Prenatal ultrasonography:
Implications for the general pediatrician

By Elaine Donoghue, MD, and John Smulian, MD, MPH

The widespread use of ultrasonography during pregnancy necessitates that pediatricians monitor conditions diagnosed before the infant was born. Understanding the significance of prenatal sonographic findings and what type of follow-up they require can be critical to both child and parents.

Approximately 4 million babies are born each year in the United States, and at least 75% of expectant mothers have ultrasonography (US) one or more times during pregnancy. That translates to some 3 million pregnancies annually in which a prenatal US examination is performed—and to about 3 million sets of parents who have clinical information about their baby before the pediatrician has even had a chance to evaluate the infant!

 

Conditions discussed in this article

Pyelectasis Isolated mild ventriculomegaly Echogenic bowel Choroid plexus cysts Echogenic cardiac foci Acknowledgment

Why are so many sonograms performed? One reason is the dramatic improvement in ultrasonographic resolution in recent years, which has made US an even more valuable screening and diagnostic tool than it had been. The medical indications for prenatal US are numerous (Table 1, available in the print edition: Adapted from: Ultrasonography in pregnancy. ACOG Technical Bulletin 1993:187). Patient demand has also increased utilization.

Results of a fetal US examination can affect clinical decisions in tangible ways. They may prompt a woman to forego an invasive test such as amniocentesis (see "The genetic sonogram: An alternative to amniocentesis?"). When a well-defined structural or functional abnormality is identified by US, obstetric or neonatal management may be altered. A finding of intrauterine growth restriction (IUGR), for example, may trigger more aggressive fetal monitoring or early delivery. Discovery of a neural tube defect might prompt the assembling of a surgical team at the time of delivery.

Many times, however, findings identified by US are subtle, and the course of follow-up action is less clear. A "finding" can be defined as any feature seen during US examination that is uncommon and deviates from the usual appearance of a fetus. Some of these "findings" are probably normal variants of fetal development and may be found in up to 10% of normal fetuses (Table 2). Some prove to be developing structural abnormalities. Many have only recently been identified and there is limited long-term, follow-up data on which to base discussions about prognosis.

 

TABLE 2 Location and frequency of abnormalities detected in prenatal ultrasonographic exams
Location of abnormality	Number*	Percentage of abnormalities attributable to specific findings*
Kidney	1,123	74% pyelectasis
Heart	785	65% echogenic foci
Head	561	66% choroid plexus cysts 13% ventriculomegaly
Umbilical cord	196	0
Abdomen	176	25% echogenic bowel
Face and neck	110	0
Extremities	106	0
Thorax	72	0
Stomach	40	0
Spine	39	0
Bladder	36	0
Abdominal wall	34	0
Total	3,350	54.6% attributable to above findings
*Abnormality or finding may not be isolated; some patients had more than one US examination
Note: Data are based on approximately 42,000 ultrasonographic examinations performed from 1997 through 1999 in the antenatal testing unit of UMDNJ–Robert Wood Johnson Medical School at Saint Peter’s University Hospital, New Brunswick, N.J.

Obstetricians have struggled with the task of counseling expectant parents when solid information is lacking. Likewise, parents have struggled to understand the counseling. Pediatricians become unwitting partners in the rapidly advancing field of ultrasonography, as they are asked to follow conditions that were diagnosed long before the child was their patient and that for the most part are discussed in the obstetric and ultrasonography literature. There is an urgent need for obstetric-pediatric partnerships to help understand the long-term implications of prenatal US findings.

When discussing this complex subject, it may be helpful to consider some of the more common ultrasonographic findings in a clinical context and place them in one of three broad categories:

• findings of known significance, such as renal pyelectasis
• findings of uncertain significance, such as mild, isolated cerebral ventriculomegaly and echogenic bowel
• findings unlikely to have clinical significance, such as echogenic cardiac foci and choroid plexus cysts.

It is important to keep in mind that "significance" is a relative term. Although decreased amniotic fluid may be significant to the obstetrician, it may not be to the pediatrician, who is not responsible for that area of care. Likewise, the significance of an increased risk of aneuploidy (an abnormal number of chromosomes) is different for the obstetrician faced with counseling a pregnant patient than it is for a pediatrician examining a phenotypically normal infant. Last, the significance to concerned parents coping with a potential problem with their baby is profoundly different. Parental anxiety may persist despite counseling and reassurance, and even after definitive prenatal testing and a normal newborn exam.

Findings of known significance

This category includes prenatal US findings serious enough to warrant evaluation after birth, even though many of these cases turn out to be normal. An example of this category is pyelectasis—mild dilation of the renal pelvis, measuring 4 to 10 mm in the anterior-posterior dimension (Figure 1).¹ The lack of dilatation of the renal calyces seen in pyelectasis typically differentiates this condition from hydronephrosis. Pyelectasis, which has an incidence of 2% to 7%, is significant because it may be caused by vesicoureteral reflux or partial urinary tract obstruction at the renal pelvis, bladder, or urethral outlet. Twenty-seven percent of cases develop into frank hydronephrosis on repeat US.² There is also a 35% increased risk for trisomy 21 when pyelectasis is found in the second trimester in high-risk mothers.

Click here to view full-size graphic

For reasons that are unclear, prenatally identified pyelectasis resolves spontaneously in many cases.³ Repeating the sonogram in the third trimester helps decrease the number of persistent cases requiring neonatal follow-up. Nevertheless, pediatricians are frequently faced with this finding and the question of what they should do. When the finding resolves before birth, no further evaluation is needed. If the finding persists, a common approach is the following⁴:

• Examine the newborn carefully for the stigmata of trisomy 21.
• Repeat the renal sonogram after the immediate postnatal period. This neonatal US should not be performed until after the first three days of life, to avoid false-negative results secondary to physiologic dehydration.
• Recommend a voiding cystourethrogram (VCUG) to rule out vesicoureteral reflux. Current recommendations are to obtain a VCUG even when the postnatal US is normal because of the high incidence of false-negative results when US is used to evaluate reflux.^4,5 This is an area of active research and debate, and recommendations may change.
• Prescribe antibiotic prophylaxis for the baby pending results of the VCUG.

The prognosis for isolated, mild pyelectasis is very good.¹

Case study. You are preparing to examine a newborn in the nursery. Pregnancy records are not available, but the mother, a 32-year-old gravida III, para II, reports having an abnormal triple screen during the pregnancy with an elevated risk for trisomy 21. Her initial and subsequent sonograms showed "extra fluid on the kidneys." Amniocentesis was performed and chromosomes were normal, according to the mother.

After you assess the newborn, you reassure the mother that the baby's physical exam was normal and there is no evidence of genetic abnormalities. The mother requests an early discharge and asks whether any special tests are needed to assess the baby's kidneys. Luckily, your hospital recently installed a computerized database of prenatal US reports. You confirm that the baby had mild, bilateral pyelectasis, which persisted on follow-up US, and that the karyotype was normal. You arrange for the baby to have a renal US in one week and a VCUG in one month and discharge the infant on prophylactic antibiotics pending the test results.

In a scenario such as this, the key actions to take are:

• Confirm prenatal findings.
• Examine the infant carefully.
• Address parental concerns.
• Arrange for follow-up diagnostic testing on an outpatient basis.

Findings of uncertain significance

This category includes findings with uncertain significance that warrant further investigation. An example is isolated mild ventriculomegaly (Figure 2), "mild" defined as cases in which the atrium of a fetal cerebral ventricle measures 10 to 15 mm in the second and third trimesters. Approximately one third of cases of frank ventriculomegaly (greater than 15 mm) are associated with a neural tube defect, 10% with aneuploidy, and more than 50% with other anomalies.⁶ The condition is also associated with congenital infections such as TORCH (toxoplasmosis, other agents, rubella, cytomegalovirus, herpes simplex).

Click here to view full-size graphic

The significance of mild ventriculomegaly that is nonprogressive and isolated is more uncertain. Although reported in only 0.15% of pregnancies scanned, it has proved to be troublesome with respect to counseling and management. In 1997, Bloom and colleagues examined 22 cases of mild, isolated ventriculomegaly in a case-control study.⁷ Cases were matched for factors known to influence development, such as age, maternal education, and household income, and Bayley and Vineland Behavior Scale evaluations were performed on the infants. Eight of 22 cases (36%) were developmentally delayed, compared with only one of the controls.

Pilu and colleagues analyzed eight other published studies and found that 8.6% of cases had malformations that were undetected on prenatal US and 11.5% had cognitive or motor delays.⁸ Vergani and colleagues found a 9% incidence of developmental delay.⁹ The range in incidence of developmental delay—from 9% to 36%—suggests that mild ventriculomegaly may serve as a subtle marker of cerebral abnormality that needs careful long-term follow-up. This likelihood should be balanced with cautious optimism, because the majority of infants do not have significant developmental delay.

Here's how to approach these cases:

• Confirm with the obstetrician that the ventriculomegaly was mild and nonprogressive.
• Confirm that no associated anomalies were detected on US.
• Find out karyotype results if a karyotype was done prenatally. If it was not, let the findings of your clinical exam (dysmorphic features) or the clinical course over time guide the decision to obtain a karyotype.
• Check whether the infant was evaluated for possible congenital infection that might cause ventriculomegaly.
• Do a careful exam for dysmorphology.
• Monitor the infant's development and head circumference carefully.
• Although the role of follow-up imaging has not been defined, consider performing a neuro-US, computed tomography, or magnetic resonance imaging on the infant in the first year of life, especially if concerns about developmental delay or abnormal head growth arise.

Another US finding of uncertain significance is echogenic bowel, generally defined as a hyperechoic, bright white appearance of fetal bowel that is similar in echodensity to fetal bone (Figure 3). This finding has been described in fetuses with aneuploidy, cystic fibrosis, congenital infections, and growth restriction, as well as in fetuses that have swallowed intra-amniotic blood. It can also be a normal variant. In 1995, MacGregor and colleagues showed that 76% of fetuses with echogenic bowel were normal; 13%, however, had growth restriction, 4% had cystic fibrosis, 4% had congenital infection, and one fetus had fetal alcohol syndrome.¹⁰ Bromley and colleagues found that 29 of 50 fetuses (58%) were normal, but eight (16%) were aneuploid and eight (16%) were growth restricted.¹¹

Click here to view full-size graphic

An approach to the follow-up of an infant with a prenatal finding of echogenic bowel would be:

• Check whether a karyotype was done prenatally. The decision to obtain a karyotype postnatally should be guided by the physical exam and the clinical course over time.
• Check whether the mother was tested for cytomegalovirus (CMV); if not, consider initiating testing if clinically indicated.
• Do a careful exam for dysmorphology.
• Follow the infant for signs and symptoms of cystic fibrosis.
• Monitor the infant's growth.

Case study. You receive notice of a newborn in the nursery. On review of the maternal pregnancy history, you note that the 24-year-old mother was a gravida I, para 0, with an unremarkable pregnancy. The infant's weight is appropriate for gestational age and the physical exam is normal. When you speak with the parents, they seem unusually anxious and focused on the baby's abdominal exam. On further questioning, they tell you that the baby had "white spots" in the abdomen on prenatal US. The mother was concerned about cystic fibrosis because she has a cousin whose child has CF. The couple declined amniocentesis because of fear of miscarriage. Genetic testing on the father was negative for known mutations, but testing on the mother was consistent with a carrier state.

After contacting the obstetrician, you learn that the mother had negative CMV serologic findings during pregnancy. You discuss the situation with the parents and inform them that, although the workup for the echogenic bowel has been reassuring so far, the baby must be monitored for CF because the father could be a carrier of an unknown mutation despite his normal test. The infant is discharged. At 5 months of age, the baby's growth is noted to be lagging. A sweat test confirms cystic fibrosis.

Findings unlikely to be clinically significant

Findings in this category may be artifacts of the imaging process or variants of normal fetal development. One such finding is choroid plexus cysts, seen as discrete hypoechoic (dark) areas within the choroid plexus (Figure 4). The choroid plexus produces cerebrospinal fluid; disruption of secretion of CSF can lead to cyst formation. The choroid plexus is easiest to examine by US in the early second trimester, when most choroid plexus cysts are found.

Click here to view full-size graphic

Choroid plexus cysts have a 2% to 4% incidence and typically resolve spontaneously by 24 weeks of gestation. They can be unilateral or bilateral and of any size.¹² These cysts were originally described in association with trisomy 18; they are identified in the second trimester in approximately 30% of trisomy 18 fetuses. However, a fetus with trisomy 18 usually has multiple anomalies identified by US. Yeo and colleagues found that, when 18 fetuses with trisomy 18 were scanned in the second trimester, the median number of abnormalities seen was seven, with no fetus having fewer than four anomalies.¹³ Digiovanni and colleagues, in their 1997 study, looked at 85 cases of isolated choroid plexus cysts.¹⁴ Karyotype or phenotype was normal in all the infants. They also checked development in 89% of the cases by DDST-II (Denver Developmental Screening Test II); all subjects had a normal screening exam.

What should a pediatrician do when confronted with an infant in whom isolated choroid plexus cysts were identified in the second trimester? If the neonatal exam is normal, simply reassure the parents. No special follow-up or evaluation is needed.

Another finding unlikely to be clinically significant is echogenic cardiac foci. ECF are bright white foci within the heart that appear around the papillary muscles of the atrioventricular valves in either ventricle (Figure 5). ECF are relatively common and are identified in 2% to 7% of all sonograms. They may represent thickening of the chordae tendineae, incomplete fenestration of the chordae, fibrosis of the papillary muscle, or microcalcifications. Rarely, they may be associated with cardiac tumors or other cardiac anomalies. ECF may increase the risk estimate for trisomy 21 in high-risk women. They can be a transient finding, however, and may be dependent on examiner technique and equipment capabilities.

Click here to view full-size graphic

For pediatricians, the significance of isolated ECF in a baby with normal chromosomes has been problematic. To address this issue, Dildy and colleagues conducted a prospective, case-controlled study in which they performed echocardiograms on 25 babies who had isolated ECF and normal chromosomes identified prenatally, and on their controls.¹⁵ They found that about half of the infants had persistent ECF but none had any clinically significant cardiac problem. Many other studies also conclude that ECF are a normal variant.^16–18

What, then, should the pediatrician do with an infant in whom echogenic cardiac foci have been found?

• Determine what kind of evaluation was done prenatally.
• Confirm that no other, associated anomalies were noted on prenatal US.
• Do a careful exam for stigmata of trisomy 21; if your exam is normal, reassure the parents.

If all these actions yield information that is reassuring, no special follow-up is necessary. Routine pediatric echocardiogram and karyotype are not indicated.

Case study. A new mother brings her 2-week-old to your office. You review the chart and note that the mother, a 39-year-old gravida I, para 0, had an uneventful pregnancy. The prenatal US, performed because of her age, was normal except for an echogenic cardiac focus. Amniocentesis was declined. Labor and delivery were normal and the physical exam was unremarkable. The baby is feeding well and gaining weight appropriately and the exam today is normal. The mother admits that she feels anxious every time the baby cries because it may "strain the heart."

Since the exam is normal, you reassure the mother that there is no problem with the infant's heart and note that further reassurance may be necessary on subsequent visits.

The importance of communication

Follow-up of prenatal US findings often involves obtaining detailed information about the pregnancy. This can be difficult to do, given busy physician schedules and complex cross-coverage arrangements. No standard protocol exists to optimize communication between pediatricians and obstetricians; each facility has its own special considerations.

A communication breakdown can have serious implications for the child's health if critical information is not transmitted.¹⁹ The potential for that to happen exists in many cases: One study that looked at the charts of newborns in whom a renal abnormality had been detected on prenatal ultrasound found that only 38% to 56% of these records contained documentation of the abnormal US finding.³

Suggestions for supplementing direct physician-to-physician communication include using computerized US databases and special flagging systems to identify infants who will need follow-up of sonogram findings. It may be helpful to target specific fetal abnormalities, such as the ones discussed in this article, as a starting point for tracking in a database. Parent-centered communication systems, such as requesting that parents bring any prenatal US reports to the first office visit, may help. Some clinicians have suggested that a medical record be started on the fetus before birth and be continued in the postpartum-pediatric period.¹⁹ Whatever communication system is used, it should be sensitive to the patient's need for confidentiality.

Last, the reliability and accuracy of prenatal US is variable: equipment capabilities and image resolution; gestational age, position of the fetus, and the presence of multiple fetuses; maternal factors such as obesity and uterine fibroids; and amniotic fluid volume can all affect findings.

The training and skill of the sonologist and his or her expertise in the area of prenatal diagnosis also have a significant impact on the accuracy of a sonogram and, in turn, on how a pregnant patient is counseled and managed. Ultrasonographic examinations of the fetus may be performed by a maternal-fetal medicine specialist (perinatologist), obstetrician, radiologist, or family practitioner, each of whom has different training and skills in prenatal diagnosis. The amount of prenatal information available to the pediatrician may vary for the same condition, depending on who did the prenatal evaluation.

Many issues raised by prenatal testing are resolved by the time the baby is born. The task of bringing closure to some of the issues that remain unresolved may fall on the shoulders of the pediatrician. He or she needs to be prepared to meet that responsibility.

 

Acknowledgment The authors wish to acknowledge Frances Rhoads, MD, Sooze Hodgson, MD, and Anthony Vintzileos, MD, for reviewing the manuscript of this article before submission.

REFERENCES

1. Broadley P, McHugo J, Morgan I, et al: The 4-year outcome following the demonstration of bilateral renal pelvic dilatation on prenatal renal ultrasound. Br J Radiol 1999;72(855):265

2. Persutte WH, Koyle M, Lenke RR, et al: Mild pyelectasis ascertained with prenatal ultrasonography is pediatrically significant. Ultrasound Obstet Gynecol 1997; 10(1):12

3. Yeo L, Smulian JC, Vintzileos AM, et al: Isolated renal abnormalities: Sonographic progression and postnatal pediatric awareness. Am J Obstet Gynecol 1998;S158: Abstract #553

4. Elder JS: Antenatal hydronephrosis: Fetal and neonatal management. Pediatr Clin North Am 1997;44(5):1299

5. Tibballs JM, DeBruyn R: Primary vesicoureteric reflux: How useful is postnatal ultrasound? Arch Dis Child 1996;75(5):444

6. Bromley B, Frigoletto FD Jr, Benacerraf BR: Mild fetal lateral ventriculomegaly: Clinical course and outcome. Am J Obstet Gynecol 1991;164(3):863

7. Bloom SL, Bloom DD, Dellanebbia C, et al: The developmental outcome of children with antenatal mild isolated ventriculomegaly. Obstet Gynecol 1997;90(1):93

8. Pilu G, Falco P, Gabrielli S, et al: The clinical significance of fetal isolated cerebral borderline ventriculomegaly: Report of 31 cases and review of the literature. Ultrasound Obstet Gynecol 1999;14(5):320

9. Vergani P, Locatelli A, Strobelt N, et al: Clinical outcome of mild fetal ventriculomegaly. Am J Obstet Gynecol 1998;178(2):218

10. MacGregor SN, Tamura R, Sabbagha R, et al: Isolated hyperechoic fetal bowel: Significance and implications for management. Am J Obstet Gynecol 1995;173(4):1254

11. Bromley B, Doubilet P, Frigoletto FD Jr, et al: Is fetal hyperechoic bowel on second-trimester sonogram an indication for amniocentesis? Obstet Gynecol 1994;85(5):647

12. Gupta JK, Cave M, Lilford RJ, et al: Clinical significance of fetal choroid plexus cysts. Lancet 1995;346(8977):724

13. Yeo L, Guzman ER, Vintzileos AM, et al: The value of a complete anatomic survey in the sonographic detection of trisomy 18. Am J Obstet Gynecol 1999;180(S 166): Abstract #585

14. DiGiovanni LM, Quinlan MP, Verp MS: Choroid plexus cyst: Infant and early childhood development outcome. Obstet Gynecol 1997;90(2):191

15. Dildy GA, Judd VE, Clark SL: Prospective evaluation of the antenatal incidence and postnatal significance of the fetal echogenic cardiac focus: A case-control study. Am J Obstet Gynecol 1996;175(4):1008

16. Petrikovsky BM, Challenger M, Wyse Li: Natural history of echogenic foci within ventricles of the fetal heart. Ultrasound Obstet Gynecol 1995;5(2):92

17. Simpson JM, Cook A, Sharland G: The significance of echogenic foci in the fetal heart: A prospective study of 228 cases. Ultrasound Obstet Gynecol 1996;8(4):225

18. Merati R, Lovotti M, Norchi S, et al: Prevalence of fetal left ventricular hyperechogenic foci in a low-risk population. Br J Obstet Gynaecol 1996;103(11):1102

19. Craig JC, Knight JF, Smith GH: Communication breakdown: A preventable cause of acute renal failure in a newborn infant. Med J Aust 1996;164(11):663

DR. DONOGHUE is clinical associate professor of pediatrics, UMDNJ–Robert Wood Johnson Medical School, New Brunswick, N.J., and attending pediatrician, St. Peter's University Hospital, New Brunswick.

DR. SMULIAN is associate professor of obstetrics, gynecology, and reproductive sciences, UMDNJ– Robert Wood Johnson Medical School, and director of perinatal research, St. Peter's University Hospital, New Brunswick.

The genetic sonogram: An alternative to amniocentesis?

Women who have an identified risk factor for aneuploidy—advanced maternal age, for example, or an abnormal second trimester biochemical triple screening—may choose to have an invasive diagnostic procedure such as amniocentesis or chorionic villus sampling for karyotyping. The risk of miscarriage with these procedures may be as high as 1%, a risk many women are unwilling to take.

Evaluating fetuses for subtle, soft markers of aneuploidy such as trisomy 21 or 18 is a relatively recent trend in ultrasonography.¹ Development of the "genetic sonogram" has given women an alternative to invasive testing. In effect, the antenatal sonogram can be used as a fetal dysmorphology examination.

Many dysmorphic features identified are not structural abnormalities, such as congenital heart disease, but rather variants of normal anatomy, such as clinodactyly or flattening of the occiput. When these or other markers are identified—or not identified—the risk for fetal aneuploidy can be adjusted based on the findings. A normal genetic sonogram may reduce the risk estimate for Down syndrome by as much as 80%; the presence of even one aneuploidy marker may increase the risk estimate over baseline.²

Women with an increased risk for aneuploidy who have a normal genetic sonogram may decide to forego amniocentesis, but approximately 3% still decide to have definitive karyotype via amniocentesis. Conversely, about 30% of women decide not to have amniocentesis even when their risk of Down syndrome remains elevated.³ Because no absolute indications exist for amniocentesis, decisions about invasive testing rest ultimately with the parents.

REFERENCES

1. Vintzileos AM, Campbell WA, Rodis JF, et al: The use of second-trimester genetic sonogram in guiding clinical management of patients at increased risk for fetal trisomy 21. Obstet Gynecol 1996;87(6):948

2. Sohl BD, Scioscia AL, Budorick NE, et al: Utility of minor ultrasonographic markers in the prediction of abnormal fetal karyotype at a prenatal diagnostic center. Am J Obstet Gynecol 1999;181(4):898

3. Vintzileos AM, Guzman EF, Smulian JC, et al: Choice of second-trimester genetic sonogram for detection of Trisomy 21. Obstet Gynecol 1997;90(2):187

 

Elaine Donoghue, John Smulian. Prenatal ultrasonography: Implications for the general pediatrician. Contemporary Pediatrics 2002;1:29.