| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From the Department of Obstetrics and Gynecology, University of Texas Southwestern Medical Center at Dallas, Dallas, Texas; and the Louisiana State University Medical Center, Shreveport, Louisiana.
Address reprint requests to: Jodi S. Dashe, MD Department of Obstetrics and Gynecology University of Texas Southwestern Medical Center at Dallas 5323 Harry Hines Boulevard Dallas, TX 75390-9032 E-mail: jodi.dashe{at}email.swmed.edu
 |
Abstract |
|---|
Top
 Abstract Materials and MethodsResultsDiscussionReferences |
|---|
Methods: In a retrospective cohort study, we analyzed consecutive live-born singletons of women who had antepartum sonography within 4 weeks of delivery and delivered between January 1, 1989 and September 30, 1996. A gestational age–specific HC/AC nomogram was derived from our sonographic database of 33,740 nonanomalous live-born singletons. Asymmetric HC/AC was defined as greater than or equal to the 95th percentile for gestational age.
Results: Among 1364 SGA infants, 20% had asymmetric HC/AC and 80% were symmetric. Asymmetric SGA infants were more likely to have major anomalies than symmetric SGA infants or AGA infants (14% versus 4% versus 3%, respectively; P < .001). After exclusion of anomalous infants, pregnancy-induced hypertension at or before 32 weeks’ gestation and cesarean delivery for nonreassuring fetal heart rate were more common in the asymmetric SGA than the AGA group (7% versus 1% and 15% versus 3%, respectively; both P < .001). A neonatal outcome composite, including one or more of respiratory distress, intraventricular hemorrhage, sepsis, or neonatal death, was more frequent among asymmetric SGA than AGA infants (14% versus 5%, P = .001). Symmetric SGA infants were not at increased risk of morbidity compared with AGA infants.
Conclusion: The minority of SGA fetuses with HC/AC asymmetry are at increased risk for intrapartum and neonatal complications.
In 1977, Campbell and Thoms1 used the sonographic head circumference to abdominal circumference ratio (HC/AC) to differentiate fetuses as "symmetric," or proportionately small, and "asymmetric," or disproportionately lagging in abdominal growth. They constructed an HC/AC nomogram from approximately 500 normal fetuses and evaluated its use in 31 fetuses at risk of uteroplacental insufficiency. Seventy percent of those fetuses had HC/AC above the 95th percentile and were termed asymmetric. Although asymmetric fetuses had relatively larger brains and were "preferentially protected from the full effects of the growth retarding stimulus," they were at significantly greater risk of severe preeclampsia, fetal distress, operative intervention, and lower Apgar scores than their symmetric counterparts.1
There has been controversy surrounding the prognosis of growth-restricted pregnancies. A recent review concluded that 70–80% of growth-restricted infants were asymmetric, whereas only 20–30% were symmetric, and that symmetrically small neonates were at greater risk of adverse outcomes.2 Although it is generally believed that symmetrically small infants are more likely to be aneuploid, to have congenital infections, and to suffer more neonatal complications, others have found just the opposite: that asymmetrically growth-restricted infants are more likely to be compromised.1,3–5
We conducted this retrospective cohort study to assess the symmetry of small for gestational age (SGA) infants and to determine the risk of adverse outcomes in asymmetric SGA infants compared with symmetric SGA and appropriate for gestational age (AGA) infants.
|
Materials and Methods |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
We previously published a validated birth weight percentile nomogram of singleton live births without major anomalies delivered at our institution.6 Distributions of birth weights for each completed week of gestation are statistically normal, and smoothed birth weight curves have been derived for each percentile.7 Approximately 97% of women received prenatal care in our system, nearly 60% beginning in the first trimester and 90% before the end of the second trimester. Gestational age was based on women’s last menstrual periods (LMP), provided that uterine fundal height corresponded to expected gestational age.8 Ultrasound was done if there were discrepancies between fundal height and LMP or if the LMP was uncertain. Obstetric estimates of gestational ages correlated well with sonographic (r = .97) and pediatric (r = .89) estimates of gestational ages.6
Infants with birth weights at or below the tenth percentile for gestational age were termed SGA for the purpose of this study. This percentile threshold was selected because it is a frequently used benchmark for disordered fetal growth, and we were previously unable to establish an alternative specific optimal fetal growth threshold for infants delivered at or before 36 weeks.4,6 For the reference group we chose infants between the 25th and 75th percentiles for gestational age, which we termed AGA.
To differentiate between asymmetric and symmetric growth restriction, we constructed a gestational age–specific nomogram for HC/AC from 33,740 singleton fetuses of women who had second- or third-trimester ultrasound scans between January 1, 1989 and September 30, 1996. We excluded from the nomogram fetal deaths and neonates found to have major anomalies before hospital discharge, regardless of whether such anomalies were diagnosed prenatally. As shown in Figure 1
, the HC/AC decreases with advancing gestation; it has a Gaussian distribution for each week of gestation from 16 through 42 weeks. Those with HC/AC at or above the 95th percentile for gestational age were termed asymmetric, and those with HC/AC below the 95th percentile for gestational age were termed symmetric. We used the 95th percentile cutoff originally described by Campbell and Thoms.1 However, we also constructed receiver operating characteristic curves for selected adverse neonatal outcomes and found that the 95th percentile discriminated more accurately between symmetric and asymmetric groups than did 90th or 99th percentile cutoffs.
|
Neonatal morbidity and outcome data were based on diagnoses by neonatal intensive care unit (NICU) faculty. Respiratory distress syndrome (RDS) diagnoses were based on the need for supplemental oxygen after the first 24 hours of life and characteristic radiographic findings. Diagnoses of bronchopulmonary dysplasia typically were made if infants needed supplemental oxygen or diuresis at 28 days of life or 36 weeks post–menstrual age. Intraventricular hemorrhages were analyzed only if grade III or IV. Diagnoses of periventricular leukomalacia were based on characteristic radiographic findings. We included only those cases of necrotizing enterocolitis confirmed during surgery. We reported cases of neonatal sepsis only if confirmed by positive blood cultures.
For continuous variables, statistical analysis was done with analysis of variance with Student-Newman-Keuls post hoc comparison. For categoric variables, Pearson’s 
2 was used, with a log-linear model for multiple comparisons. Multiple logistic regression was applied to adjust the comparison of outcomes for labor induction. P < .05 was statistically significant. Statistical analysis was done with the SAS system (SAS Institute, Cary, NC).
|
Results |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
shows the distribution of these infants according to whether they were asymmetric SGA, symmetric SGA, or AGA. Using birth weight percentiles for our population, we found that 1364 (16%) of infants were at or below the tenth percentile (SGA). Among those, 1090 (80%) were symmetric and 274 (20%) were asymmetric. Three thousand eight hundred seventy-three infants were AGA. Also shown is the corresponding prevalence of major anomalies. More major anomalies were detected among asymmetric SGA infants than among symmetric SGA infants or AGA infants (14% versus 4% versus 3%, respectively; P < .001). This difference persisted after adjustment for neonates with ventral wall defects, whose abdomens might be expected to be small. Aneuploidy also was more prevalent among asymmetric SGA infants (3%) than symmetric SGA (1%) or AGA infants (less than 0.1%) (P < .001).
|
. Mean maternal age at delivery and the proportions of women under 15 and over 35 years old did not differ between the AGA and SGA groups or between the symmetric and asymmetric SGA subsets. There was no SGA propensity among black or white women, although the asymmetric SGA subset contained fewer Hispanic women than either of the other groups. Weight at delivery was lower among mothers of SGA infants than those of AGA infants, and SGA infants were more likely than AGA infants to be born to nulliparas. There were no significant differences between the asymmetric and symmetric SGA subsets in maternal weight at delivery or nulliparity.
|
). The most common indication was a suspected lag in fetal growth, which was more common in both SGA subsets than in the AGA group, but was not different between the asymmetric and symmetric SGA subsets. Ultrasound for uncertain gestational age was more common in the AGA group than in either SGA subset. Evaluation for possible anomalies was more common in asymmetric SGA than symmetric SGA fetuses, although not significantly more common than in AGA fetuses. No other indication for ultrasound was significantly different among the groups.
As shown in Figure 3, the mean birth weight was significantly lower in the asymmetric group than in the symmetric group at each gestational age beginning at 28 weeks. Gestational age at delivery was earlier in asymmetric SGA infants than either symmetric SGA or AGA infants, and asymmetric SGA infants were more likely to deliver at or before 32 weeks’ gestation (Table 2). Preterm induction of labor (at most 36 weeks) was more common in asymmetric SGA than AGA infants. Intrapartum hypertension treated with magnesium sulfate for eclampsia prophylaxis was significantly more common in mothers of SGA infants than AGA infants, but was not more common among the asymmetric SGA subset. Intrapartum hypertension that necessitated delivery at or before 32 weeks’ gestation was significantly more common in the asymmetric SGA group than in the symmetric SGA or AGA groups. Cesarean delivery for nonreassuring fetal heart rate tracing was significantly more common with SGA than AGA fetuses and was nearly twice as frequent among pregnancies complicated by asymmetric as opposed to symmetric fetal growth restriction.
|
|
). These differences in low Apgar scores and delivery room intubations represent few infants and might not have clinical significance; however, the difference in admissions to the NICU (18% of asymmetric SGA infants versus 7% of AGA infants) might be important clinically. There were no statistically significant differences between AGA infants and symmetric SGA infants for those outcomes. Few infants had umbilical artery pH at or below 7.1 or 7.0, and there were no differences in the likelihood of such values among the groups.
|
. Asymmetric SGA infants were significantly more likely than AGA infants to be diagnosed with RDS (9% of asymmetric SGA infants versus 3% of AGA infants; P < .001). Asymmetric SGA infants also had a significantly greater likelihood of grade III or IV intraventricular hemorrhage and death in the immediate neonatal period, although both outcomes were so infrequent that such differences might not be clinically significant. The overall prevalence of most neonatal morbidities was low, so we also analyzed a neonatal outcome composite, to include one or more of RDS, bronchopulmonary dysplasia, intraventricular hemorrhage, periventricular leukomalacia, necrotizing enterocolitis, sepsis, or neonatal death. The neonatal composite was significantly more common in the asymmetric SGA group than in the AGA group (14% versus 5%, P < .001) and remained significantly more common among asymmetric SGA infants than AGA infants when the analysis was restricted to preterm infants (at most 36 weeks’ gestation; 32% versus 15%, P < .001). For each of the outcomes listed, there were no significant differences between the symmetric SGA group and the AGA group.
|
).
|
|
Discussion |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Among the limitations of our study is that we included only pregnancies in which ultrasound was done within 4 weeks of delivery, so complicated pregnancies might have been more likely to come to our attention. It is unlikely that the prevalence of asymmetrically small fetuses would have increased if more low-risk women were included. Another possible bias was in gestational age assignment. We used the same obstetric estimate of gestational age that was used by the obstetricians who cared for our patients, and we have previously validated this method.6 Thirteen percent of women had their gestational ages confirmed by ultrasound within the first 20 weeks of gestation, and those with asymmetric SGA remained at significantly increased risk of morbidity.
It has long been considered that head and abdomen proportions in SGA infants indicate the timing and nature of the insult, with the assumption that extrinsic causes lead to asymmetric and intrinsic causes to symmetric growth restriction. Although those generalizations are interesting conceptually, fetal growth patterns are more complex.11 The most common extrinsic insult is placental insufficiency, typically caused by hypertension. Asymmetric growth restriction was more common with pregnancy-induced hypertension in some studies but not in others.1,10 We found intrapartum hypertension associated with asymmetric growth in infants delivered at 32 weeks’ gestation or less; however, asymmetry was not more common among infants delivered in the setting of maternal hypertension later in gestation.
Another assumption has been that processes that inhibit mitosis, such as anomalies, result in symmetric growth restriction.4 We found that major anomalies, including aneuploidy, were more often associated with asymmetric than symmetric growth restriction, even after exclusion of some anomalies expected to decrease size of the abdomen. Nicolaides et al12 observed asymmetry, rather than symmetry, in growth-restricted infants with aneuploidy. We did not compile data on prenatal detection of anomalies, so we do not want our data to be misinterpreted to suggest that asymmetry is a marker for anomalies, although prospective sonographic studies of anomaly detection rates in fetuses with asymmetric and symmetric HC/AC would be of interest.
Our results consistently indicate that the prognosis of SGA infants with asymmetric growth is poorer than that of symmetrically grown infants and much worse than that of AGA infants. Symmetrically SGA infants compare favorably with normally grown AGA infants. This is in keeping with the understanding that most symmetrically grown infants below the tenth percentile are merely constitutionally small. Asymmetry of the fetal HC/AC indicates pronounced growth impairment. The increased morbidity of asymmetric SGA infants might reflect both earlier gestational age at delivery and lower weight for gestational age.
|
Footnotes |
|---|
Received January 18, 2000. Received in revised form April 5, 2000. Accepted April 27, 2000.
|
References |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
2. Lin CC, Santolaya-Forgas JS. Current concepts of fetal growth restriction: Part I. Causes, classification, and pathophysiology. Obstet Gynecol 1998;92:1044–55.[Abstract]
3. Villar J, de Onis M, Kestler E, Bolanos F, Cerezo R, Bernedes H. The differential neonatal morbidity of the intrauterine growth retardation syndrome. Am J Obstet Gynecol 1990;163:151–7.[Medline]
4. Lockwood CJ, Weiner SJ. Assessment of fetal growth. Clin Perinatol 1986;13:3–35.[Medline]
5. Patterson RM, Pouliot MR. Neonatal morphometrics and perinatal outcome: Who is growth retarded? Am J Obstet Gynecol 1987;157: 691–3.[Medline]
6. McIntire DD, Bloom SL, Casey BM, Leveno KJ. Birthweight in relation to morbidity and mortality among newborn infants. N Engl J Med 1999;340:1234–8.
7. Velleman PE. Definition and comparison of robust linear data smoothing algorithms. J Am Stat Assoc 1980;75:609–15.
8. Jimenez JS, Tyson JE, Reisch JS. Clinical measures of gestational age in normal pregnancies. Obstet Gynecol 1983;61:438–43.
9. Opitz JM, Gilbert EF. Pathogenic analysis of congenital anomalies in humans. Pathobiol Annu 1982;12:301–49.[Medline]
10. Lin CC, Su SJ, River LP. Comparison of associated high-risk factors and perinatal outcome between symmetric and asymmetric fetal intrauterine growth retardation. Am J Obstet Gynecol 1991;164: 1535–42.[Medline]
11. Cunningham FG, MacDonald PC, Gant NF, Leveno KJ, Gilstrap LC, Hankins GDV, et al, eds. Fetal growth restriction. In: Williams obstetrics. 20th ed. East Norwalk, Connecticut: Appleton & Lange, 1997:841–5.
12. Nicolaides KH, Snijders RJM, Noble P. Cordocentesis in the study of growth-retarded fetuses. In: Divon MY, ed. Abnormal fetal growth. New York: Elsevier Science, 1991:164–8.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
