Neonatal Outcomes are Similar between Patients with Resolved and Those with Persistent Oligohydramnios

Abstract

Objective

Oligohydramnios (defined as amniotic fluid volume < 5 cm or deepest vertical pocket < 2 cm) is regarded as an ominous finding on prenatal ultrasound. Amniotic fluid, however, is not static, and to date, there have been no studies comparing perinatal outcomes in patients who are diagnosed with oligohydramnios that resolves and those who have persistent oligohydramnios.

Study Design

This is a secondary analysis of a National Institutes of Health-funded retrospective cohort study of singleton gestations delivered at a tertiary care hospital between 2002 and 2013 with mild hypertensive disorders and/or fetal growth restriction (FGR). Maternal characteristics, delivery, and neonatal information were abstracted by trained research nurses. Patients with a diagnosis of oligohydramnios were identified, and those with resolved versus persistent oligohydramnios at the time of delivery were compared. The primary outcome was a composite of neonatal resuscitation at delivery: administration of oxygen, bag–mask ventilation, continuous positive airway pressure, intubation, chest compression, or cardiac medication administration. Secondary outcomes included FGR, timing, and mode of delivery.

Results

Of 527 women meeting study criteria, 42 had oligohydramnios that resolved prior to delivery, whereas 485 had persistent oligohydramnios. There were no significant differences in patient demographics between groups. The gestational age at diagnosis was significantly lower for patients with resolved versus persistent oligohydramnios (median: 33.0 [interquartile range, IQR: 29.1–35.9] vs. 38.0 [IQR: 36.4–39.3], p<0.001). There was not a substantial difference in rate of neonatal resuscitation (41 vs. 32%, p = 0.31). Patients with resolved oligohydramnios were more likely to have developed FGR than those with persistent oligohydramnios (55 vs. 36%, p<0.02). There were no significant differences for gestational age at delivery, birth weight, or neonatal intensive care unit admission.

Conclusion

Patients whose oligohydramnios resolved were diagnosed earlier yet had similar rates of neonatal resuscitation but higher rates of FGR than those who had persistent oligohydramnios.

Oligohydramnios is defined as amniotic fluid index (AFI) of ≤5 cm or deepest vertical pocket (DVP) of ≤2cm and is diagnosed at rates of 0.5 to 11%.^1–5 Previous studies investigating perinatal outcomes in patients diagnosed with oligohydramnios in the preterm period have identified a strong association with adverse perinatal outcomes.^2,5–8 Nevertheless, preterm delivery for isolated oligohydramnios has been shown to lead to more neonatal intensive care unit (NICU) admissions and lower birthweight without a reduction in neonatal respiratory morbidity, neonatal death, and other complications of prematurity.⁵

Additionally, resolution of oligohydramnios occurs in up to 8.3% of pregnancies.⁵ However, to date, no studies have been performed which evaluate the outcomes of pregnancies in which oligohydramnios is diagnosed and then resolves as compared with those with persistent oligohydramnios. We hypothesized that patients with oligohydramnios that resolves on subsequent prenatal imaging will have improved perinatal outcomes when compared with patients with persistent oligohydramnios.

Methods

This is a secondary analysis of a large National Institutes of Health–funded retrospective cohort study of women who delivered singleton gestations at a tertiary care hospital between January 2002 and March 2013.⁹ The original study was performed to assess outcomes of iatrogenic delivery between 34 and 38 weeks’ gestation on perinatal outcomes among pregnancies complicated by fetal growth restriction (FGR), maternal diabetes, or hypertension. FGR was defined as estimated fetal weight of <10% as determined by biometric ultrasound.¹⁰ Maternal diabetes included both preexisting disease diagnosed prior to pregnancy as well gestational diabetes. Hypertensive disorders included chronic hypertension, gestational hypertension, and preeclampsia. Oligohydramnios was identified after diagnosis on ultrasound by AFI of <5cm or DVP<2cm.^1,3 Resolved oligohydramnios was determined if either AFI or DVP normalized on subsequent ultrasound. Data from hospital discharge, identified by International Classification of Diseases, 9th Edition codes, were used to identify eligible patients. Trained research nurses performed detailed chart abstraction.

For this analysis, participants were included if they were diagnosed with oligohydramnios. The primary outcome was a composite variable of neonatal resuscitation at delivery, composed of administration of oxygen, bag and mask ventilation, or continuous positive airway pressure, as well as intubation, chest compression, or cardiac medication administration. All data were analyzed in SAS (Cary, NC). Dichotomous outcomes were compared using t-test or Fisher’s exact test. Continuous outcomes were compared using Wilcoxon rank-sum, and statistical significance was defined as p-value <0.05. A multivariable modified Poisson regression was performed to account for hypertension and pregestational diabetes, conditions that are known to increase the risk of oligohydramnios and FGR. Additionally, we controlled for gestational age at diagnosis of oligohydramnios.

Results

Among 11,916 women included in the parent study, a total of 527 women (4.4%) were diagnosed with oligohydramnios. Of these, 42 women had resolved oligohydramnios (8.0%), and 485 had persistent oligohydramnios. There were no significant differences in patient demographics or medical comorbidities between groups (Table 1).

Table 1.

Characteristics of patients with persistent and resolved oligohydramnios

Characteristics of persistent vs. resolved oligohydramnios
	Persistent oligohydramnios (N=485)	Resolved oligohydramnios (N=42)	p-Value
Gestational age at diagnosis of oligo
Median (IQR)	38.0 (36.4–39.3)	33.0 (29.1–35.9)	<0.001a
Maternal age at delivery
Median (IQR)	29.2 (23.9–34.3)	27.8 (23.9–32.1)	0.19a
Maternal race
Black	50 (10.3)	2 (4.8)	0.61b
White	250 (51.6)	25 (59.5)
Hispanic	135 (27.8)	10 (23.8)
Other	50 (10.3)	5 (11.9)
Maternal BMI
Median (IQR)	32.8 (28.2–37.9)	32.8 (28.1–37.6)	0.85a
Maternal medical comorbidities
Chronic hypertension	49 (10.1)	2/41 (4.9)	0.41b
Pregestational diabetes	17 (3.5)	1/41 (2.4)	1.00b
Renal disease	6 (1.2)	1 (2.4)	0.44b
Autoimmune disease	11 (2.3)	1 (2.4)	1.00b
Tobacco use	76 (15.7)	8 (19.1)	0.52b
Alcohol use	30 (6.2)	2 (4.8)	1.00b
Other substance use	14 (2.9)	0	0.62b

Abbreviations: BMI, body mass index; IQR, interquartile range; NICU, neonatal intensive care unit.

Note: Categorical data are (N%) unless otherwise noted.

^aWilcoxon rank-sum test.

^bFisher’s exact test.

In terms of the primary outcome, there was no difference in composite neonatal resuscitation (41 vs. 32%, p=0.31). There were also no differences in the individual components of the composite neonatal resuscitation outcome between groups (Table 2). The gestational age at diagnosis of oligohydramnios was significantly lower for patients with resolved versus persistent oligohydramnios (median: 33.0 weeks [interquartile range, IQR: 29.1–35.9] vs. 38.0 weeks [IQR: 36.4–39.3], p<0.001). Additionally, patients with resolved oligohydramnios were more likely to be diagnosed with FGR than those with persistent oligohydramnios (55 vs. 36%, p<0.02; Table 2).

Table 2.

Outcomes persistent and resolved oligohydramnios

	Persistent oligohydramnios (N=485)	Resolved oligohydramnios (N=42)	p-Value
Primary outcomes
Composite neonatal resuscitationa	157 (32.4)	17 (40.5)	0.31b
Oxygen	143/484 (29.6)	15 (35.7)	0.39b
Bag and mask with oxygen	47/483 (9.7)	2 (4.8)	0.41b
CPAP	26/483 (5.4)	1 (2.4)	0.71b
Intubation	19/484 (3.9)	1 (2.4)	1.00b
Chest compressions	8/484 (1.7)	1 (2.4)	0.53b
Cardiac medications	2/484 (0.41)	0 (0)	1.00b
Secondary neonatal outcomes
Betamethasone administration	50 (10.4)	6 (14.3)	0.43c
Fetal growth restriction	176 (36.3)	23 (54.8)	0.02b
Gestational age at delivery
Median (IQR)	38.3 (37.1–39.4)	39.0 (37.3–39.6)	0.45c
Birthweight (g)
Median (IQR)	2,870 (2,430–3,300)	2,860 (2,445–3,360)	0.86c
5-min Apgar <5	3 (0.57)	0	1.00b
NICU admission	110 (22.7)	7 (16.7)	0.44b
Diagnosis of RDS	33 (6.8)	1 (2.4)	0.50b
Neonatal death	3 (0.62)	0	1.00b
Secondary obstetric outcomes
Type of labor, N
Spontaneous	10 (2.1)	7 (16.7)	<0.001b
Spontaneous/augmented	21 (4.3)	4 (9.5)
Prelabor CS	123 (25.4)	9 (21.4)
Induction	331 (68.3)	22 (52.4)
Mode of delivery, N (%)	(n=479)	(n=42)	0.14b
Vaginal	229 (47.8)	25 (59.5)
Cesarean	229 (47.8)	14 (33.3)
OVD	21 (4.4)	3 (7.1)

Abbreviations: CPAP, continuous positive airway pressure; CS, cesarean section; IQR, interquartile range; NICU, neonatal intensive care unit; OVD, operative vaginal delivery; RDS, respiratory distress syndrome.

Note: Categorical data are (N%) unless otherwise noted.

^aIncludes oxygen, bag and mask with oxygen, CPAP, intubation, chest compressions, and cardiac medications.

^bFisher’s exact test.

^cWilcoxon rank-sum test.

Patients with persistent oligohydramnios were more likely to undergo induction of labor (68.3%) than patients with resolved oligohydramnios (52.4%, p<0.001), but there was no difference in mode of delivery between groups.

A multivariable Poisson regression was performed to account for gestational age at diagnosis as well as hypertensive and diabetic disorders and the results remained similar between groups (Table 3).

Table 3.

Adjusted risk ratio of composite neonatal resuscitation among patients with persistent oligohydramnios

	RR (95% CI) Persistent oligo vs. resolved (ref)	p-Value
Crude model (n=527)	0.80 (0.54–1.18)	0.26
Crude model, no missing covariates listed below (n=505)	0.79 (0.52–1.15)	0.21
Adjusted for GA at diagnosis (per wk)	1.16 (0.73–1.83)	0.14
Adjusted for chronic hypertension	0.78 (0.52–1.15)	0.21
Adjusted for pregestational diabetes	0.77 (0.52–1.15)	0.21
Adjusted for GA at diagnosis, chronic hypertension, and pregestational diabetes	1.15 (0.73–1.82)	0.55

Abbreviations: CI, confidence interval; GA, gestational age; RR, relative risk.

Notes: Resolved oligohydramnios served as the reference group for this analysis. Multivariable modified Poisson regression performed.

Discussion

In this cohort, resolution of oligohydramnios occurred in 8% of patients. There were no significant differences in primary outcome of need for neonatal resuscitation between patients with resolved versus persistent oligohydramnios. However, when compared with patients with persistent oligohydramnios, those with resolved oligohydramnios were diagnosed with oligohydramnios at an earlier gestational age and had higher associated rates of FGR. These findings suggest that women with resolved oligohydramnios should receive close prenatal surveillance prior to delivery even after amniotic fluid levels normalize.

Overall rates of NICU admission in this cohort were 2 to 3 times higher than the 6.8% reported for term infants without complicated pregnancies.¹¹ Prior research evaluating the outcomes of pregnancies complicated by FGR and oligohydramnios have shown rates of NICU admission up to 1.4 to 2.13 times as compared with those without oligohydramnios.^12,13 As amniotic fluid is crucial in the development of fetal lungs, this may account for why rates are of NICU admission so much higher in pregnancies complicated by oligohydramnios.

Strengths of our study include our population of socioeconomically diverse high-risk women as well as the extensive size of our database, which allowed for the study of uncommon outcomes. All data abstraction was performed by trained nursing staff and had multiple quality checks, which increases the internal validity of the data.

This study is limited in that the parent study from which this study originated included only participants with complicated pregnancies we are unable to assess truly isolated oligohydramnios. The fact that the parent study was restricted to high-risk pregnancies may reduce the generalizability of our findings to settings with predominantly low-risk cohorts; however, given oligohydramnios is more commonly diagnosed among high-risk pregnancies, these findings are still pertinent to many settings. Details on severity of oligohydramnios by level of AFI or DVP were not obtained for the original study cohort and are unable to be evaluated in this secondary analysis. Additionally, the proximity of the timing from diagnosis to delivery in the persistent oligohydramnios group may confound any association with neonatal outcomes and highlights the need for prospective studies. Timing of diagnosis of FGR is also not able to be assessed as the parent dataset only noted the presence of FGR and did not note age at diagnosis and therefore temporality between oligohydramnios and FGR cannot be assessed. Another limitation is the potential for type 2 error: despite the large size of the parent database, there were relatively small number of patients included in this secondary analysis. Indeed, a post hoc power calculation concluded we had <80% power to detect a difference between study groups. Lastly, like any retrospective study, causality cannot be established; as such, we can only identify an association between women with resolved oligohydramnios and development of FGR.

Conclusion

Patients whose oligohydramnios resolved were diagnosed at early gestational ages and had similar rates of neonatal resuscitation but higher rates of FGR than those who had persistent oligohydramnios. Further prospective research is necessary to evaluate this association.

Key Points.

• When diagnosed earlier in pregnancy, oligohydramnios was more likely to resolve prenatally.

• Patients who were diagnosed with oligohydramnios earlier in pregnancy had higher rates of FGR.

• There were no differences in the rates of the composite outcome of need for neonatal resuscitation when comparing those with resolved versus those with persistent oligohydramnios. No differences in composite neonatal morbidity were noted between those with resolved versus persistent oligohydramnios.