OBJECTIVE: To investigate the etiology of oligohydramnios in postterm pregnancy using Doppler velocimetry.
METHODS: Renal and umbilical artery Doppler velocimetry were performed in women with singleton postterm (287 days' or more gestation) pregnancies. The renal and umbilical artery Doppler resistance index (RI) and end‐diastolic velocity were measured. Stepwise logistic regression and the two‐tailed t test were used to determine whether the Doppler indices correlated with oligohydramnios (amniotic fluid index less than 5 cm).
RESULTS: We studied 147 well‐dated, singleton, postterm pregnancies, of which 21 (14.3%) had oligohydramnios. For the study cohort, the mean (±standard deviation) gestational age at Doppler was 41.4 ± 0.45 weeks and at delivery 41.8 ± 0.47 weeks. Stepwise logistic regression using renal and umbilical artery Doppler indices found the renal RI to be the only significant predictor of oligohydramnios: β = −10.4186, P < .05 (odds ratio [95% confidence interval (CI)] = 0, 0.88). The renal artery RI was significantly higher in cases with oligohydramnios (RI: mean (±standard error) = 0.8843 ± 0.11 versus 0.8601 ± 0.05, P ≤ 0.05). A renal artery Doppler end‐diastolic velocity below the mean for gestation significantly increased the risk of oligohydramnios: relative risk (95% CI), 1.5 (1.1, 2.0).
CONCLUSION: Renal artery Doppler was more predictive of oligohydramnios than the umbilical RI. The reduced renal artery end‐diastolic velocity suggests that increased arterial impedance is an important factor in the development of oligohydramnios in prolonged pregnancies.
Elevated fetal renal artery Doppler resistance index due to reduced end&#x2010;diastolic velocity suggests that reduced renal perfusion is the cause of oligohydramnios in prolonged pregnancy.
Department of Obstetrics and Gynecology, Mersin University, Mersin, Turkey; Department of Obstetrics and Gynecology, Yale University School of Medicine, New Haven, Connecticut; and Department of Obstetrics and Gynecology, University of Cincinnati, Cincinnati, Ohio.
Address reprint requests to: Ray O. Bahado‐Singh, MD, University of Cincinnati, Department of Obstetrics and Gynecology, 231 Albert Sabin Way, ML 0526, Cincinnati, OH 45267; E‐mail: email@example.com
Received October 16, 2001. Received in revised form April 25, 2002. Accepted May 13, 2002.
Prolonged gestation remains a significant issue in contemporary perinatal practice, occuring in 3–12% of pregnancies, with a recurrence risk of 50%.1 Postmaturity syndrome, which occurs in prolonged gestation, has long been known to elevate the risk for perinatal complications,2 such as meconium aspiration syndrome and perinatal asphyxia.
Antepartum surveillance is widely used in the management of these pregnancies.1 An important component of such biophysical monitoring is the ultrasound assessment of amniotic fluid volume. Oligohydramnios is a strong predictor of adverse outcome in prolonged gestation, and monitoring fluid volume has been reported to significantly reduce perinatal complications.3 Despite the significance of amniotic fluid volume in postterm assessment and perinatal outcome, however, the mechanism of the development of oligohydramnios remains uncertain and controversial. Some have proposed redistribution of blood flow away from the fetal kidney4 or increased renal tubular sensitivity to arginine vasopressin5 as the etiology of oligohydramnios in prolonged pregnancy.
The purpose of the current study was to determine whether there was a statistically significant relationship between renal artery blood flow, as represented by the Doppler indices, and reduced amniotic fluid volume in postterm pregnancy.
MATERIALS AND METHODS
Institutional Review Board approval and verbal consent were obtained for the performance of the study. Enrollees were patients presenting to the Antepartum Testing Unit at the Department of Obstetrics and Gynecology at Yale University School of Medicine between January 29, 1995, and December 31, 1996. All women carried a singleton pregnancy and underwent clinically indicated biophysical testing due to postterm status (≥287 days' or more gestation). The components of the biophysical profile included fetal breathing movement, nonstress test, fetal movement, tone, and amniotic fluid volume as described by Manning et al.6
The length of gestation was determined based on precisely recalled menstrual dates or a first trimester ultrasound or an early second trimester ultrasound performed prior to 20 weeks. We excluded patients with medical complications that are known to affect fetal growth, such as diabetes, chronic hypertension, or other vascular disorders, along with those with imprecise dates or the diagnosis of fetal growth restriction as an indication for testing. Further, only singleton nonanomalous cases were included in the study. Patients who declined consent were not included.
The amniotic fluid volume was estimated by a sonographer using the amniotic fluid index (AFI)7 with a value of less than 5 cm, and clinical observation was used to define oligohydramnios. Patients were recruited by a perinatologist (RB) or the antepartum testing nurses. Final selection of patients was performed by a perinatologist (RB).
Doppler velocimetry was performed by a single investigator (RB). Doppler measurements were performed either before or after the amniotic fluid volume estimation. The patient was placed in the left lateral position, and an axial image of the fetal abdomen was obtained at the level of the fetal kidneys. Using color flow Doppler, the renal artery was identified as it approached the kidney from the aorta. A straight segment of the vessel was identified and the Doppler gate placed within the lumen of the vessel. We attempted to obtain an angle of insonation as close as possible to 0°. A minimum of three consecutive wave forms was used to calculate the resistance index (RI) ([systolic‐diastolic] systolic). The wall filter was set at 50 Hz to preserve the end‐diastolic component of the wave form. Measurements were performed with a 3.5‐MHz or 5.0‐mHz probe using the Acuson XP‐128 (Mountain View, CA), and the ATL Ultramark 9 (Bothell, WA). Measurements were performed during periods of fetal apnea and inactivity.
Umbilical artery Doppler velocimetry was also performed, and the RI was measured. Neither the renal nor umbilical artery Doppler velocimetry results were shared with the managing obstetrician.
The mean Doppler velocity is known to correlate significantly with blood volume flow. We therefore evaluated the correlation between this index and amniotic fluid index.
Pregnancy and neonatal outcome were ascertained after delivery by reviewing the appropriate medical records. Adverse outcomes were defined as one or more of the following: cesarean delivery for fetal distress, 5‐minute Apgar score less than 7, neonatal intensive care unit stay greater than 24 hours, or perinatal death.
The groups were compared using the two‐tailed t test. Normality of distribution of the Doppler indices was confirmed using the Kolmogorov‐Smirnov test. The Spearman correlation coefficient was used to determine whether the Doppler indices significantly correlated with gestational age within the gestational age range of the study patients. Stepwise multiple logistic regression was performed to determine which Doppler index significantly predicted oligohydramnios. In addition, a boxplot of renal artery RI values in the groups with and without oligohydramnios was constructed. P < .05 was considered to be statistically significant.
There was a total of 147 study cases. The mean (±standard deviation [SD]) maternal age was 28.0 ± 6.4 years. Forty‐five percent of the study cases were parous. Of these, 35% were primiparous, and the rest were multiparous. The mean (±SD) gestational age at Doppler were 41.4 ± 0.45 weeks. One hundred and twenty‐three (83.7%) of the patients were from 41 to 41 6/7 weeks at Doppler, and 24 (16.3%) were 42 weeks or more. The mean (± SD) gestational age at delivery was 41.8 ± 0.47 weeks. Mean (± SD) birth weight was 3629.7 ± 468.8 g.
Sixty‐five percent of the patients were referred from private obstetricians, whereas the remainder were referred from the high‐risk and residents clinics. No specific index of socioeconomic status was evaluated; however, the study group was thought to be similar to the hospital's overall delivery population, which is approximately 70% middle class, with the remainder being lower‐middle class and indigent patients.
There were no cases with absent or reversed umbilical end‐diastolic velocity. There were a total of 23 adverse outcomes, with no perinatal deaths.
There were 21 pregnancies (14.3%) with oligohydramnios. There was no significant difference in the number of primigravidas in oligohydramnios versus normal amniotic fluid groups (9/21 versus 73/124, respectively, P = .7). The mean gestational age at Doppler study was not significantly different between those with low and normal fluid volumes (41.3 ± 0.46 weeks versus 41.4 ± 0.44, respectively, P = .2), whereas the age of delivery was (41.4 ± 0.47 weeks versus 41.8 ± 0.45 weeks, P = .002). The women with oligohydramnios also had neonates with significantly lower gestational age at delivery and birth weight than others (3414.0 ± 455.5 g compared to 3665.7 ± 463.0 g, P < .03).
Stepwise logistic regression (using renal artery RI, peak systolic and end‐diastolic velocities, and umbilical artery RI as the independent variables) found that the renal artery RI was the only significant predictor of oligohydramnios: β = −10.4186, P < .05, (odds ratio [95% confidence interval (CI)] = 0, 0.876). The renal artery RI was significantly higher in the group with oligohydramnios (RI mean [± standard error] = 0.8843 ± 0.11 versus 0.8601 ± 0.05, P < .05, respectively) (Figure 1).
Within the study population, there was no significant correlation between the renal artery Doppler velocity measurements or the RI and gestational age.
A renal artery end‐diastolic velocity below the mean significantly increased the risk of oligohydramnios: relative risk (95% CI) 1.5 (1.1, 2.0). The presence of oligohydramnios (AFI < 5 cm) significantly increased the risk of adverse outcome compared with normal fluid volume: 7/16 (43.7%) versus 16/85 (18.8%), χ2 = 4.76, P < .05. Finally, the mean Doppler velocity had a statistically significant positive correlation with the amniotic fluid index (r = .21, P = .04). The renal artery RI was not significantly different in cases with and without perinatal complications: 0.856 multiples of the median (MoM) versus 0.868 MoM (P = .19).
Using a larger number of study cases than previously reported, and by examining the primary (peak systolic velocity and end‐diastolic) Doppler velocities and derived (resistance index) Doppler indices, we were able to demonstrate a statistically significant increase in the renal artery RI in postdates pregnancies with oligohydramnios. Our data also indicate that the basis of the increase in the RI is a reduction in the fetal renal artery end‐diastolic velocity. A reduction of the Doppler end‐diastolic velocity below the mean was associated with a statistically significant increase in the risk of having oligohydramnios.
There are several potential mechanisms by which changes in the kidney can cause oligohydramnios in prolonged pregnancies. One is fetal hypovolemia and hemoconcentration due to changes in the placental fetal fluid balance, as proposed by Rightmire and Campbell.8 Bar‐Hava et al5 considered oligohydramnios to be the result of excessive tubular reabsorption of urine due to above‐normal sensitivity to vasopressin in prolonged pregnancies.5 Another possibility is a localized increase in impedance in the renal vessels, which can be measured using Doppler velocimetry. The study results reported here appear consistent with the latter mechanism. Our findings support those of Veille et al, who found elevated renal artery systolic/diastolic ratios in postdates pregnancies with oligohydramnios.4
On the other hand, our study is at variance with that of Bar‐Hava et al,5 who found no changes in the umbilical, middle cerebral, or renal artery Doppler velocities in oligohydramnios cases despite a reduction in the average birth weight of the group. Their findings agree with the report of Battaglia et al.9 The latter group, however, did find a significant reduction of descending aorta mean blood velocity in oligohydramnios cases. This indicates that circulatory adjustments were caused by or associated with oligohydramnios in prolonged gestation.
The finding of reduced birth weight in the oligohydramnios cases, along with increased renal artery resistance index, indicates a similarity in the pathophysiology to intrauterine growth restriction. Reduced fetal growth is a well‐recognized complication in postmaturity syndrome, and is thought to be due to placental senescence. Rightmire10 analyzed the published literature and compared the screening performance of several different antepartum testing modalities, namely, umbilical artery Doppler, electronic fetal heart rate monitoring, and amniotic fluid volume for predicting outcome in postdates pregnancies. Doppler studies had the best statistical performance. This provides further evidence that significant hemodynamic changes are a feature of such pregnancies. A potential confounding variable in such studies is the inclusion of cases with fetal growth impairment due to other causes apart from postmaturity. In the current study, we eliminated cases with underlying maternal disorders known to be associated with fetal growth restriction, thus reducing the potential for confusing the cause of the Doppler changes.
The Doppler changes described were subtle and seen in only a minority of the postterm pregnancies. This finding could explain the failure to detect such changes in the preceding studies. More important than the small number of postdates cases enlisted in the preceding studies is the even smaller number of oligohydramnios cases, further reducing the likelihood of detecting differences in the renal Doppler indices. Another possibility is that the studies that found no significant difference between groups were more effective in eliminating patients with conditions that can cause fetal growth restriction (apart from postdatism) and therefore a more appropriate study population to assess the isolated effect of postdatism. We do not believe this to be the case, because we eliminated pregnancies with other recognized causes of fetal growth restriction in our study cohort. Moreover, in general clinical practice, few patients with recognizable significant underlying disorders that lead to growth restriction are allowed to remain undelivered by 41 weeks. Finally, the fact that the renal artery and not umbilical Doppler velocimetry significantly predicted oligohydramnios suggests that the etiology of the oligohydramnios may reside at the renal more than the placental level.
As noted above, the changes observed on Doppler measurements were subtle and might not therefore be clinically significant or useful for the routine evaluation of prolonged pregnancies.
1. American College of Obstetricians and Gynecologists. The management of post-term pregnancy. ACOG practice patterns no. 6. Washington, DC: American College of Obstetricians and Gynecologists, 1997.
2. Clifford SH. Postmaturity—with placental dysfunction. J Pediatr 1954;41:1:1–13.
3. Johnson JM, Harman CR, Lange IR, Manning FA. Biophysical profile scoring in the management of the postterm pregnancy. Am J Obstet Gynecol 1986;154:269–73.
4. Veille JC, Penry M, Mueller-Heubach E. Fetal renal pulsed Doppler waveform in prolonged pregnancies. Am J Obstet Gynecol 1993;169:882–4.
5. Bar-Hava I, Divon MY, Sardo M, Barnhard Y. Is oligohydramnios in post-term pregnancy associated with redistribution of fetal blood flow? Am J Obstet Gynecol 1995;173:519–22.
6. Manning FA, Morrison I, Harman CR, Lange IR, Menticoglu S. Fetal assessment based on fetal biophysical profile scoring. Experience in 19,221 referred high-risk pregnancies: II. An analysis of false-negative fetal deaths. Am J Obstet Gynecol 1987;157:880–4.
7. Moore TR, Cayle JE. Amniotic fluid index in normal human pregnancy. Am J Obstet Gynecol 1990;162:1168–73.
8. Rightmire DA, Campbell S. Fetal and maternal Doppler blood flow parameters in post-term pregnancies. Obstet Gynecol 1987;69:891–4.
9. Battaglia C, Larocca E, Lanzani A, Coukos G, Genazzani R. Doppler velocimetry in prolonged pregnancy. Obstet Gynecol 1991;77:213–6.
10. Rightmire DA. Doppler velocimetry for post-dated pregnancies. In: Maulik D, ed. Doppler ultrasound in obstetrics and gynecology. New York: Springer-Verlag, 1997: 337–48.