Elevated maternal serum AFP and pregnancy outcome
Volume 167 I'\umber 6
10. Varner MW, Hauser KS, Shaw K. Current trends in liveborn singleton birth weight. J Iowa Med Soc 1983;73: 447-50. 11. Brumfield CG, Cloud GA, Davis RO, Finley SC, Hauth JL, Boots L. The relationship between maternal serum and amniotic fluid a-fetoprotein in women undergoing early amniocentesis. AM J OBSTET GYNECOL 1990; 163:903-5. 12. Barford DA, Dickerman LH, Johnson WE. a-Fetoprotein: relationship between maternal serum and amniotic fluid levels. AMJ OBSTET GYNECOL 1985;151:1038-41. 13. Gill 1J, Repetti CF. Immunologic and genetic factors influencing reproduction. Am J Pathol 1979;95:465-570. 14. Greenberg F, Faucett WA, Rose E, et al. Congenital deficiency of alpha-fetoprotein. Pediatr Res 1989;25:52A. 15. Oblekwe BC, Malek N, Kitau MJ, Chard T. Maternal and fetal alpha fetoprotein (AFP) levels at term. Acta Obstet Gynecol Scand 1985;64:251-3. 16. Hubinont C, Fisk NM, Nicolini U, Rodeck CH, Johnson
17.
18.
19. 20.
RD. Fetal alpha-fetoprotein concentration in growth retardation. Br.J Obstet Gynecol 1989;96:1233-4. Caballero G, Vekemans M, Lopez del Campo J C, Robyn C. Serum a-fetoprotein in adults, in women during pregnancy, in children at birth, and during the first week of life: a sex difference. AM J OBSTET GYNECOI. 1977; 127: 384-9. Zeltzer PM, Neerhont RC, Fonkalsrud EW, Stiehm ER. Differentiation between neonatal hepatitis and biliary atresia by measuring serum alpha-fetoprotein. Lancet 1974; 1: 373-5. Katz VL, Cefalo RC, McCune BK, Moos MK. Elevated second trimester maternal serum alpha-fetoprotein and cytomegalovirus infection. Obstet Gynecol 1986;68:580-1. Hay DL, Barrie JV, Davison GB, et al. The relation between maternal serum alpha fetoprotein levels and fetomaternal haemorrhage. Br J Obstet Gynaecol 1979;86: 516-20.
Measurement of amniotic fluid volume: Accuracy of ultrasonography techniques Everett F. Magann, MD; Thomas E. Nolan, MD,b L. Wayne Hess, MD; Rick W. Martin, MD; Neil S. Whitworth, PhD; and John C. Morrison, MD
jackson, Mississippi, and Augusta, Georgia OBJECTIVE: Our purpose was to determine amniotic fluid volume by the dye-dilution technique and compare it with the amniotic fluid index, largest vertical pocket, and two-diameter pocket (defined as vertical x horizontal of the largest vertical pocket). STUDY DESIGN: This prospective study involved 40 women undergoing amniocentesis in late pregnancy to detect fetal lung maturity or evidence of chorioamnionitis. The amniotic fluid volume was quantified ultrasonographically by means of the amniotic fluid index, largest vertical pocket, and two-diameter pocket. During amniocentesis the fluid volume was calculated by the dye-dilution technique of Charles and Jacoby. RESULTS: Ultrasonographic measurements by amniotic fluid index, largest vertical pocket, and two-diameter pocket correctly predicted normal amniotic fluid and hydramnios (74%). A new measurement, two-diameter pocket, gave a significantly more accurate estimate of oligohydramnios than did amniotic fluid index (p < 0.002) or largest vertical pocket (p < 0.0003). CONCLUSION: All three indices are moderately accurate in identifying normal amniotic fluid volume and hydramnios. Two-diameter pocket is the most accurate test to predict oligohydramnios. (AM J OBSTET GYNECOL 1992;167:1533-7.)
Key words: Amniotic fluid, oligohydramnios, ultrasonography
From the Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, University of MississiPPi Medical Center: and the Department of Obstetrics and Gynecology, Medical College of Georgia.' Supported in part by the Vicksburg Hospital Medical Foundation. Presented in part at the Thirty-ninth Annual Meeting of the Society for Gynecologic Investigation, San Antonio, Texas, March 18-21, 1992. Reprint requests: Everett F. Magann, MD, Department of Obstetrics and Gynecology, University of Mississippi Medical Center, 2500 N. State St., Jackson, MS 39216-4505. 6/6/41725
Documentation of abnormalities of amniotic fluid volume may provide valuable information to enhance fetal health assessment. Decreased amounts of amniotic fluid, particularly in the third trimester, have been associated with multiple fetal risks, including death, pulmonary hypoplasia, growth retardation, and other complications. ,·3 In contrast increased amounts of amniotic fluid are correlated with fetal anomalies, twins, and insulin-dependent diabetes'" 5 A simple, inexpen-
1534 Magann et al.
sive, clinically useful test that will accurately determine amniotic fluid volume is needed. Unfortunately, even under research conditions techniques to confirm this measurement late in pregnancy are lacking. Ultrasonography is noninvasive and clinically can quantitate the amniotic fluid volume. It can be used to evaluate amniotic fluid volume by measurement of a single pocket of amniotic fluid or by means of a semiquantitative four-quadrant technique." 6, 7 Volume assessment by these techniques can be performed quickly, is easily taught, and is reproducible. 7 However, late in pregnancy, when it is most needed clinically, the error rate of ultrasonography is at its highest. Dye-dilution methods can be used to determine the amount of amniotic fluid in the same way the technology is used in calculating cardiac output." Amniotic fluid volume, as measured by a dye-dilution technique, has been correlated with the amniotic fluid as determined by ultrasonography in women undergoing second-trimester abortions. 9 Currently these techniques are the most accurate methods used to measure amniotic fluid volume. However, they are time-consuming, cumbersome, and invasive. Unfortunately, correlation of a single pocket of amniotic fluid or the amniotic fluid index with the actual volume of fluid as determined by dye-dilution techniques in the third trimester of pregnancy has not been accomplished. The purpose of this study is to correlate the amount of amniotic fluid as measured by ultrasonographic assessment with the actual volume determined by dyedilution techniques in late pregnancy. In addition, we calculated the pocket volume as a new measurement in the evaluation of amniotic fluid volume. Material and methods
This study comprised 40 women in the third trimester of pregnancy undergoing amniocentesis for the evaluation of fetal lung maturity or to detect subclinical chorioamnionitis in amniotic fluid. Inclusion criteria were as follows: (1) pregnancy between 28 and 37 weeks' gestation, (2) informed consent, (3) intact membranes, not in labor, and (4) maternal age> 15 years. Exclusion criteria were (1) lack of desire to participate in this study, (2) pregnancies < 28 weeks' gestation, and (3) parturients < 15 years old. Patients who agreed to participate in the study were then asked to sign a consent form approved by the Investigational Review Board of the University of Mississippi Medical Center. Ultrasonography was performed in all cases by the same physician with an Ultramark-4 device (Advanced Technology Laboratories, Bothell, Wash.). The amniotic fluid volume was assessed by means of the largest pocket method, as described by Chamberlain et aI.' In this technique the vertical and transverse diameters of the largest pocket of amniotic fluid were measured and
December 1992 Am J Obstet Gynecol
recorded. The depth of the pocket was measured at a right angle to the uterine contour. A depth of < 1 cm was classified as decreased, 1.1 to 2 cm as marginal (0 to 2 cm as oligohydramnios), 2.1 but 8.1 cm as hydramnios. Amniotic fluid volume was also determined by the two-diameter pocket. This volume consists of the vertical dimensions multiplied by the horizontal diameter of the largest pocket. The two-diameter pocket is a new measurement and a variation of the largest vertical pocket technique. The two-diameter pocket measurements corresponding to oligohydramnios, normal fluid, and hydramnios were compared with the actual fluid volume. These values were then compared with amniotic fluid index values on a receiver-operator characteristic curve and the ranges for each fluid volume were determined. A two-diameter pocket of 0 to 15 cm was oligohydramnios, 15.1 to 50 cm was normal fluid, and > 50.1 cm was hydramnios. The amniotic fluid volume was also assessed by the four-quadrant technique as described by Phelan. 6 The uterus was divided into four quadrants: by the umbilicus transversely (into upper and lower halves) and by the linea nigra vertically (into right and left halves). The maximum vertical diameter of the largest pocket in each quadrant without an aggregate of cord or fetal extremities was measured in centimeters. A depth of 0 to 5 cm equates oligohydramnios, 5.1 to 8 cm low, 8.1 to 18 cm normal, and > 18 cm high. Mter assessment of amniotic fluid volume by ultrasonography, amniocentesis was performed. Mter amniotic fluid was obtained for clinical studies, 2 ml of a 20% aqueous solution of aminohippurate sodium (400 mg, Merck, West Point, Pa.) was injected into the amniotic cavity. The amniocentesis needle was left in the amniotic cavity and continuous ultrasonographic monitoring of needle placement and fetal position was performed. The plunger of the syringe was withdrawn and depressed several times to promote rapid mixing of aminohippurate sodium with amniotic fluid. Three milliliters of amniotic fluid was collected 20 to 30 minutes after injection of aminohippurate sodium. It has been our experience, and also has been reported by others, 8 that optimal mixing of amniotic fluid has occurred by this time. The samples were frozen at - 20° C until assayed for aminohippurate sodium concentrations. Aminohippurate sodium concentrations and amniotic fluid volumes were determined with the diazo dye-reaction and spectrophotometric technique of Charles and Jacoby."' 10 The intraassay and interassay coefficient of variation was 1.9% and 3.5%, respectively. Serial dilutions of samples yielded aminohippurate sodium concentrations that were 101.8% of the expected value, and aminohippurate sodium added to a pool of amniotic fluid was recovered at a rate of 110.0%. The data were evaluated by X2 analysis. Between-
Amniotic fluid volume
Volume 167 J',;umhcr 6
1535
Table I. Demographic characteristics Race (yr)
Gestational age (wk)
Gravidity
White
23.8 ± 5.7
34.3 ± 2.1
1.9 ± 1.2
5
Age
No. 40
I
Black
Native American
I
31
4
Table II. Amniotic fluid volume test performance characteristics Sensitivity (%)
Oligohydramnios Amniotic fluid index Largest vertical pocket Pocket volume Normal amniotic fluid Amniotic fluid index Largest vertical pocket Pocket volume Polyhydramnios Amniotic fluid index Largest vertical pocket Pocket volume
Specificity (%)
6.7 0.0 60.0
100 100 84
Test efficiency (%)
65 62.5 75
73.7 94.7 73.7
28.6 14.3 42.6
50 52.5 65
83.3 50 50
85.3 97 97
85 90 90
group differences with p < 0.05 were considered statistically significant. Results
The patients' ages ranged from 17 to 39 with a mean of 23.8 ± 5.7 years, gestational ages ranged from 29 to 37 weeks with a mean of 34.3 ± 2.1 weeks, and gravidity ranged from 1 to 5 with a mean of l.9 ± l.2 (Table I). Thirty-one of the patients were black, four were Native American, and five were white. No evidence of intrauterine growth retardation or chorioamnionitis was present in this population. Preterm labor was present in 23 patients who underwent amniocentesis to evaluate the amniotic fluid for signs of subclinical infection, and the remaining 17 were assessed for fetal lung maturity. Ultrasonographic measurements were compared with amniotic fluid volume, as determined by the dye-dilution technique. Fluid volumes as quantitated by dye dilution were defined as oligohydramnios < 500 ml (n = 15), normal fluid 500 to 1500 ml (n = 19), or hydramnios > 1500 ml (n = 6). Normal fluid volumes (as confirmed by dye dilution) were correctly detected ultrasonographically by the two-diameter pocket in 61 %, by amniotic fluid index in 50%, and by largest vertical pocket in 50% of cases. These values were not different statistically. Hydramnios was identified by the pocket volume in (4/6) 67%, by amniotic fluid index (5/6) in 84%, and by largest vertical pocket (3/6) in 50% of the patients (P = 0.136). The test efficiency of the two-diameter pocket was 75% compared with 65% with the amniotic fluid index and 62.5% with the largest vertical pocket when applied
only to subjects with oligohydramnios (Table II). The two-diameter pocket gave a significantly more accurate estimate of oligohydramnios than did the amniotic fluid index (P < 0.002) and the largest vertical pocket (P < 0.0003). A receiver-operator characteristic curve for oligohydramnios was constructed for the two-diameter pocket and amniotic fluid index. The largest vertical pocket failed to identify any patients with oligohydramnios and was not used. A graphic representation (Fig. 1) demonstrates the superiority of the two-diameter pocket compared with the amniotic fluid index over their entire range of values. Comment
Amniotic fluid volume is an important adjunct in assessing fetal well-being and predicting certain perinatal disorders. I. 3. 4 The incidence of major congenital anomalies, insulin~dependent diabetes, multiple pregnancies, premature labor, and abruptio placentae is increased in women who have hydramnios. 5 Oligohydramnios and placental insufficiency are significantly related in the prolonged pregnancy. I I. 12 Oligohydramnios in the second trimester of pregnancy is associated with increased perinatal morbidity.13 The determination of amniotic fluid volume whether normal, oligohydramnios, or hydramnios has become an important diagnostic assessment in the overall evaluation of the fetus in late pregnancy. Older techniques of determining amniotic fluid volume were cumbersome and required leaving a needle in place or the performance of two amniocenteses. In spite of their accuracy, their cumbersome nature limited
1536 Magann et al.
December 1992 Am J Obstet Gynecol
1.0 0.9
0
0.8 0.7
-'w
>- 0.6
.s;
:,::;
cQ)
(f)
0.5 0.4
t:. Pocket Volume AFI
0.3
o
0.2 0.1 0.0
0.0 0.1
0.2
0.3
0.4 0.5 0.6 0.7
0.8
0.9
1.0
1 - Specificity Fig. 1. Receiver-operator characteristic curve comparing amniotic fluid index (AFl) versus pocket volume in analysis of oligohydramnios.
usefulness. They have been replaced by ultrasonographic estimations. The largest vertical pocket, as developed by Chamberlain et aI.,I continues to be used as one of the five parameters evaluated in the biophysical profile. I4 The amniotic fluid index was used by Rutherford et al. 6 to show a statistically significant increase in meconium staining, cesarean section for fetal distress, and Apgar scores > 7 at 1 and 5 minutes when the index value was :5 5. The amniotic fluid index has been compared by Moore I5 to the largest vertical pocket technique and found to be a better predictor of abnormal amounts of amniotic fluid (either oligohydramnios or hydramnios). In this study oligohydramnios and hydramnios, as defined by the amniotic fluid index, were compared with measurements of the largest vertical pocket. He found that the largest vertical pocket of < 3 em when compared with the amniotic fluid index from 1178 high-risk patients predicted those infants below the 5th percentile only 51 % of the time. The largest vertical pocket of ~ 8 em also predicted hydramnios poorly when compared with 95% for the amniotic fluid index group. This study used two ultrasonographically determined evaluations of amniotic fluid volume. The actual volume of amniotic fluid was not known. Ultrasonographic measurements have not been cor-
related with actual fluid volume confirmed by dye diluton techniques in the third trimester, so that the definitions of normal, oligohydramnios, and hydramnios as they relate to amniotic fluid index and the largest vertical pocket have not been validated. The amniotic fluid volume determined by dye dilution technique compared very favorably with amniotic fluid volume versus gestational age and previously collected and published data on amniotic fluid volume changes throughout pregnancy.I6 Our study confirms that the amniotic fluid index with a sensitivity of 6.7% and the largest vertical pocket with a sensitivity of 0% are poor predictors of oligohydramnios. The two-diameter pocket correctly identified oligohydramnios 60% of the time, significantly better in comparison with the average vertical pocket and amniotic fluid index. A receiveroperator characteristic curve that provides a clear graphic analysis of the performance of diagnostic tests showed that the two-diameter pocket is the superior test in comparison to the amniotic fluid index over nearly the entire range of values. The receiver-operator characteristic curve did not reveal an amniotic fluid volume ultrasonographic index in oligohydramnios that could be used to increase the sensitivity of the test without leading to an unacceptably high false-positive rate. Any of the three ultrasonographic measurements evaluate
Volume 167 I\umber 6
normal and increased amounts of amniotic fluid equally well. The two-diameter pocket in comparison with the amniotic fluid index and the largest vertical pocket is the superior test in the evaluation of oligohydramnios. We thank Denise Montgomery for skilled technical assistance in conducting this investigation. REFERENCES I. Chamberlain MB, Manning GA, Morrison I, Hannan CR, Lange IR. Ultrasound evaluation of amniotic fluid. I. The relationship of marginal and decreased amniotic fluid volume to perinatal outcome. AM J OBSTET GYNECOL 1984; 150:245-9. 2. Hill LM, Breckle R, Wolfgram KR, O'Brien PC. Oligohydramnios ultrasonically detected incidence and subsequent fetal outcome. AMJ OBSTET GYNECOL 1984;147:40710. 3. Mercer LJ, Brown LG, Petres RE, Messer RH. A survey of pregnancies complicated by decreased amniotic fluid. AM J OBSTET GYNECOL 1984; 149:355-61. 4. Hill LM, Breckle R, Thomas ML, et al. Polyhydramnios: ultrasonically detected prevalence and neonatal outcome. Obstet Gynecol 1987;69:21-5. 5. Chamberlain PF, Manning FA, Morrison L, Harman CR, Lange IR. Ultrasound evaluation of amniotic fluid. II. The relationship of increased amniotic fluid volume to perinatal outcome. AM J OBSTET GYNECOL 1984; 150:250-4. 6. 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;70:353-6.
AmniotiC fluid volume
1537
7. Moore TR, Cayle JE. The amniotic fluid index in normal human pregnancy. AM J OBSTET GYNECOL 1990;162:116873. 8. Charles D, Jacoby HE. Preliminary data on the use of sodium amino hippurate to detennine amniotic fluid volume. AM J OBSTEI GYNECOL 1966;95:266-9. 9. Schiff E, Ben-Baruchj G, Kushnir 0, Mashiach S. Standardized measurement of amniotic fluid by volume correlation of sonography with dye dilutional technique. Obstet Gynecol 1990;76:44-6. 10. QueenanJT, Thompson W, Whitfield CR, Shah SI. Amniotic fluid volumes in nonnal pregnancies. AM J OBSTET GYNECOL 1972;1l4:34-8. II. PhelanJP, Platt LD, Yeh S, Broussard P, Paul RH. The role of ultrasound assessment of amniotic fluid volume in the management of the postdate pregnancy. AM J OBSTET GYNECOL 1985;151:304-8. 12. Beischer NA, Brown JR, Townsend L. Studies in prolonged pregnancies. AM J OBSTET GYNECOL 1969; 103:496503. 13. Mercer LJ, Brown LG. Fetal outcome with oligohydramnios in the second trimester. Obstet Gynecol 1986;67: 840-2. 14. Manning FA, Baskett TF, Morrison I, Lange I. Fetal biophysical profile scoring: a prospective study in 1184 high risk patients. AM J OBSTET GYNECOL 1981; 140:289-94. 15. Moore TR. Superiority of the four-quadrant sum over the single largest pocket technique in ultrasonographic identification of abnormal amniotic fluid volumes. AM J OBSTET GYNECOL 1990;163:762-7. 16. Brace RA, WolfEJ. Normal amniotic fluid volume changes throughout pregnancy. AM J OBSTET GYNECOI. 1989;161: 382-8.
Volume 167 I'\umber 6
10. Varner MW, Hauser KS, Shaw K. Current trends in liveborn singleton birth weight. J Iowa Med Soc 1983;73: 447-50. 11. Brumfield CG, Cloud GA, Davis RO, Finley SC, Hauth JL, Boots L. The relationship between maternal serum and amniotic fluid a-fetoprotein in women undergoing early amniocentesis. AM J OBSTET GYNECOL 1990; 163:903-5. 12. Barford DA, Dickerman LH, Johnson WE. a-Fetoprotein: relationship between maternal serum and amniotic fluid levels. AMJ OBSTET GYNECOL 1985;151:1038-41. 13. Gill 1J, Repetti CF. Immunologic and genetic factors influencing reproduction. Am J Pathol 1979;95:465-570. 14. Greenberg F, Faucett WA, Rose E, et al. Congenital deficiency of alpha-fetoprotein. Pediatr Res 1989;25:52A. 15. Oblekwe BC, Malek N, Kitau MJ, Chard T. Maternal and fetal alpha fetoprotein (AFP) levels at term. Acta Obstet Gynecol Scand 1985;64:251-3. 16. Hubinont C, Fisk NM, Nicolini U, Rodeck CH, Johnson
17.
18.
19. 20.
RD. Fetal alpha-fetoprotein concentration in growth retardation. Br.J Obstet Gynecol 1989;96:1233-4. Caballero G, Vekemans M, Lopez del Campo J C, Robyn C. Serum a-fetoprotein in adults, in women during pregnancy, in children at birth, and during the first week of life: a sex difference. AM J OBSTET GYNECOI. 1977; 127: 384-9. Zeltzer PM, Neerhont RC, Fonkalsrud EW, Stiehm ER. Differentiation between neonatal hepatitis and biliary atresia by measuring serum alpha-fetoprotein. Lancet 1974; 1: 373-5. Katz VL, Cefalo RC, McCune BK, Moos MK. Elevated second trimester maternal serum alpha-fetoprotein and cytomegalovirus infection. Obstet Gynecol 1986;68:580-1. Hay DL, Barrie JV, Davison GB, et al. The relation between maternal serum alpha fetoprotein levels and fetomaternal haemorrhage. Br J Obstet Gynaecol 1979;86: 516-20.
Measurement of amniotic fluid volume: Accuracy of ultrasonography techniques Everett F. Magann, MD; Thomas E. Nolan, MD,b L. Wayne Hess, MD; Rick W. Martin, MD; Neil S. Whitworth, PhD; and John C. Morrison, MD
jackson, Mississippi, and Augusta, Georgia OBJECTIVE: Our purpose was to determine amniotic fluid volume by the dye-dilution technique and compare it with the amniotic fluid index, largest vertical pocket, and two-diameter pocket (defined as vertical x horizontal of the largest vertical pocket). STUDY DESIGN: This prospective study involved 40 women undergoing amniocentesis in late pregnancy to detect fetal lung maturity or evidence of chorioamnionitis. The amniotic fluid volume was quantified ultrasonographically by means of the amniotic fluid index, largest vertical pocket, and two-diameter pocket. During amniocentesis the fluid volume was calculated by the dye-dilution technique of Charles and Jacoby. RESULTS: Ultrasonographic measurements by amniotic fluid index, largest vertical pocket, and two-diameter pocket correctly predicted normal amniotic fluid and hydramnios (74%). A new measurement, two-diameter pocket, gave a significantly more accurate estimate of oligohydramnios than did amniotic fluid index (p < 0.002) or largest vertical pocket (p < 0.0003). CONCLUSION: All three indices are moderately accurate in identifying normal amniotic fluid volume and hydramnios. Two-diameter pocket is the most accurate test to predict oligohydramnios. (AM J OBSTET GYNECOL 1992;167:1533-7.)
Key words: Amniotic fluid, oligohydramnios, ultrasonography
From the Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, University of MississiPPi Medical Center: and the Department of Obstetrics and Gynecology, Medical College of Georgia.' Supported in part by the Vicksburg Hospital Medical Foundation. Presented in part at the Thirty-ninth Annual Meeting of the Society for Gynecologic Investigation, San Antonio, Texas, March 18-21, 1992. Reprint requests: Everett F. Magann, MD, Department of Obstetrics and Gynecology, University of Mississippi Medical Center, 2500 N. State St., Jackson, MS 39216-4505. 6/6/41725
Documentation of abnormalities of amniotic fluid volume may provide valuable information to enhance fetal health assessment. Decreased amounts of amniotic fluid, particularly in the third trimester, have been associated with multiple fetal risks, including death, pulmonary hypoplasia, growth retardation, and other complications. ,·3 In contrast increased amounts of amniotic fluid are correlated with fetal anomalies, twins, and insulin-dependent diabetes'" 5 A simple, inexpen-
1534 Magann et al.
sive, clinically useful test that will accurately determine amniotic fluid volume is needed. Unfortunately, even under research conditions techniques to confirm this measurement late in pregnancy are lacking. Ultrasonography is noninvasive and clinically can quantitate the amniotic fluid volume. It can be used to evaluate amniotic fluid volume by measurement of a single pocket of amniotic fluid or by means of a semiquantitative four-quadrant technique." 6, 7 Volume assessment by these techniques can be performed quickly, is easily taught, and is reproducible. 7 However, late in pregnancy, when it is most needed clinically, the error rate of ultrasonography is at its highest. Dye-dilution methods can be used to determine the amount of amniotic fluid in the same way the technology is used in calculating cardiac output." Amniotic fluid volume, as measured by a dye-dilution technique, has been correlated with the amniotic fluid as determined by ultrasonography in women undergoing second-trimester abortions. 9 Currently these techniques are the most accurate methods used to measure amniotic fluid volume. However, they are time-consuming, cumbersome, and invasive. Unfortunately, correlation of a single pocket of amniotic fluid or the amniotic fluid index with the actual volume of fluid as determined by dye-dilution techniques in the third trimester of pregnancy has not been accomplished. The purpose of this study is to correlate the amount of amniotic fluid as measured by ultrasonographic assessment with the actual volume determined by dyedilution techniques in late pregnancy. In addition, we calculated the pocket volume as a new measurement in the evaluation of amniotic fluid volume. Material and methods
This study comprised 40 women in the third trimester of pregnancy undergoing amniocentesis for the evaluation of fetal lung maturity or to detect subclinical chorioamnionitis in amniotic fluid. Inclusion criteria were as follows: (1) pregnancy between 28 and 37 weeks' gestation, (2) informed consent, (3) intact membranes, not in labor, and (4) maternal age> 15 years. Exclusion criteria were (1) lack of desire to participate in this study, (2) pregnancies < 28 weeks' gestation, and (3) parturients < 15 years old. Patients who agreed to participate in the study were then asked to sign a consent form approved by the Investigational Review Board of the University of Mississippi Medical Center. Ultrasonography was performed in all cases by the same physician with an Ultramark-4 device (Advanced Technology Laboratories, Bothell, Wash.). The amniotic fluid volume was assessed by means of the largest pocket method, as described by Chamberlain et aI.' In this technique the vertical and transverse diameters of the largest pocket of amniotic fluid were measured and
December 1992 Am J Obstet Gynecol
recorded. The depth of the pocket was measured at a right angle to the uterine contour. A depth of < 1 cm was classified as decreased, 1.1 to 2 cm as marginal (0 to 2 cm as oligohydramnios), 2.1 but 8.1 cm as hydramnios. Amniotic fluid volume was also determined by the two-diameter pocket. This volume consists of the vertical dimensions multiplied by the horizontal diameter of the largest pocket. The two-diameter pocket is a new measurement and a variation of the largest vertical pocket technique. The two-diameter pocket measurements corresponding to oligohydramnios, normal fluid, and hydramnios were compared with the actual fluid volume. These values were then compared with amniotic fluid index values on a receiver-operator characteristic curve and the ranges for each fluid volume were determined. A two-diameter pocket of 0 to 15 cm was oligohydramnios, 15.1 to 50 cm was normal fluid, and > 50.1 cm was hydramnios. The amniotic fluid volume was also assessed by the four-quadrant technique as described by Phelan. 6 The uterus was divided into four quadrants: by the umbilicus transversely (into upper and lower halves) and by the linea nigra vertically (into right and left halves). The maximum vertical diameter of the largest pocket in each quadrant without an aggregate of cord or fetal extremities was measured in centimeters. A depth of 0 to 5 cm equates oligohydramnios, 5.1 to 8 cm low, 8.1 to 18 cm normal, and > 18 cm high. Mter assessment of amniotic fluid volume by ultrasonography, amniocentesis was performed. Mter amniotic fluid was obtained for clinical studies, 2 ml of a 20% aqueous solution of aminohippurate sodium (400 mg, Merck, West Point, Pa.) was injected into the amniotic cavity. The amniocentesis needle was left in the amniotic cavity and continuous ultrasonographic monitoring of needle placement and fetal position was performed. The plunger of the syringe was withdrawn and depressed several times to promote rapid mixing of aminohippurate sodium with amniotic fluid. Three milliliters of amniotic fluid was collected 20 to 30 minutes after injection of aminohippurate sodium. It has been our experience, and also has been reported by others, 8 that optimal mixing of amniotic fluid has occurred by this time. The samples were frozen at - 20° C until assayed for aminohippurate sodium concentrations. Aminohippurate sodium concentrations and amniotic fluid volumes were determined with the diazo dye-reaction and spectrophotometric technique of Charles and Jacoby."' 10 The intraassay and interassay coefficient of variation was 1.9% and 3.5%, respectively. Serial dilutions of samples yielded aminohippurate sodium concentrations that were 101.8% of the expected value, and aminohippurate sodium added to a pool of amniotic fluid was recovered at a rate of 110.0%. The data were evaluated by X2 analysis. Between-
Amniotic fluid volume
Volume 167 J',;umhcr 6
1535
Table I. Demographic characteristics Race (yr)
Gestational age (wk)
Gravidity
White
23.8 ± 5.7
34.3 ± 2.1
1.9 ± 1.2
5
Age
No. 40
I
Black
Native American
I
31
4
Table II. Amniotic fluid volume test performance characteristics Sensitivity (%)
Oligohydramnios Amniotic fluid index Largest vertical pocket Pocket volume Normal amniotic fluid Amniotic fluid index Largest vertical pocket Pocket volume Polyhydramnios Amniotic fluid index Largest vertical pocket Pocket volume
Specificity (%)
6.7 0.0 60.0
100 100 84
Test efficiency (%)
65 62.5 75
73.7 94.7 73.7
28.6 14.3 42.6
50 52.5 65
83.3 50 50
85.3 97 97
85 90 90
group differences with p < 0.05 were considered statistically significant. Results
The patients' ages ranged from 17 to 39 with a mean of 23.8 ± 5.7 years, gestational ages ranged from 29 to 37 weeks with a mean of 34.3 ± 2.1 weeks, and gravidity ranged from 1 to 5 with a mean of l.9 ± l.2 (Table I). Thirty-one of the patients were black, four were Native American, and five were white. No evidence of intrauterine growth retardation or chorioamnionitis was present in this population. Preterm labor was present in 23 patients who underwent amniocentesis to evaluate the amniotic fluid for signs of subclinical infection, and the remaining 17 were assessed for fetal lung maturity. Ultrasonographic measurements were compared with amniotic fluid volume, as determined by the dye-dilution technique. Fluid volumes as quantitated by dye dilution were defined as oligohydramnios < 500 ml (n = 15), normal fluid 500 to 1500 ml (n = 19), or hydramnios > 1500 ml (n = 6). Normal fluid volumes (as confirmed by dye dilution) were correctly detected ultrasonographically by the two-diameter pocket in 61 %, by amniotic fluid index in 50%, and by largest vertical pocket in 50% of cases. These values were not different statistically. Hydramnios was identified by the pocket volume in (4/6) 67%, by amniotic fluid index (5/6) in 84%, and by largest vertical pocket (3/6) in 50% of the patients (P = 0.136). The test efficiency of the two-diameter pocket was 75% compared with 65% with the amniotic fluid index and 62.5% with the largest vertical pocket when applied
only to subjects with oligohydramnios (Table II). The two-diameter pocket gave a significantly more accurate estimate of oligohydramnios than did the amniotic fluid index (P < 0.002) and the largest vertical pocket (P < 0.0003). A receiver-operator characteristic curve for oligohydramnios was constructed for the two-diameter pocket and amniotic fluid index. The largest vertical pocket failed to identify any patients with oligohydramnios and was not used. A graphic representation (Fig. 1) demonstrates the superiority of the two-diameter pocket compared with the amniotic fluid index over their entire range of values. Comment
Amniotic fluid volume is an important adjunct in assessing fetal well-being and predicting certain perinatal disorders. I. 3. 4 The incidence of major congenital anomalies, insulin~dependent diabetes, multiple pregnancies, premature labor, and abruptio placentae is increased in women who have hydramnios. 5 Oligohydramnios and placental insufficiency are significantly related in the prolonged pregnancy. I I. 12 Oligohydramnios in the second trimester of pregnancy is associated with increased perinatal morbidity.13 The determination of amniotic fluid volume whether normal, oligohydramnios, or hydramnios has become an important diagnostic assessment in the overall evaluation of the fetus in late pregnancy. Older techniques of determining amniotic fluid volume were cumbersome and required leaving a needle in place or the performance of two amniocenteses. In spite of their accuracy, their cumbersome nature limited
1536 Magann et al.
December 1992 Am J Obstet Gynecol
1.0 0.9
0
0.8 0.7
-'w
>- 0.6
.s;
:,::;
cQ)
(f)
0.5 0.4
t:. Pocket Volume AFI
0.3
o
0.2 0.1 0.0
0.0 0.1
0.2
0.3
0.4 0.5 0.6 0.7
0.8
0.9
1.0
1 - Specificity Fig. 1. Receiver-operator characteristic curve comparing amniotic fluid index (AFl) versus pocket volume in analysis of oligohydramnios.
usefulness. They have been replaced by ultrasonographic estimations. The largest vertical pocket, as developed by Chamberlain et aI.,I continues to be used as one of the five parameters evaluated in the biophysical profile. I4 The amniotic fluid index was used by Rutherford et al. 6 to show a statistically significant increase in meconium staining, cesarean section for fetal distress, and Apgar scores > 7 at 1 and 5 minutes when the index value was :5 5. The amniotic fluid index has been compared by Moore I5 to the largest vertical pocket technique and found to be a better predictor of abnormal amounts of amniotic fluid (either oligohydramnios or hydramnios). In this study oligohydramnios and hydramnios, as defined by the amniotic fluid index, were compared with measurements of the largest vertical pocket. He found that the largest vertical pocket of < 3 em when compared with the amniotic fluid index from 1178 high-risk patients predicted those infants below the 5th percentile only 51 % of the time. The largest vertical pocket of ~ 8 em also predicted hydramnios poorly when compared with 95% for the amniotic fluid index group. This study used two ultrasonographically determined evaluations of amniotic fluid volume. The actual volume of amniotic fluid was not known. Ultrasonographic measurements have not been cor-
related with actual fluid volume confirmed by dye diluton techniques in the third trimester, so that the definitions of normal, oligohydramnios, and hydramnios as they relate to amniotic fluid index and the largest vertical pocket have not been validated. The amniotic fluid volume determined by dye dilution technique compared very favorably with amniotic fluid volume versus gestational age and previously collected and published data on amniotic fluid volume changes throughout pregnancy.I6 Our study confirms that the amniotic fluid index with a sensitivity of 6.7% and the largest vertical pocket with a sensitivity of 0% are poor predictors of oligohydramnios. The two-diameter pocket correctly identified oligohydramnios 60% of the time, significantly better in comparison with the average vertical pocket and amniotic fluid index. A receiveroperator characteristic curve that provides a clear graphic analysis of the performance of diagnostic tests showed that the two-diameter pocket is the superior test in comparison to the amniotic fluid index over nearly the entire range of values. The receiver-operator characteristic curve did not reveal an amniotic fluid volume ultrasonographic index in oligohydramnios that could be used to increase the sensitivity of the test without leading to an unacceptably high false-positive rate. Any of the three ultrasonographic measurements evaluate
Volume 167 I\umber 6
normal and increased amounts of amniotic fluid equally well. The two-diameter pocket in comparison with the amniotic fluid index and the largest vertical pocket is the superior test in the evaluation of oligohydramnios. We thank Denise Montgomery for skilled technical assistance in conducting this investigation. REFERENCES I. Chamberlain MB, Manning GA, Morrison I, Hannan CR, Lange IR. Ultrasound evaluation of amniotic fluid. I. The relationship of marginal and decreased amniotic fluid volume to perinatal outcome. AM J OBSTET GYNECOL 1984; 150:245-9. 2. Hill LM, Breckle R, Wolfgram KR, O'Brien PC. Oligohydramnios ultrasonically detected incidence and subsequent fetal outcome. AMJ OBSTET GYNECOL 1984;147:40710. 3. Mercer LJ, Brown LG, Petres RE, Messer RH. A survey of pregnancies complicated by decreased amniotic fluid. AM J OBSTET GYNECOL 1984; 149:355-61. 4. Hill LM, Breckle R, Thomas ML, et al. Polyhydramnios: ultrasonically detected prevalence and neonatal outcome. Obstet Gynecol 1987;69:21-5. 5. Chamberlain PF, Manning FA, Morrison L, Harman CR, Lange IR. Ultrasound evaluation of amniotic fluid. II. The relationship of increased amniotic fluid volume to perinatal outcome. AM J OBSTET GYNECOL 1984; 150:250-4. 6. 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;70:353-6.
AmniotiC fluid volume
1537
7. Moore TR, Cayle JE. The amniotic fluid index in normal human pregnancy. AM J OBSTET GYNECOL 1990;162:116873. 8. Charles D, Jacoby HE. Preliminary data on the use of sodium amino hippurate to detennine amniotic fluid volume. AM J OBSTEI GYNECOL 1966;95:266-9. 9. Schiff E, Ben-Baruchj G, Kushnir 0, Mashiach S. Standardized measurement of amniotic fluid by volume correlation of sonography with dye dilutional technique. Obstet Gynecol 1990;76:44-6. 10. QueenanJT, Thompson W, Whitfield CR, Shah SI. Amniotic fluid volumes in nonnal pregnancies. AM J OBSTET GYNECOL 1972;1l4:34-8. II. PhelanJP, Platt LD, Yeh S, Broussard P, Paul RH. The role of ultrasound assessment of amniotic fluid volume in the management of the postdate pregnancy. AM J OBSTET GYNECOL 1985;151:304-8. 12. Beischer NA, Brown JR, Townsend L. Studies in prolonged pregnancies. AM J OBSTET GYNECOL 1969; 103:496503. 13. Mercer LJ, Brown LG. Fetal outcome with oligohydramnios in the second trimester. Obstet Gynecol 1986;67: 840-2. 14. Manning FA, Baskett TF, Morrison I, Lange I. Fetal biophysical profile scoring: a prospective study in 1184 high risk patients. AM J OBSTET GYNECOL 1981; 140:289-94. 15. Moore TR. Superiority of the four-quadrant sum over the single largest pocket technique in ultrasonographic identification of abnormal amniotic fluid volumes. AM J OBSTET GYNECOL 1990;163:762-7. 16. Brace RA, WolfEJ. Normal amniotic fluid volume changes throughout pregnancy. AM J OBSTET GYNECOI. 1989;161: 382-8.
