan Odeh contributed to the project’s development. Mohamed Elafawi, Olga Braude, and Sarah Abramov contributed to the data collection. Inshirah Sgayer, Mohamed Elafawi, Olga Braude, and Sarah Abramov contributed to the data analysis. Inshirah Sgayer wrote the first draft of the manuscript study. Lior Lowenstein and Marwan Odeh edited the manuscript. All the authors read and approved the submitted version of the manuscript.

The data that support the findings of this study are not publicly available due to their containing information that could compromise the privacy of research participants but are available from the corresponding author (I.S.) upon reasonable request.