American Journal of Perinatology VOLUME 9
NUMBER 5/6
SEPT/NOV 1992
AMNIOTIC FLUID INDEX: CORRELATION WITH AMNIOTIC FLUID VOLUME Iffath A. Hoskins, M.D., Peter G. McGovern, M.D., Steven A. Ordorica, M.D., Faith J. Frieden, M.D., and Bruce K. Young, M.D.
ABSTRACT
Clinicians have long recognized that assessment of amniotic fluid (AF) status is an integral part of fetal testing. AF is easily identified by current diagnostic ultrasonographic methods. Objective AF determination is based on identification of the largest pocket of fluid and measurement of its vertical diameter. This method is subject to error due to variations in pocket sizes created by fetal movements and, thus, is only of real value in detecting the extremes of fluid volume distribution and variation within the normal range. Recently, the four quadrant AF index (AFI) assessment has been described by Phelan et al.1 It appears to have the additional benefit of a more detailed and thus accurate evaluation of the intrauterine contents and fluid volume. The present study was undertaken to evaluate the correlation between the two different forms of AF assessments and to compare their accuracy relative to each other in normal and abnormal gestations. MATERIALS AND METHODS We prospectively performed AF assessments on all patients referred to the Prenatal Diagnostic Unit for various obstetric indications. All patients were divided into
Downloaded by: Universite de Sherbrooke. Copyrighted material.
We calculated the amniotic fluid indexes (AFIs) of 310 women on 459 occasions. Normative data were analyzed and compared with data in several high-risk groups. In the normal gestations there was a progressive increase in AFI with advancing gestation until 32 weeks, after which there was a decline. The mean AFIs in abnormal gestations varied with the clinical diagnoses. These values were compared to those obtained by assessing amniotic fluid volume (AFV), that is a pocket more than 2 cm. There were 51 patients with abnormal AFVs. Forty-two had decreased fluid, six also had decreased AFIs; nine had increased AFVs and five (all with diabetes) also had increased AFIs. Thus, AFIs in normal pregnancies showed an orderly pattern of change with gestational age, and there was no accurate correlation between AFI and AFV. Thus, using AFV alone may lead to false interpretations of amniotic fluid status.
one of two groups according to the indication for referral. Those referred for ultrasonographic confirmation of their dates, evaluation of placental location, or amniocenteses for chromosome and lung maturity studies provided the data for normative assessments. Women who were referred with clinical diagnoses of medical complications provided data for the high-risk assessments. All patients who were found to have multiple gestations, premature rupture of membranes, or suspected or proven fetal anomalies were excluded from the investigation. All AF assessments were conducted using the ATL Ultramark IV ultrasound scanner, with the patients being placed in the semi-Fowler position. Each patient underwent a four quadrant AFI assessment using the technique described by Phelan et al.1 For patients with gestational ages (GA) less than 26 weeks, the area below an arbitrary line across the uterine fundus was divided into four equal quadrants and measurements made with a sector scanner. For patients with GA greater than 26 weeks, the maternal abdomen over the uterus was divided into four quadrants. The umbilicus was used as the reference point for dividing into upper and lower halves and the linea nigra as the reference point for dividing into right and left halves. The transducer was placed on the maternal abdomen, along the longitudinal axis and perpendicular to the floor. The
Department of Obstetrics and Gynecology, New York University Medical Center, New York, New York Reprint requests: Dr. Hoskins, Department of Obstetrics and Gynecology, New York University Medical Center, 550 First Avenue, Room 9 North 1, New York, NY 10016 Copyright © 1992 by Thieme Medical Publishers, Inc., 381 Park Avenue South, New York, NY 10016. All rights reserved.
315
vertical diameter of a single largest AF pocket in each quadrant was measured in centimeters. The AFI was the total of the measurements obtained in each of the four quadrants. In order to minimize intraobserver variation, all patients included in the study had at least two AFI assessments, with the mean being taken as the true value. These results were then compared to the more traditional method of qualitatively assessing AF volume (AFV) by measuring, on each patient, any single pocket of fluid of at least 2 cm in vertical diameter. Pockets containing loops of umbilical cord were excluded from the study. The AFI was defined as normal if the value was 6 to 17 cm, decreased for a value of less than 6 cm and increased if the total was greater than or equal to 18 cm. The AFV was considered to be normal if there were fluid pockets whose vertical diameters were 2 to 4 cm, it was decreased if the AF pockets were less than 2 cm. and, if there were fluid pockets more than 4 cm in vertical diameter, it was called increased. Normative data were analyzed by least squares linear regression and compared to data from several high-risk groups. RESULTS
There were 310 women in this study who underwent AF assessments on 459 occasions, from 12 to 43 weeks GA. Of these, 302 assessments were performed on 218 women with normal gestations. The remaining 157 were on 92 patients with the following diagnoses: diabetes mellitus (31), chronic hypertension (25), intrauterine growth retardation (23), postdates (13). Analysis of the normative data showed that there was an orderly pattern with advancing gestational age of a progressively increasing AFI (mean, 15.3 ± 5.5 cm) (Table 1). This trend persisted until 32 weeks' GA, after which there was a slight decline in the mean value (13.5 ± 4.2 cm). The mean AFIs in the abnormal gestations varied with the clinical diagnoses of the patients. This is shown in Table 2. Of the 218 women with normal gestations, assessed on 302 occasions, 192 (88.1%) had normal AFVs. Twenty-six (11.9%) had abnormal values, with four of these (all in the third trimester) having increased amounts of fluid by this assessment method (Table 3). Of the 92 patients with abnormal gestations, 67 (72.8%) had normal and 25 (27.2%) had abnormal AFV assessments. Of these 25 patients five had AF pockets more than 4 cm. All these women had diabetes mellitus.
Table 1. Gestational Age (weeks)
12-16 17-21 22-26 27-31 32-36 37-41 42-44 Total 316
Amniotic Fluid Index: Normal Gestations No. Patients 7 9 17 35 57 63 30 218+
*Mean ± SEM. But there were 302 occasions.
+
Amniotic Fluid Index (cm)*
10.5 11.7 12.5 14.8 14.0 10.7 11.2
± ± ± ± ± ± ±
2.3 3.0 4.5 4.0 4.5 5.5 3.8
Sept/Nov 7992 Table 2.
Gestations Normal < 3 2 weeks > 3 2 weeks Abnormal Diabetes Hypertension Intrauterine growth retardation Postdates
Amniotic Fluid Index No. Patients 218 83 135 92 31 25 23 13
Mean ± SEM
15.3 ± 5.5 13.5 ± 4.2 19.2 ± 3.4 12.9 ± 5.1 12.1 ± 4.6 12.0 ± 4.0
In Tables 4 and 5 we show the correlation between the two methods of AF assessment. In the patients with normal pregnancies who were less than 32 weeks GA, there were no abnormal AFIs, but there were four women who had decreased AFVs, that is, pockets of fluid less than 2 cm in vertical diameter. In the women who had normal pregnancies and were more than 32 weeks GA, we found six with decreased AFIs (less than 6 cms) but 18 patients in this same subgroup were found to have decreased AFVs (less than 2 cm pockets). None of the women had increased AFIs, but four patients with apparently normal pregnancies had evidence of increased AFV (more than 4 cm pockets). In the abnormal pregnancies, four women with diabetes mellitus had increased AFIs. These same women were also found to have increased AFVs. There was one additional diabetic patient who had an increased AFV but had a normal AFI. There were six patients with decreased AFIs in this group (three intrauterine growth retardation [IUGR], three postdates) but there were 20 identified as having decreased AFVs (hypertension, 2; IUGR, 13; postdates, 5). DISCUSSION Ultrasonographic evaluation of the fetal environment, including assessment of AF status, is an integral part of antepartum fetal studies and provides insight into the presence or potential for fetal morbidity. Since alterations in AF amounts have been associated clinically with potentially serious obstetric situations, such as IUGR, diabetes mellitus, postdates, it is evident that studying this variable is extremely important in assessing the integrity of the fetoplacental unit. To date, there are no accurate correlations between the various noninvasive techniques to quantitate AF amounts. In 1987 Phelan et al1 published their findings using the four quadrant AFI in the third trimester of pregnancy. Subsequently, they described its values throughout gestation.2 Following this, Rutherford et al3 described its value as an adjunct to antepartum fetal heart rate testing in 330 high-risk pregnancies. In our study, we initially assessed the usefulness of the AFI in normal and abnormal gestations in our patient population (low socioeconomic status, inner city hospital) and then correlated these findings with AFV measurements. We found a progressive increase in the AFI until 32 weeks' GA and then a gradual decline in these values until term. This is in contrast to the findings of Phelan et al2 who noted a progressive rise in the AFI until 26 weeks' GA and then a plateau. It was only after 38 weeks' gestation that they were able to demonstrate a decline in the AFI. However, Gadd4 showed a progressive rise
Downloaded by: Universite de Sherbrooke. Copyrighted material.
AMERICAN JOURNAL OF PERINATOLOCY/VOLUME 9, NUMBER 5/6
AMNIOTIC FLUID ASSESSMENTS/Hoskins, et al. Table 3.
Amniotic Fluid Volumes: Normal and Abnormal Gestations
Gestation
No. Patients
Normal 32 weeks Abnormal Diabetes Hypertension Intrauterine growth retardation Postdates
Decreased Pocket (4 cm)
79 113
4 18
— 4
26 23 10 8
— 2 13 5
5 — — —
218 83 135 92 31 25 23 13
Correlation of Amniotic Fluid Index with Amniotic Fluid Volume
Amniotic Fluid Index
Amniotic Fluid Volume
Abnormal No. Patients
Gestation Normal 32 weeks Abnormal Diabetes Hypertension Intrauterine growth retardation Postdates
Normal 6-17
218 83 135 92 31 25 23 13
Table 5.
Abnormal
Decreased Increased
NUMBER 5/6
SEPT/NOV 1992
AMNIOTIC FLUID INDEX: CORRELATION WITH AMNIOTIC FLUID VOLUME Iffath A. Hoskins, M.D., Peter G. McGovern, M.D., Steven A. Ordorica, M.D., Faith J. Frieden, M.D., and Bruce K. Young, M.D.
ABSTRACT
Clinicians have long recognized that assessment of amniotic fluid (AF) status is an integral part of fetal testing. AF is easily identified by current diagnostic ultrasonographic methods. Objective AF determination is based on identification of the largest pocket of fluid and measurement of its vertical diameter. This method is subject to error due to variations in pocket sizes created by fetal movements and, thus, is only of real value in detecting the extremes of fluid volume distribution and variation within the normal range. Recently, the four quadrant AF index (AFI) assessment has been described by Phelan et al.1 It appears to have the additional benefit of a more detailed and thus accurate evaluation of the intrauterine contents and fluid volume. The present study was undertaken to evaluate the correlation between the two different forms of AF assessments and to compare their accuracy relative to each other in normal and abnormal gestations. MATERIALS AND METHODS We prospectively performed AF assessments on all patients referred to the Prenatal Diagnostic Unit for various obstetric indications. All patients were divided into
Downloaded by: Universite de Sherbrooke. Copyrighted material.
We calculated the amniotic fluid indexes (AFIs) of 310 women on 459 occasions. Normative data were analyzed and compared with data in several high-risk groups. In the normal gestations there was a progressive increase in AFI with advancing gestation until 32 weeks, after which there was a decline. The mean AFIs in abnormal gestations varied with the clinical diagnoses. These values were compared to those obtained by assessing amniotic fluid volume (AFV), that is a pocket more than 2 cm. There were 51 patients with abnormal AFVs. Forty-two had decreased fluid, six also had decreased AFIs; nine had increased AFVs and five (all with diabetes) also had increased AFIs. Thus, AFIs in normal pregnancies showed an orderly pattern of change with gestational age, and there was no accurate correlation between AFI and AFV. Thus, using AFV alone may lead to false interpretations of amniotic fluid status.
one of two groups according to the indication for referral. Those referred for ultrasonographic confirmation of their dates, evaluation of placental location, or amniocenteses for chromosome and lung maturity studies provided the data for normative assessments. Women who were referred with clinical diagnoses of medical complications provided data for the high-risk assessments. All patients who were found to have multiple gestations, premature rupture of membranes, or suspected or proven fetal anomalies were excluded from the investigation. All AF assessments were conducted using the ATL Ultramark IV ultrasound scanner, with the patients being placed in the semi-Fowler position. Each patient underwent a four quadrant AFI assessment using the technique described by Phelan et al.1 For patients with gestational ages (GA) less than 26 weeks, the area below an arbitrary line across the uterine fundus was divided into four equal quadrants and measurements made with a sector scanner. For patients with GA greater than 26 weeks, the maternal abdomen over the uterus was divided into four quadrants. The umbilicus was used as the reference point for dividing into upper and lower halves and the linea nigra as the reference point for dividing into right and left halves. The transducer was placed on the maternal abdomen, along the longitudinal axis and perpendicular to the floor. The
Department of Obstetrics and Gynecology, New York University Medical Center, New York, New York Reprint requests: Dr. Hoskins, Department of Obstetrics and Gynecology, New York University Medical Center, 550 First Avenue, Room 9 North 1, New York, NY 10016 Copyright © 1992 by Thieme Medical Publishers, Inc., 381 Park Avenue South, New York, NY 10016. All rights reserved.
315
vertical diameter of a single largest AF pocket in each quadrant was measured in centimeters. The AFI was the total of the measurements obtained in each of the four quadrants. In order to minimize intraobserver variation, all patients included in the study had at least two AFI assessments, with the mean being taken as the true value. These results were then compared to the more traditional method of qualitatively assessing AF volume (AFV) by measuring, on each patient, any single pocket of fluid of at least 2 cm in vertical diameter. Pockets containing loops of umbilical cord were excluded from the study. The AFI was defined as normal if the value was 6 to 17 cm, decreased for a value of less than 6 cm and increased if the total was greater than or equal to 18 cm. The AFV was considered to be normal if there were fluid pockets whose vertical diameters were 2 to 4 cm, it was decreased if the AF pockets were less than 2 cm. and, if there were fluid pockets more than 4 cm in vertical diameter, it was called increased. Normative data were analyzed by least squares linear regression and compared to data from several high-risk groups. RESULTS
There were 310 women in this study who underwent AF assessments on 459 occasions, from 12 to 43 weeks GA. Of these, 302 assessments were performed on 218 women with normal gestations. The remaining 157 were on 92 patients with the following diagnoses: diabetes mellitus (31), chronic hypertension (25), intrauterine growth retardation (23), postdates (13). Analysis of the normative data showed that there was an orderly pattern with advancing gestational age of a progressively increasing AFI (mean, 15.3 ± 5.5 cm) (Table 1). This trend persisted until 32 weeks' GA, after which there was a slight decline in the mean value (13.5 ± 4.2 cm). The mean AFIs in the abnormal gestations varied with the clinical diagnoses of the patients. This is shown in Table 2. Of the 218 women with normal gestations, assessed on 302 occasions, 192 (88.1%) had normal AFVs. Twenty-six (11.9%) had abnormal values, with four of these (all in the third trimester) having increased amounts of fluid by this assessment method (Table 3). Of the 92 patients with abnormal gestations, 67 (72.8%) had normal and 25 (27.2%) had abnormal AFV assessments. Of these 25 patients five had AF pockets more than 4 cm. All these women had diabetes mellitus.
Table 1. Gestational Age (weeks)
12-16 17-21 22-26 27-31 32-36 37-41 42-44 Total 316
Amniotic Fluid Index: Normal Gestations No. Patients 7 9 17 35 57 63 30 218+
*Mean ± SEM. But there were 302 occasions.
+
Amniotic Fluid Index (cm)*
10.5 11.7 12.5 14.8 14.0 10.7 11.2
± ± ± ± ± ± ±
2.3 3.0 4.5 4.0 4.5 5.5 3.8
Sept/Nov 7992 Table 2.
Gestations Normal < 3 2 weeks > 3 2 weeks Abnormal Diabetes Hypertension Intrauterine growth retardation Postdates
Amniotic Fluid Index No. Patients 218 83 135 92 31 25 23 13
Mean ± SEM
15.3 ± 5.5 13.5 ± 4.2 19.2 ± 3.4 12.9 ± 5.1 12.1 ± 4.6 12.0 ± 4.0
In Tables 4 and 5 we show the correlation between the two methods of AF assessment. In the patients with normal pregnancies who were less than 32 weeks GA, there were no abnormal AFIs, but there were four women who had decreased AFVs, that is, pockets of fluid less than 2 cm in vertical diameter. In the women who had normal pregnancies and were more than 32 weeks GA, we found six with decreased AFIs (less than 6 cms) but 18 patients in this same subgroup were found to have decreased AFVs (less than 2 cm pockets). None of the women had increased AFIs, but four patients with apparently normal pregnancies had evidence of increased AFV (more than 4 cm pockets). In the abnormal pregnancies, four women with diabetes mellitus had increased AFIs. These same women were also found to have increased AFVs. There was one additional diabetic patient who had an increased AFV but had a normal AFI. There were six patients with decreased AFIs in this group (three intrauterine growth retardation [IUGR], three postdates) but there were 20 identified as having decreased AFVs (hypertension, 2; IUGR, 13; postdates, 5). DISCUSSION Ultrasonographic evaluation of the fetal environment, including assessment of AF status, is an integral part of antepartum fetal studies and provides insight into the presence or potential for fetal morbidity. Since alterations in AF amounts have been associated clinically with potentially serious obstetric situations, such as IUGR, diabetes mellitus, postdates, it is evident that studying this variable is extremely important in assessing the integrity of the fetoplacental unit. To date, there are no accurate correlations between the various noninvasive techniques to quantitate AF amounts. In 1987 Phelan et al1 published their findings using the four quadrant AFI in the third trimester of pregnancy. Subsequently, they described its values throughout gestation.2 Following this, Rutherford et al3 described its value as an adjunct to antepartum fetal heart rate testing in 330 high-risk pregnancies. In our study, we initially assessed the usefulness of the AFI in normal and abnormal gestations in our patient population (low socioeconomic status, inner city hospital) and then correlated these findings with AFV measurements. We found a progressive increase in the AFI until 32 weeks' GA and then a gradual decline in these values until term. This is in contrast to the findings of Phelan et al2 who noted a progressive rise in the AFI until 26 weeks' GA and then a plateau. It was only after 38 weeks' gestation that they were able to demonstrate a decline in the AFI. However, Gadd4 showed a progressive rise
Downloaded by: Universite de Sherbrooke. Copyrighted material.
AMERICAN JOURNAL OF PERINATOLOCY/VOLUME 9, NUMBER 5/6
AMNIOTIC FLUID ASSESSMENTS/Hoskins, et al. Table 3.
Amniotic Fluid Volumes: Normal and Abnormal Gestations
Gestation
No. Patients
Normal 32 weeks Abnormal Diabetes Hypertension Intrauterine growth retardation Postdates
Decreased Pocket (4 cm)
79 113
4 18
— 4
26 23 10 8
— 2 13 5
5 — — —
218 83 135 92 31 25 23 13
Correlation of Amniotic Fluid Index with Amniotic Fluid Volume
Amniotic Fluid Index
Amniotic Fluid Volume
Abnormal No. Patients
Gestation Normal 32 weeks Abnormal Diabetes Hypertension Intrauterine growth retardation Postdates
Normal 6-17
218 83 135 92 31 25 23 13
Table 5.
Abnormal
Decreased Increased
