| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
by the American Institute of Ultrasound in Medicine J Ultrasound Med 22:359-363 • 0278-4297 Fetuses Subsequently Born Premature Are Smaller Than Gestational Age–Matched Fetuses Not Born PrematureDepartments of Radiology (P.M.D., C.B.B.), Obstetrics and Gynecology (L.W.-H.), and Newborn Medicine (S.R.), Brigham and Women’s Hospital and Harvard Medical School, Boston, Massachusetts USA. Address correspondence and reprint requests to Peter M. Doubilet, MD, PhD, Department of Radiology, Brigham and Women’s Hospital, 75 Francis St, Boston, MA 02115 USA.
Objective. To determine whether singleton fetuses in pregnancies that subsequently deliver prematurely are smaller than singleton fetuses of the same gestational age who are not born premature. Methods. Our study population consisted of singleton pregnancies meeting the following criteria: at least 1 first-trimester sonogram (to ensure accurate dating), at least 1 sonogram after 24 weeks’ gestation, and a known delivery date. Controlling for gestational age at sonography using analysis of covariance, we compared very premature fetuses (delivery at 24–29.9 weeks; n = 26) and moderately premature fetuses (delivery at 30–36.9 weeks; n = 306) with nonpremature fetuses (n = 1838) with respect to the following sonographic parameters: abdominal diameter, femur length, biparietal diameter, and estimated fetal weight. Results. On 24- to 29.9-week sonograms, fetuses who were subsequently born very premature had significantly smaller fetal measurements than did fetuses who were not born premature (P < .05 for all parameters). Even after excluding 9 very premature fetuses with a risk factor for uteroplacental insufficiency (e.g., toxemia or hypertension), the remaining 17 fetuses had significantly smaller abdominal diameters, femur lengths, and estimated fetal weights than nonpremature fetuses (P < .05). Sonographic parameters in moderately premature fetuses were smaller than in nonpremature fetuses on 30- to 36.9-week sonograms (P < .05 for all parameters), but the only parameters that differed significantly between these 2 groups on 24- to 29.9-week sonograms were abdominal diameter and femur length. Conclusions. Singleton fetuses subsequently born premature are smaller than gestational age–matched fetuses not born premature, even in the absence of an identifiable cause of growth restriction. The lag in growth appears to occur in the last few weeks before delivery.
Key Words: birth weight • obstetric sonography • prematurity Abbreviations: AD, abdominal diameter • ANCOVA, analysis of covariance • BPD, biparietal diameter • EFW, estimated fetal weight • FL, femur length • GA, gestational age
A number of studies have suggested that there is a relationship between small fetal size and preterm delivery. One study compared fetal measurements obtained on 32-week sonograms in 2 groups of patients, those giving birth before term (before 37 weeks’ gestation) and those giving birth at term, and found that the measurements were smaller in the former group.1 In another study based on obstetric sonographic examinations performed in women who subsequently gave birth prematurely, the biparietal diameter (BPD) and abdominal circumference were found to fall below the 10th percentile in considerably more than 10% of cases.2 When birth weights of premature neonates were compared with weight norms developed from estimated fetal weights (EFWs), considerably more than 10% fell below the 10th percentile.3 Birth weights of premature neonates have been found to be lower relative to an estimate of their "individual growth potential" than in neonates born at term.4 In women with preterm labor treated with tocolytic therapy, the ratio of femur length (FL) to abdominal circumference (a measure that tends to be elevated in growth-restricted fetuses5) was found to be higher in those who gave birth prematurely than those in whom tocolytic therapy was successful.6 These studies provide evidence that there is an association between poor fetal growth and early delivery. We conducted a study to confirm the relationship between small fetal size and prematurity by directly comparing sonographic measurements of singleton fetuses who are subsequently born premature with those of gestational age (GA)–matched singleton fetuses who are born at term. Our goals, in addition, were to address a number of questions left unanswered by prior studies: Does the tendency to small fetal size in pregnancies with premature delivery hold regardless of the degree of prematurity, or does it hold only for very premature cases (those with delivery before 30 weeks’ gestation)? Does the tendency hold in pregnancies with premature delivery without risk factors for uteroplacental insufficiency? At what point in gestation does fetal growth become subnormal in pregnancies that ultimately deliver prematurely?
From our database of obstetric sonographic examinations performed at our institution since 1989, we extracted a study population of singleton pregnancies meeting the following criteria: at least 1 first-trimester sonogram (to ensure accurate dating), at least 1 sonogram between 24 and 37 weeks’ gestation, and a known delivery date. A total of 2170 pregnancies met these criteria. For each case in the study population, the initial first-trimester sonogram was used as the basis for GA at the time of all subsequent sonograms and at birth; that is, the GA at the time of any subsequent sonogram or at birth was calculated by summing the GA determined at the initial first-trimester scan and the time elapsed since that scan. For each case in the study population, we identified either 1 or 2 study sonograms: the earliest sonogram obtained at 24 to 29.9 weeks (if one was obtained during this period) and the earliest sonogram obtained at 30 to 36.9 weeks (if one was obtained during this period). All study sonographic examinations were performed according to our institutional protocol for third-trimester scans, which includes measurement of the abdominal diameter (AD), FL, and BPD. These measurements were used to calculate7 the EFW from 24 weeks onward. For each study sonogram, we recorded GA, AD, FL, BPD, and EFW. Each case was classified according to the GA at delivery: very premature if delivered at 24 to 29.9 weeks (n = 26), moderately premature if at 30 to 36.9 weeks (n = 306), and nonpremature if at 37 weeks or beyond (n = 1838). Controlling for GA at sonography using analysis of covariance (ANCOVA), we compared very premature fetuses with nonpremature fetuses with respect to the following parameters obtained on a 24- to 29.9-week sonogram: AD, FL, BPD, and EFW. We also carried out the same comparisons after excluding those very premature cases complicated by toxemia and hypertension. Likewise, we compared moderately premature and nonpremature fetuses with respect to those sonographic parameters listed above, controlling for GA at sonography using ANCOVA, using measurements obtained on 24- to 29.9-week sonograms, and again using measurements obtained on 30- to 36.9-week sonograms.
In a comparison of sonograms obtained at 24 to 29.9 weeks in pregnancies that ultimately delivered very prematurely with 24- to 29.9-week sonograms in pregnancies that did not deliver prematurely, all fetal measurement parameters were significantly smaller in the group that delivered very prematurely (Table 1  , "All cases"). In particular, when GA was controlled for with the use of ANCOVA, 24- to 29.9-week fetuses who were subsequently born very premature were smaller than 24- to 29.9-week fetuses who were subsequently born nonpremature with respect to AD, FL, BPD, and EFW (P < .05, all parameters).
In 9 of the 26 pregnancies that delivered very prematurely, there were risk factors for uteroplacental insufficiency (hypertension or toxemia). Even after excluding these 9 cases (Table 1  , "Very premature without risk factors"), the remaining 17 fetuses in the very premature category had significantly smaller ADs, FLs, and EFWs than the nonpremature fetuses (P < .05, ANCOVA). There was no significant difference in BPDs.
Fetuses who were subsequently born moderately premature (between 30 and 36.9 weeks) were significantly smaller than nonpremature fetuses with respect to all sonographic parameters (AD, FL, BPD, and EFW) on sonograms obtained at 30 to 36.9 weeks’ gestation (P < .05, ANCOVA; Table 2
Our study confirms and expands on prior investigations that suggested a relationship between small fetal size and preterm delivery. A number of elements of our analysis distinguish it from prior studies. Unlike most prior studies, ours shows the relationship between fetal size and preterm delivery by directly comparing sonographic measurements in 2 groups of fetuses in a single study population: those who were subsequently born premature and those who were subsequently born at or near term. We also differentiated very premature from moderately premature pregnancies and evaluated sonographic measurements at 2 periods in pregnancy (24–29.9 and 30–36.9 weeks’ gestation), thus providing information about the timing of the lag in growth. One prior study looked at sonographic measurements in premature versus nonpremature pregnancies,1 but that study was restricted to sonograms performed at 32 weeks’ gestation and to pregnancies delivering after 32 weeks. Our study shows that the tendency to small size in premature fetuses applies to both very premature fetuses (those born before 30 weeks’ gestation) and those born moderately premature (at 30–36.9 weeks’ gestation). In particular, on scans performed at 24 to 29.9 weeks, sonographic measurements were smaller, on average, in fetuses subsequently born very premature than in fetuses subsequently born at term. Similarly, on scans performed at 30 to 36.9 weeks, sonographic measurements were smaller, on average, in fetuses subsequently born moderately premature than in fetuses subsequently born at term. Our findings also shed some light on the onset of the lag in growth of premature fetuses. Moderately premature fetuses were smaller than nonpremature fetuses with respect to all parameters (AD, FL, BPD, and EFW) on sonograms obtained shortly before delivery but only differed significantly with respect to AD and FL on earlier sonograms obtained between 24 and 29.9 weeks’ gestation. This suggests that much of the lag in growth occurs in the last few weeks before delivery. A definitive demonstration of the timing of a growth lag in premature fetuses, however, would require a different study design than ours: a study involving a large number of pregnant women scanned at close intervals. Some cases of premature delivery are precipitated by maternal complications such as hypertension or toxemia, which are risk factors for uteroplacental insufficiency and small fetal size. Even in the absence of these maternal problems, however, premature fetuses are smaller than their nonpremature counterparts. This suggests that an as yet unidentified factor may lead to both poor fetal growth and premature delivery. One consequence of our study is that the finding of small fetal size on a prenatal sonogram before term indicates an elevated risk of preterm delivery. This corroborates a prior study in which the rate of preterm delivery was found to be greater than twice as high in pregnancies with EFWs at or below the 40th percentile on 24- to 34-week sonograms compared with those in which the EFWs were above the 40th percentile.8 This is of more theoretical interest than clinical utility, however, because sonographic assessment of fetal size does not permit one to reliably predict that preterm delivery will or will not occur. The tendency to small size in fetuses born prematurely has an important, clinically relevant implication concerning interpretation of weight percentile tables. Most such tables are derived from birth weights of neonates of known GA.9–15 However, on the basis of our study and others, neonates born before term are smaller than GA-matched fetuses in utero, because fetuses born before 37 weeks are smaller, on average, than those who remain in utero. It follows that neonatal birth weight percentiles are different from fetal weight percentiles before term. In particular, before 37 weeks, more than 50% of fetuses will have weights above the 50th percentile for neonates, and fewer than 10% will have weights below the 10th percentile for neonates. Hence, if a 27-week fetus has a sonographically estimated weight at the 50th percentile in a table based on birth weights, this sonographic finding should not be interpreted as meaning that the fetal weight is at the 50th percentile for all 27-week fetuses. Instead, it should be interpreted as indicating that, if delivery were to occur shortly after the sonogram, the neonate would be expected to be at the 50th percentile for neonates of that GA. In conclusion, fetuses who are subsequently born premature are smaller than GA-matched fetuses who are born at or near term. The relationship between early delivery and small size holds even in the absence of an identifiable cause of growth restriction. The lag in growth appears to occur mainly in the last few weeks before delivery.
Received December 18, 2002, from the Departments of Radiology (P.M.D., C.B.B.), Obstetrics and Gynecology (L.W.-H.), and Newborn Medicine (S.R.), Brigham and Women’s Hospital and Harvard Medical School, Boston, Massachusetts USA. Revision requested January 2, 2003. Revised manuscript accepted for publication January 13, 2003.
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Top
Abstract
Introduction
37 Weeks)