Prenatal detection of gastrointestinal bubbles since early pregnancy: Clues to correct diagnosis

To characterize gastrointestinal bubbles detected since early pregnancy and to describe corresponding diagnoses.

than third-trimester detection as very inaccurate for jejunal and ileal atresia. 5 Duodenal obstruction can also be caused by compression from extrinsic structures such as annular pancreas 6 or malrotationassociated peritoneal fibrous bands (Ladd bands). 7 A differential diagnosis for gastrointestinal cysts or "bubbles" is enteric duplication cysts, which are rare malformations that can arise anywhere in the gastrointestinal tract. It is recommended that prenatally diagnosed enteric duplications be resected after birth as they can cause complications. 8 In the current study, we present our experience with two, three, and four gastrointestinal bubbles detected prenatally since early second trimester (14-17 weeks) and the corresponding diagnoses. 3-to 10-MHz vaginal transducer. 9 Second-trimester scans were performed transabdominally using a Samsung SW80 2-to 6-MHz abdominal transducer. 7 All fetuses with an early detection of bubbles were rescanned transabdominally at the second trimester.

| MATERIAL S AND ME THODS
Amniocentesis was performed for all fetuses with fixed bubbles and karyotype was evaluated (in one case of chromosomal microarray was also performed).
Data regarding prenatally diagnosed associated anomalies and genetic abnormalities was gathered. Data on pregnancy outcomebirth/abortion/termination of pregnancy (TOP) for confirmation of the diagnosis was retrieved, and only cases in which data were available on pregnancy outcome were included.
The study was approved by the institutional Helsinki committee (approval # 0210-10-RMB). Patient consent for participation was waived because of the retrospective nature of the study and deidentification of participants.   Ten cases of transient bubbles were detected through early scans during the study period (approximately 1:2700 of early scans). In these cases, two to four bubbles appeared and disappeared within a few seconds. Repeat scans were performed in all cases as well as postnatal follow-up. All later scans and newborns were normal.

| RE SULTS
Most of the fixed bubbles were detected through early scans.
However, in three cases, the bubbles were detected through later scans.
A fixed double bubble was detected in nine scans. In three of these cases, duodenal atresia was diagnosed without additional structural abnormalities. Two of the three duodenal atresia cases were detected in early scans, and trisomy 21 was confirmed following amniocentesis. TOP was performed. In the third case, the In this case, TOP was performed.
Two cases of fixed three bubbles were detected during the study period, both in later scans. In one case, detection of bubbles occurred during the second anatomical scan at gestational week 23 and was followed by amniocentesis with a normal karyotype.
Malrotation with obstructing Ladd bands was diagnosed postnatally and the newborn underwent a successful operative repair.
Mild polyhydramnios was detected in this case. In the second case, in which amniocentesis had been performed at gestational week 17 with a normal karyotype, the bubbles were detected during fetal growth assessment at gestational week 30, and intrauterine fetal demise was detected at 32 weeks of gestation. However, autopsy was not performed, and diagnosis could not be confirmed.
Therefore, only one case of fixed three bubbles was included in the study.
In two cases, fixed four bubbles were detected early in the second trimester. TOP was performed in both cases. Autopsy revealed jejunal atresia in both cases, with one of them being an apple peel jejunal atresia (Figure 4).

| DISCUSS ION
We have demonstrated that gastrointestinal bubbles detected since early pregnancy (14-17 weeks) can represent various conditions. Specific characteristics are important and can distinguish between normal physiological peristalsis and various pathological conditions.
In a previous publication, three cases of a transient double bubble in three normal fetuses had been reported. 10 In the current study, we have depicted more robust data of 10 cases of transient two to four bubbles in normal fetuses. We have followed all 10 fetuses with later scans and postnatally and no pathologies were detected. We therefore suggest that to confirm upper gastrointestinal obstruction it is important to demonstrate dilated bowel for a few minutes, and that a transient nature of bubbles is reassuring with no need for further investigation. Fixed bubbles have been described in many previous studies. However, previous series described cases that were detected later than 20 weeks of gestation. 5,8,11,12 We have demonstrated that when the early detected bubbles were fixed, the diagnosis was accurate. Early (14-17 weeks) transvaginal scans are vastly performed in some countries. These early scans have been reported to be diagnostic for many congenital malformations. 13 In this study, we have demonstrated early scans to be diagnostic also for high gastrointes- Also, a cephalic-posterior bubble is suspicious for esophageal duplication.
The "double wall" is a sonographic criterion for intestinal duplication cysts, which has been described for both postnatal 15,16 and prenatal 8 scans. We have identified the double wall on the later scans in cases of duodenal and esophageal duplications.
In one case, a pancreatic cyst was diagnosed. In this case, one bubble represented the stomach and the second bubble represented a pancreatic cyst. This fetus was also diagnosed with Jeune syndrome. In this syndrome, a characteristic thoracic dystrophy along with limb shortening and polydactyly can sometimes be accompanied by pancreatic cysts. 17 Therefore, the combination of sonographic features associated with Jeune syndrome along with a double bubble (without a double wall) should raise the suspicion for a pancreatic cyst.
We have demonstrated that triple bubble and quadribubble, as expected, represent a more distal obstruction. A similar finding was recently reported as the C-sign of jejunal atresia; however, during the third trimester, 18 John et al. found in their retrospective series of 58 fetuses with prenatal suspected or postnatal confirmed small bowel atresia that when small bowel atresia was suspected before the third trimester, accuracy was poor. They suggested that an ultrasound after 32 weeks should be performed to confirm the presence of both polyhydramnios and bowel dilatation. 8 As opposed to these findings, we found accurate correlation between early prenatal detection of a quadribubble and the diagnosis of jejunal atresia.
Our study is limited by the retrospective collection of data over a long period, which might lead to errors of data collection and missed cases. In addition, in this study, only one physician performed all scans. This physician is very experienced and skilled in both early and late scans and therefore the detection rate reported might not represent the detection rate in the general population, which might be lower. Despite these limitations, a relatively large number of cases is described and can be used to guide physicians who perform prenatal scans.
In conclusion, prenatal detection of intestinal bubbles can start at early transvaginal scans and can accurately discriminate between physiological and pathological upper gastrointestinal conditions. Transient bubbles are physiological, and no further investigation is needed. Fixed double bubbles are suspicious for either duodenal atresia or esophageal or duodenal duplication cysts. Three or four bubbles are suspicious for more distal intrinsic (jejunal atresia) or extrinsic (malrotation associated) obstructions.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data sharing is not applicable to this article as no new data were created or analyzed in this study.

AUTH O R CO NTR I B UTI O N S
Osnat Zmora: design, planning, data analysis, and manuscript writing. Ron Beloosesky: design, planning, conduct, data analysis, and manuscript writing. Nizar Khatib: design, planning, conduct, and manuscript revision. Yuval Ginsberg: design, planning, conduct, and manuscript revision. Ayala Gover: design, planning, and manuscript revision. Moshe Bronshtein: design, planning, conduct, data analysis, and manuscript writing.

CO N FLI C T O F I NTE R E S T S TATE M E NT
The authors declare no conflicts of interest.