The amniotic fluid index (AFI) has been an integral component of fetal assessment during antepartum ultrasound examination for more than 20 years.1 Decreased amniotic fluid, or oligohydramnios, is typically defined as an AFI below 5 cm, which represents a value below the first percentile.2–5 In term and near-term gestations, this 5-cm threshold has been associated with increased rates of complications, including small for gestational age neonate, nonreassuring fetal heart rate, stillbirth, and neonatal death.3,6
The amniotic fluid volume is most abundant in the early third trimester, subsequently decreasing until term.2,4,7 Before 34 weeks, a value of 8 cm is below the fifth percentile for gestational age.4 AFI values between 5 and 8 cm have been termed “borderline.”8 Potential risks associated with borderline amniotic fluid in the preterm period are not fully understood. Gumus and colleagues recently reported that pregnancies with borderline AFI were at increased risk for preterm birth, fetal distress, and birth weight below the tenth percentile.9 Our objective was to better characterize risks associated with AFI of 8 cm or less in a large cohort of preterm pregnancies, comparing outcomes in such pregnancies to those with normal amniotic fluid volume.
MATERIALS AND METHODS
This was a review of consecutive singleton pregnancies that received ultrasound examinations at 24 0/7 through 33 6/7 weeks of gestation from August 1997 to March 2008 and were subsequently delivered of liveborn neonates or stillborn fetuses at our hospital. We routinely perform AFI measurements at this gestational age range using the method described by Phelan and colleagues.1 Pregnancies with AFI 5.0 cm or less were diagnosed with oligohydramnios. For study purposes, pregnancies with AFI 5.1 to 8.0 cm were considered to have borderline amniotic fluid volume. However, this term was not used for clinical decision-making during the study period, nor did sonography reports describe the amniotic fluid volume as abnormal. Pregnancies with AFI 8.1 to 24 cm were considered to have normal amniotic fluid volume. It is our standard practice to use color Doppler to identify intervening loops of umbilical cord when measuring the AFI. We excluded pregnancies with hydramnios, defined as AFI greater than 24 cm, those with ruptured membranes at the time of ultrasound examination, and those in which the fetus was not alive at the time of the ultrasound examination. If more than one ultrasound examination was performed, the lowest AFI was used for analysis.
Standard ultrasound examinations were performed by registered diagnostic medical ultrasonographers, either at our hospital-based central ultrasound unit or within our neighborhood prenatal clinics, with images transmitted electronically to our central unit and overread by faculty who specialize in obstetric ultrasonography. This is done in accordance with American Institute of Ultrasound in Medicine Practice Accreditation Standards. All of our ultrasound sites are American Institute of Ultrasound in Medicine-accredited.
Pregnancy outcomes were obtained by merging our ultrasound database with an obstetric database that is prospectively maintained for quality assurance and includes neonatal outcomes for all women delivered at our hospital. In this database, major anomalies are defined as those that are potentially life-threatening or require surgery. Outcomes for neonates with major anomalies were analyzed separately. Fetal growth restriction was based on birth weight percentiles derived from our population, adjusted for maternal race or ethnicity and neonatal sex.10
Statistical analyses were performed with SAS 9.2. Group comparisons were made with the Pearson χ2 tests for contingency tables and analysis of variance for comparison of means. Logistic regression was used to adjust for gestational age at delivery where appropriate. P<.05 was considered statistically significant. This study was approved by the Institutional Review Board of the University of Texas Southwestern Medical Center.
There were 28,555 women with singleton pregnancies who received ultrasound examinations between 24 0/7 and 33 6/7 weeks during the study period and were subsequently delivered at our hospital. The mean gestational age at ultrasound examination was 29.2 plus or minus 2.9 weeks (mean gestational age plus or minus standard deviation). The indication for the examination was either estimation of gestational age or evaluation of fetal growth in 22,315 women (78%). Other indications included hypertension, diabetes, fetal abnormality, placenta previa, and preterm labor. The gestational age at the ultrasound examination was slightly but significantly greater in pregnancies with oligohydramnios and borderline AFI than in pregnancies with normal AFI, 30.1 and 30.5 weeks compared with 29.2 weeks, respectively, P<.001.
Maternal demographic characteristics are presented in Table 1 according to AFI category. Women with oligohydramnios were older (1 to 2 years) than those with borderline AFI or normal fluid, P=.015. There were no significant differences between AFI groups with respect to maternal race or ethnicity. Pregnancy hypertension and diabetes were each more prevalent in pregnancies with oligohydramnios than in those with normal AFI, both P<.001, with no significant differences between the borderline and normal AFI groups. Stillbirths were more common in pregnancies complicated by oligohydramnios (5%) than with either borderline AFI or normal fluid (less than 1% in each group), P<.001. Major malformations were significantly more common in the setting of oligohydramnios or borderline AFI than with normal fluid, 25% compared with 10% compared with 2%, respectively, P<.001. Subsequent analyses were restricted to pregnancies with nonanomalous liveborn neonates.
Delivery data and intrapartum complications are presented in Table 2. Preterm births at or before 36 weeks, 34 weeks, and 32 weeks were all more common in the oligohydramnios and borderline AFI groups than in the cohort with normal fluid, all P<.001. Preterm births were further analyzed according to whether the delivery was indicated, for example, labor induced for a maternal or fetal complication, or whether it resulted from spontaneous preterm labor. Overall, 80% of preterm births resulted from spontaneous preterm labor, and 20% were indicated. Both types of preterm births were significantly more common in pregnancies with oligohydramnios and borderline AFI than in those with normal amniotic fluid, all P<.001. Indicated preterm births made up a larger proportion of the total number of preterm births in the oligohydramnios and borderline AFI cohorts than in those with normal fluid, 32% compared with 36% compared with 19%, respectively, P<.001. In 2% of the pregnancies (425/27,717), preterm labor developed before the ultrasound examination. After exclusion of these pregnancies and reanalysis, the rate of spontaneous preterm birth remained significantly greater in the oligohydramnios and borderline AFI groups than in pregnancies with normal AFI, 61 of 110 (61%) and 63 of 175 (36%) compared with 1,900 of 26,947 (7%), respectively, P<.001. Cesarean deliveries, both overall and those performed for nonreassuring fetal status, were significantly more common in pregnancies with oligohydramnios and borderline amniotic fluid volume than in those with normal AFI, both P<.001.
Fetal growth restriction, defined as birth weight at or below the third percentile, was 5 times more common in the setting of borderline AFI (odds ratio [OR] 5.2, 95% confidence interval [CI]: 3.9, 6.9) and 9 times more common in the setting of oligohydramnios (OR 9.0, 95% CI: 7.0, 11.5) than in pregnancies with normal AFI, P<.001. Birth weight at or below the tenth percentile was also more prevalent in pregnancies with oligohydramnios or borderline AFI than in the normal AFI group, P<.001. Neonatal death was more common in pregnancies with oligohydramnios and borderline AFI than with normal AFI in initial analyses; however, this finding was no longer significant after adjusting for gestational age at delivery, adjusted P=.065.
There are three main findings from this study of the relationship between amniotic fluid volume and pregnancy outcome in preterm gestations. First, major fetal malformations were much more common in pregnancies with decreased amniotic fluid volume, with a 5-fold increase in prevalence with borderline AFI and a more than 10-fold increase with oligohydramnios. After exclusion of pregnancies with malformations, we also found that preterm births, both indicated and those resulting from spontaneous preterm labor, were significantly increased in women with AFI of 8 cm or less, which includes pregnancies considered to have oligohydramnios as well as those with “borderline” amniotic fluid. Preterm births remained greatly increased after excluding pregnancies in which the woman had preterm labor before her ultrasound examination. Finally, the prevalence of fetal growth restriction, based on birth weight at or below the third percentile for gestational age, was markedly increased in pregnancies with borderline AFI (5-fold increase) and oligohydramnios (9-fold increase).
Other investigators have also noted an increase in small for gestational age neonates and preterm birth when AFI is decreased. Gumus and colleagues studied 90 pregnancies with AFI between 5 and 10 cm identified during the third trimester and found an approximately 2-fold increased risk for preterm birth, intrapartum fetal distress, and birth weight below the tenth percentile, compared with pregnancies with AFI greater than 10 cm.9 Banks and Miller reviewed 75 pregnancies with AFI between 5 and 10 cm identified within 1 week of delivery and found increased likelihood of birth weight below the tenth percentile compared with pregnancies with larger AFI, 13% compared with 4%.11 Similarly, Kwon and colleagues studied 217 pregnancies found to have an AFI between 5 and 8 cm within 2 weeks of delivery and reported increased rates of small for gestational neonates (31% compared with 12%), cesarean delivery performed for nonreassuring fetal testing (6% compared with 2%), and earlier gestational age at delivery (36 weeks compared with 38 weeks) compared with AFI greater than 8 cm.12 Although these studies had findings similar to ours, they differed in two important ways: they were not restricted to preterm gestations, and the latter two studies included only pregnancies shortly before delivery.
Strengths of this series are the large number of pregnancies, more than 25,000, the use of normograms derived from our population to ascertain growth restriction, and the fact that pregnancies with AFI between 5 and 8 cm were not diagnosed with an amniotic fluid abnormality or managed according to the AFI value during the study period. One limitation is that all ultrasound examinations were performed for specific indications, such as evaluation of size-date discrepancies, a practice consistent with national guidelines which have not called for routine sonography at this gestational age range.13 Thus our findings do not represent a population-based AFI distribution, and it is likely that pregnancies with complications are overrepresented. However, our findings would apply to the practitioner performing an indicated ultrasound examination who unexpectedly indentifies oligohydramnios or an AFI between 5 and 8 cm.
Most clinicians would agree that AFI values less than 5, usually described as “oligohydramnios,” require further evaluation and possibly intervention. Given our findings that AFI values between 5 and 8 cm, usually described as “borderline,” are also associated with pregnancy complications, what is the clinician to do when encountering a “borderline” AFI in a preterm gestation? Our interpretation in our practice is that because we could not associate increased mortality nor morbidity with borderline AFI values compared with normal indices, in the absence of suspected fetal growth restriction, that intervention is not warranted for borderline values. Whether or not interventions such as antepartum testing of fetal condition in the absence of growth restriction might be of benefit can be addressed only by appropriate prospective studies. Nonetheless, our findings of increased preterm birth and fetal growth restriction when borderline AFI is encountered in preterm gestations suggest that it may be necessary to redefine what is considered an abnormal AFI during this gestational period.
1. Phelan JP, Smith CV, Broussard P, Small M. Amniotic fluid volume assessment with the four-quadrant technique at 36–42 weeks' gestation. J Reprod Med 1987;32:540–2.
2. Rutherford SE, Phelan JP, Smith CV, Jacobs N. The four-quadrant assessment of amniotic fluid volume: an adjunct to antepartum fetal heart rate testing. Obstet Gynecol 1987;70(3 pt 1):353–6.
3. Casey BM, McIntire DD, Bloom SL, Lucas MJ, Santos R, Twickler DM, et al. Pregnancy outcomes after antepartum diagnosis of oligohydramnios at or beyond 34 weeks' gestation. Am J Obstet Gynecol 2000;182:909–12.
4. Moore TR, Cayle JE. The amniotic fluid index in normal human pregnancy. Am J Obstet Gynecol 1990;162:1168–73.
5. Callen PW. Amniotic fluid volume: its role in fetal health and disease. In: Callen PW, editor. Ultrasonography in obstetrics and gynecology. 5th ed. Philadelphia (PA): Saunders-Elsevier; 2008:762–4.
6. Locatelli A, Vergani P, Toso L, Verderio M, Pezzullo JC, Ghidini A. Perinatal outcome associated with oligohydramnios in uncomplicated term pregnancies. Arch Gynecol Obstet 2004;269:130–3.
7. Magann E, Sanderson M, Martin J, Chauhan S. The amniotic fluid index, single deepest pocket, and two-diameter pocket in normal human pregnancy. Am J Obstet Gynecol 2000;182:1581–8.
8. Moore TR. Clinical assessment of amniotic fluid. Clin Obstet Gynecol 1997;40:303–13.
9. Gumus II, Koktener A, Turhan NO. Perinatal outcomes of pregnancies with borderline amniotic fluid index. Arch Gynecol Obstet 2007;276:17–9.
10. McIntire DD, Bloom SL, Casey BM, Leveno KJ. Birth weight in relation to morbidity and mortality among newborn infants. N Engl J Med 1999;340:1234–8.
11. Banks EH, Mill er DA. Perinatal risks associated with borderline amniotic fluid index. Am J Obstet Gynecol 1999;180(6 pt 1):1461–3.
12. Kwon JY, Kwon HS, Kim YH, Park YW. Abnormal Doppler velocimetry is related to adverse perinatal outcome for borderline amniotic fluid index during third trimester. J Obstet Gynaecol Res 2006;32:545–9.
13. Ultrasonography in Pregnancy. ACOG Practice Bulletin No. 101. American College of Obstetricians and Gynecologists. Obstet Gynecol 2009;113(2 pt 1):451–61.
Figure. No caption available.