Abstract
BACKGROUND:
A sonographically large fetal stomach has been associated with gastrointestinal obstruction, per case reports, and is often followed up with serial ultrasound examinations. The frequency of this phenomenon has not been systematically studied, resulting in challenges in counseling parents about the prognosis and making cost-benefit analysis of serial ultrasound follow-up difficult to assess.
OBJECTIVE:
This study aimed to determine the frequency at which an enlarged fetal stomach as the sole abnormality on fetal ultrasound reflects a bowel obstruction to aid in parental counseling and determine the best practice for follow-up.
STUDY DESIGN:
We performed a retrospective cohort study of all prenatal sonographic cases in which a large fetal stomach was visualized between January 1, 2002, and June 1, 2016. The inclusion criteria required a fetal diagnosis of a large stomach, defined as an increased measurement in ≥2 dimensions based on a nomogram, that resulted in a liveborn delivery within the Johns Hopkins Health System. We excluded pregnancy loss, pregnancy termination, and cases delivered outside of the Johns Hopkins Health System. Cases were subclassified as isolated or complex based on the absence or presence of additional ultrasound findings at initial presentation of the enlarged stomach. The perinatal outcomes and maternal demographics were determined and compared between isolated and complex cases.
RESULTS:
Of 57,346 total cases with ultrasound examinations in the Johns Hopkins Health System within the study time frame, 348 fetuses had enlarged stomachs, with 241 (69.3%) who met the inclusion criteria as follows: 161 (66.8%) isolated and 80 (33.2%) complex. Of the 161 isolated cases, 1 resulted in neonatal small bowel obstruction (0.62%). Of note, 158 of the isolated large stomach cases (98.1%) had no postnatal abnormalities of any kind. Of the 80 complex cases, 18 (22.5%) resulted in neonatal gastrointestinal obstruction (14 cases of duodenal atresia and 4 cases of jejunal atresia). Those with isolated findings were significantly less likely to deliver preterm (n=24 [14.9%] vs n=35 [43.8%]; P<.001), be complicated by polyhydramnios (n=18 [11.2%] vs n=23 [28.8%]; P<.001), have a neonatal intensive care unit admission (n=31 [19.3%] vs n=76 [95.0%]; P<.01), or have a major surgical procedure (n=2 [1.2%] vs n=66 [82.5]; P<.001) compared with complex cases.
CONCLUSION:
We found that 0.62% of isolated large fetal stomachs (1 of 161) were associated with neonatal intestinal obstruction. Of the complex cases with an enlarged stomach, 18 of 80 (22.5%) were found to have a gastrointestinal obstruction; by definition, none of these complex cases began as an isolated large stomach as their initial ultrasound finding, but rather had other concurrent sonographic abnormalities, including a double bubble sign and intestinal dilation. With a prevalence of <1% resulting in the development of a small bowel obstruction, our results suggest that, when isolated, a large stomach does not seem to warrant serial prenatal ultrasound follow-up or postnatal imaging and is likely to reflect an incidental finding.
Keywords: duodenal atresia, enlarged fetal stomach, fetal anomaly, prenatal counseling, small bowel obstruction, stomach enlargement, ultrasound, ultrasound markers
Introduction
Fetal gastrointestinal (GI) obstruction is a serious condition requiring a neonatal intensive care unit (NICU) admission for surgical correction.1 There are various etiologies, such as small and large bowel atresia, malrotation, and volvulus. Prenatal sonographic diagnosis aids in appropriate referral to a tertiary care center, expeditious neonatal management, and parental counseling.2–4
Sonographically, the pathognomonic finding of duodenal atresia is a “double bubble sign,” indicating the presence of the stomach adjacent to a second noncommunicating bubble, the duodenum.1,2,4 However, these and more distal bowel obstructions can present more insidiously and be difficult to diagnose.1,2,5 For example, per case reports, an isolated enlarged may be suggestive of early obstruction and is reported to parents as a potential marker for GI obstruction with serial ultrasound (US) examinations often recommended.6,7
Mothers-to-be have increased anxiety after being notified of a sonographic marker such as this, which may have negative health impacts on the fetus and pregnancy.8–10 Thus, minimization of this potential iatrogenic harm must be attempted via appropriate prognostic counseling. To the best of our knowledge, the frequency at which a large stomach is a marker for intestinal obstruction has not yet been determined, making prenatal counseling challenging.11 In addition, the recommendation for serial US examinations increases medical costs with undetermined cost-benefit analysis.12
We sought to quantify the association of a large fetal stomach and formation of GI obstruction to provide better prognostic counseling for this finding and potentially reduce patient anxiety. We examined outcomes in fetuses with isolated enlarged fetal stomachs and those with other sonographic abnormalities accompanying a large stomach to better understand the frequency and types of associated anomalies and their outcomes compared with those with an isolated large stomach.
Materials and Methods
Study design and population
This is a retrospective database cohort of cases with a large fetal stomach on prenatal US at any given gestational age born between January 1, 2002, and June 1, 2016, in the Johns Hopkins Health System. Under the institutional review board (IRB) approval, cases with stomach enlargement were selected from our prenatal US database. All cases were reevaluated with stomach dimensions verified by the authors. The inclusion criteria required a fetal stomach of >2 standard deviations above the mean in at least 2 planes based on a published stomach size nomogram.11 US evaluations were performed by American Registry for Diagnostic Medical Sonography–certified sonographers on a GE Voluson E8 or E10 machine (General Electric, Boston, MA) accredited by the American Institute of Ultrasound in Medicine.
Cases were stratified by the presence or absence of other sonographic findings. Those devoid of other sonographic findings aside from a large stomach were classified as isolated whereas those with an additional abnormality at the time of the first enlarged fetal stomach were classified as complex. These formed the 2 comparison groups for this study.
Data collection
Demographics and outcomes were extracted from the medical record of both mothers and infants (Table 1, Table 2). Indications for delivery included preeclampsia, preterm premature ruptured membranes, non-reassuring fetal status, and other obstetrical considerations. Isolated cases were stratified by enlarged stomach resolution or persistence before birth, and outcomes were compared (Table 3). Complex cases were stratified into those with GI-related postnatal diagnoses and those whose postnatal diagnoses were not GI-related (GI vs non-GI) (Table 4). Isolated cases were assessed for NICU admission and were further statistically evaluated with respect to maternal demographics and postnatal outcomes (Tables 5 and 6).
TABLE 1.
Complex vs isolated maternal demographics
| Isolated (n=161) | Complex (n=80) | P value | |
|---|---|---|---|
| Maternal age, y, mean (range) | 29.1 (16–45) | 27.4 (16–42) | .05a |
| Race | .02b | ||
| White | 66 (41.0) | 49 (61.3) | |
| African American | 63 (39.1) | 23 (28.8) | |
| Asian | 23 (14.3) | 4 (5.0) | |
| Other | 9 (5.6) | 4 (5.0) | |
| Gravity and parity | .13b | ||
| Nulliparous | 43 (26.7) | 29 (36.3) | |
| Multiparous | 118(73.3) | 51 (63.8) | |
| Previous abortion or miscarriage | .88b | ||
| Yes | 68 (42.2) | 33 (41.3) | |
| No | 93 (57.8) | 47 (58.8) | |
| Maternal diabetes | .13c | ||
| Pregestational | 3(1.9) | 2 (2.5) | |
| Gestational | 10(6.2) | 6 (7.5) | |
| None | 139(86.3) | 72 (90.0) | |
| Not available | 9 (5.6) | 0(0) | |
| Infection during pregnancy | .93b | ||
| Yes | 59 (36.0) | 35 (43.8) | |
| UTI | 8 | 3 | |
| HIV | 2 | 0 | |
| GBS | 43 | 19 | |
| Chlamydia | 9 | 5 | |
| HSV | 8 | 5 | |
| Other | 8 | 9 | |
| No | 88 (54.7) | 34 (425) | |
| Not available | 14(8.7) | 11 (13.8) | |
| Substance use during pregnancy | .21b | ||
| Yes | 15 (9.3) | 8(10.0) | |
| Tobacco | 13 | 8 | |
| Opiates | 2 | 3 | |
| Marijuana | 8 | 3 | |
| Other | 0 | 3 | |
| No | 137(85.1) | 70 (87.5) | |
| Not available | 9 (5.6) | 2 (2.5) | |
| BMI | .62c | ||
| Underweight | 6 (3.7) | 2 (2.5) | |
| Normal | 65 (40.4) | 36 (45.0) | |
| Overweight or obese | 65 (40.4) | 33 (41.3) | |
| Morbidly obese | 12(7.5) | 2 (2.5) | |
| Not available | 13 (8.1) | 7 (8.8) |
Values are expressed as number (percentage) unless indicated otherwise.
BMI, body mass index; GBS, Guillain-Barre syndrome; HSV, herpes simplex virus; UTI, urinary tract infection.
Student t test;
Chi-squared analysis;
Fisher exact test.
TABLE 2.
Complex vs isolated postnatal outcomes
| Isolated (n=161) | Complex (n=80) | Pvalue | |
|---|---|---|---|
| Sex | .39a | ||
| Male | 71 (44.1) | 40 (50.0) | |
| Female | 90 (55.9) | 40 (50.0) | |
| Weight, kg, mean (range) | 3.15 (1.38–4.66) | 2.70 (1.63–4.66) | <.01b |
| Preterm | 2.30 | 2.31 | .94b |
| Term | 3.30 | 3.00 | <.01b |
| Preterm or term birth | <.001a | ||
| Preterm | 24 (14.9) | 35 (43.8) | |
| Term | 137(85.1) | 45 (56.3) | |
| Gestational age at delivery, wk, mean (range) | 38.5 (28.1–41.3) | 36.8 (31–41.3) | <.01b |
| Amniotic fluid | <.001c | ||
| Normal | 143 (88.8) | 54 (67.5) | |
| Polyhydramnios | 18 (11.2) | 23 (28.8) | |
| Oligohydramnios | 0 (0) | 3 (3.8) | |
| Stomach size resolved before birth? | <.01a | ||
| Yes | 106 (65.8) | 21 (26.3) | |
| No | 55 (34.2) | 59 (73.8) | |
| Mode of Delivery | .43a | ||
| Vaginal | 101 (62.7) | 46 (57.5) | |
| Cesarean | 60 (37.3) | 34 (42.5) | |
| Medical indication for delivery | <.01c | ||
| None | 109(67.7) | 39 (48.8) | |
| PREEC | 8 (5.0) | 2 (2.5) | |
| PPROM | 12 (7.5) | 5 (6.25) | |
| Nonreassuring fetal status | 21 (13.0) | 28 (35.0) | |
| Other | 10 (6.2) | 6 (7.5) | |
| Not available | 1 (0.6) | 0 (0) | |
| NICU admission | <.01c | ||
| Yes | 31 (19.3) | 76 (95.0) | |
| Preterm | 17 (54.8) | 34 (44.7) | <.01c |
| Term | 14 (45.2) | 42 (55.3) | <.01c |
| No | 130 (80.7) | 3 (3.8) | |
| Not applicable | 0 (0) | 1 (1.3) | |
| LOS, d | |||
| Mean (all) | 2.21 | 41.4 | <.01b |
| Mean (NICU) | 11.77 | 43.01 | <.01b |
| Imaging indicated | <.01c | ||
| Yes | 47 (29.2) | 79 (98.8) | |
| Abdominal US | 22 | 63 | |
| X-ray | 31 | 74 | |
| Echo | 7 | 5 | |
| Other | 1 | 3 | |
| No | 114 (70.8) | 0 (0) | |
| Not applicable | 0 (0) | 1d (1.3) | |
| Surgery indicated | <.001c | ||
| Yes | 2 (1.2)b | 66 (82.5) | |
| No | 159 (98.8) | 13 (16.3) | |
| Not applicable | 0 (0) | 1e (1.3) | |
| Postnatal diagnosis | <.001a | ||
| None | 158 (98.1) | 0 (0) | |
| GI | 1 (0.62) | 56 (70.0) | |
| Duodenal atresia or web | 1 | 14 (25.0) | |
| Duodenal atresia+heterotaxy | 0 | 2 | |
| Duodenal atresia+trisomy 21 | 0 | 3 | |
| Duodenal atresia+other | 0 | 1 | |
| Jejunal atresia | 0 | 4 7.1) | |
| Duodenal hernia | 0 | 4 (7.1) | |
| Gastroschisis | 0 | 31 (55.4) | |
| Omphalocele | 0 | 2 (3.6) | |
| Heterotaxy only | 0 | 1 (1.8) | |
| Non-GI | 2 (1.2) | 24 (30.0) | |
Values are expressed as number (percentage) unless indicated otherwise.
GI, gastrointestinal; LOS, length of stay; NICU, neonatal intensive care unit; PPROM, preterm premature rupture of the membranes; PREEC, preeclampsia; US, ultrasound.
Chi-squared analysis;
Student t test analysis;
Fisher exact test analysis;
One complex case with severe fetal malformation, including renal agenesis, complex cardiac anomalies, a large stomach, and fetal hydrops was born and received comfort care only owing to poor prognosis and therefore received no NICU admission, imaging, or postnatal surgical procedure;
The 2 surgeries indicated in the isolated group were 1 aortic stenosis repair and 1 repair of a duodenal.
TABLE 3.
Isolated large stomach outcomes stratified by resolution of enlarged stomach before birtha
| Resolved (n=106) | Not resolved (n=55) | Pvalue | |
|---|---|---|---|
| Sex | .67b | ||
| Male | 48 (45.3) | 23 (41.8) | |
| Female | 58 (54.7) | 32 (58.2) | |
| Weight, kg, mean (range) | 3.19 (1.62–4.45) | 3.06 (1.47–4.66) | .19c |
| Preterm or term birth | .31b | ||
| Preterm | 18(17.0) | 6 (10.9) | |
| Term | 88 (83.0) | 49(89.1) | |
| Gestational age at delivery, wk, mean (range) | 38.6 (32.6–41.3) | 38.3 (28.1–41.1) | .13c |
| Amniotic fluid | <.01c | ||
| Normal | 102 (96.2) | 41 (74.5) | |
| Polyhydramnios | 4 (3.8) | 14(25.5) | |
| Oligohydramnios | 0(0) | 0(0) | |
| Mode of delivery | .39c | ||
| Vaginal | 69(65.1) | 32 (58.2) | |
| Cesarean | 37 (34.9) | 23 (41.8) | |
| Medical indication for delivery | .51d | ||
| None | 76 (71.7) | 33 (60.0) | |
| PREEC | 5 (4.7) | 3 (5.5) | |
| PPROM | 7 (6.6) | 5 (9.1) | |
| Nonreassuring fetal status | 13 (12.3) | 8 (14.5) | |
| Other | 5 (4.7) | 5 (9.1) | |
| Not available | 0(0) | 1 (1.8) | |
| NICU admission | .31d | ||
| Yes | 18(17.0) | 13 (23.6) | |
| Preterm | 11 | 4 | |
| Term | 7 | 9 | |
| No | 88 (83.0) | 42 (76.4) | |
| Not applicable | 0(0) | 0(0) | |
| LOS, d | |||
| Mean (all) | 1.25 | 4.08 | .04c |
| Mean (NICU) | 7.38 | 18.36 | .07c |
| Imaging indicated | .282d | ||
| Yes | 28 (26.4) | 19(34.5) | |
| No | 78 (73.6) | 36 (65.5) | |
| Not applicable | 0(0) | 0(0) | |
| Surgery indicated | .16d | ||
| Yes | 0(0) | 2 (3.6) | |
| No | 106 (100.0) | 53 (96.4) | |
| Not applicable | 0(0) | 0(0) | |
| Postnatal diagnosis | .34b | ||
| None | 106 (100.0) | 54 (98.2) | |
| GI | 0(0) | 1 (1.8) | |
| Duodenal atresia or web | 0(0) | 1 | |
| Non-GI | 0(0) | 0(0) | |
Values are expressed as number (percentage) unless indicated otherwise.
GI, gastrointestinal; LOS, length of stay; NICU, neonatal intensive care unit; PPROM, preterm premature rupture of the membranes; PREEC, preeclampsia.
The isolated large stomach group was divided into those whose enlarged stomachs resolved to within normal limits before birth and those whose did not, and the postnatal outcomes were analyzed between those 2 subgroups;
Chi-squared analysis;
Student t test analysis;
Fisher exact test analysis.
TABLE 4.
Complex case outcomes stratified by GI vs non-GI postnatal diagnosisa
| Non-GI (n=28) | GI (n=52) | Pvalue | |
|---|---|---|---|
| Sex | 1.0b | ||
| Male | 14(50.0) | 26 (50.0) | |
| Female | 14(50.0) | 26 (50.0) | |
| Weight, kg, mean (range) | 3.01 (2.06–3.75) | 2.52 (1.30–3.97) | <.01c |
| Preterm or term birth | <.01b | ||
| Preterm | 1 (3.6) | 34 (85.4) | |
| Term | 27 (96.4) | 18(34.6) | |
| Gestational age at delivery, wk, mean (range) | 38.5 (31.1–41.0) | 35.9 (29.6–41.3) | <.01d |
| Amniotic fluid | .28a | ||
| Normal | 22 (78.6) | 32 (61.5) | |
| Polyhydramnios | 5 (17.9) | 18(34.6) | |
| Oligohydramnios | 1 (3.6) | 2 (3.8) | |
| Stomach size resolved before birth? | <.01c | ||
| Yes | 18(64.3) | 3 (5.8) | |
| No | 10(35.7) | 49 (94.2) | |
| Mode of delivery | .60c | ||
| Vaginal | 15 (53.6) | 31 (59.6) | |
| Cesarean | 13 (46.4) | 21 (40.4) | |
| Medical indication for delivery | <.01d | ||
| None | 22 (78.6) | 17(32.7) | |
| PREEC | 0(0) | 2 (3.8) | |
| PPROM | 0(0) | 5 (9.6) | |
| Nonreassuring fetal status | 3 (10.7) | 25 (48.1) | |
| Other | 3 (10.7) | 3 (5.8) | |
| Not available | 0(0) | 0(0) | |
| NICU admission | |||
| Yes | 25 (89.3) | 51 (98.1) | .04d |
| Preterm | 1 | 33 | .26d |
| Term | 24 | 18 | 1.0d |
| No | 3 (10.7) | 0(0) | |
| Not applicable | 0(0) | 1e (1.9) | |
| LOS, d | |||
| Mean (all) | 26.3 | 49.6 | <.01c |
| Mean (NICU) | 29.5 | 49.6 | .03c |
| Imaging indicated | 1.0d | ||
| Yes | 28 (100.0) | 51 (98.1) | |
| No | 0(0) | 0(0) | |
| Not applicable | 0(0) | 1e (1.9) | |
| Surgery indicated | <.01d | ||
| Yes | 16(57.1) | 50 (96.2) | |
| No | 12 (42.9) | 1 (1.9) | |
| Not applicable | 0(0) | 1e (1.9) | |
| Postnatal diagnosis | Not applicable | ||
| None | 0(0) | 0(0) | |
| GI | 0(0) | 52 (100.0) | |
| Duodenal atresia or web | 0(0) | 14(26.9) | |
| Duodenal atresia+heterotaxy | 0 | 2 | |
| Duodenal atresia+trisomy 21 | 0 | 3 | |
| Duodenal atresia+other | 0 | 1 | |
| Jejunal atresia | 0(0) | 4 (7.7) | |
| Gastroschisis | 0(0) | 31 (59.6) | |
| Omphalocele | 0(0) | 2 (3.8) | |
| Heterotaxy only | 0(0) | 1 (1.9) | |
| Non-GI | 28 (100.0) | 0(0) | |
Values are expressed as number (percentage) unless indicated otherwise.
GI, gastrointestinal; LOS, length of stay; NICU, neonatal intensive care unit; PPROM, preterm premature rupture of the membranes; PREEC, preeclampsia.
The complex large stomach group was divided into those with postnatal diagnoses involving the GI system and those whose postnatal diagnoses did not involve the GI system, and the postnatal outcomes were compared between these 2 subgroups. GI diagnoses included heterotaxy, omphalocele, gastroschisis, and small bowel atresias. Non-GI cases included diaphragmatic hernia and renal, cranial, and cardiac anomalies;
Chi-squared analysis;
Student t test analysis;
Fisher exact test analysis;
One complex case with severe fetal malformation, including renal agenesis, complex cardiac anomalies, a large stomach, and fetal hydrops was born and received comfort care only owing to poor prognosis and therefore received no NICU admission, imaging, or postnatal surgical procedure.
TABLE 5.
Demographics for isolated cases with NICU vs no NICU admissiona
| NICU admission (n=31) | No NICU admission (n=130) | P value | |
|---|---|---|---|
| Maternal age, y, mean (range) | 30.9 (17–42) | 28.7 (16–45) | .10b |
| Race | .06c | ||
| Black | 8 (25.8) | 55 (42.3) | |
| White | 19(61.3) | 47 (36.2) | |
| Asian | 4(12.9) | 19(14.6) | |
| Other | 0(0) | 9 (6.9) | |
| Gravity and parity | .56c | ||
| Nulliparous | 7 (22.6) | 36 (27.7) | |
| Multiparous | 24 (77.4) | 94 (72.3) | |
| Previous abortion or miscarriage | .40c | ||
| Yes | 11 (35.5) | 57 (43.8) | |
| No | 20 (64.5) | 73 (56.2) | |
| Maternal diabetes | .85d | ||
| Pregestational | 0(0) | 3 (2.3) | |
| Gestational | 1 (3.2) | 9 (6.9) | |
| None | 29 (93.5) | 110(84.6) | |
| Not available | 1 (3.2) | 8 (6.2) | |
| Infection during pregnancy | .93c | ||
| Yes | 9 (29.0) | 50 (38.5) | |
| UTI | 0 | 8 | |
| HIV | 2 | 4 | |
| GBS | 5 | 34 | |
| Chlamydia | 1 | 8 | |
| HSV | 1 | 7 | |
| Other | 1 | 7 | |
| No | 18(58.1) | 70 (53.8) | |
| Not available | 4(12.9) | 10(7.7) | |
| Substance use during pregnancy | .141c | ||
| Yes | 6(19.4) | 9 (6.9) | |
| Tobacco | 6 | 7 | |
| Opiates | 1 | 1 | |
| Marijuana | 4 | 6 | |
| Other | 0 | 0 | |
| No | 24 (77.4) | 113 (86.9) | |
| Not available | 1 (3.2) | 8 (6.9) | |
| BMI | .18c | ||
| Underweight | 1 (3.2) | 5 (3.8) | |
| Normal | 8 (25.8) | 57 (43.8) | |
| Overweight or obese | 18(58.1) | 47 (36.2) | |
| Morbidly obese | 1 (3.2) | 11 (8.5) | |
| Not available | 3 (9.7) | 10(6.2) |
Values are expressed as number (percentage) unless indicated otherwise.
BMI, body mass index; GBS, Guillain-Barre syndrome; HSV, herpes simplex virus; NICU, neonatal intensive care unit; UTI, urinary tract infection.
The isolated large stomach group was divided into those who were admitted to the NICU and those who were not, and the maternal demographics were analyzed between these 2 comparison groups;
Student t test;
Chi-squared analysis;
Fisher exact test.
TABLE 6.
Outcomes for isolated cases with NICU vs no NICU admissiona
| NICU admission (n=31) | No NICU admission (n=130) | P value | |
|---|---|---|---|
| Sex | .50b | ||
| Male | 12 (38.7) | 59 (45.4) | |
| Female | 19(61.3) | 71 (54.6) | |
| Weight, kg, mean (range) | 2.57 (1.38–4.25) | 3.29 (2.14–4.54) | <.01c |
| Preterm | 2.09 | 2.81 | <.01c |
| Term | 3.15 | 3.32 | .24c |
| Preterm or term Birth | <.01b | ||
| Preterm | 17(54.8) | 7 (5.4)d | |
| Term | 14(45.2) | 123 (94.6) | |
| Gestational age at delivery, wk, mean (range) | 36.2 (28.1–41) | 39.1 (35–41.3) | <.01c |
| Preterm | 33.9 (28.1–36.7) | 36.3 (35–36.9) | <.01c |
| Term | 39.1 (37–41) | 39.2 (37–41.3) | .54c |
| Amniotic | .35e | ||
| Fluid | |||
| Normal | 26 (83.9) | 117(90.0) | |
| Polyhydramnios | 5 (16.1) | 13 (10.0) | |
| Oligohydramnios | 0(0) | 0(0) | |
| Stomach size resolved before birth? | .31b | ||
| Yes | 18(58.1) | 88 (67.7) | |
| No | 13 (41.9) | 42 (32.3) | |
| Mode of delivery | .01b | ||
| Vaginal | 12 (38.7) | 89 (68.5) | |
| Cesarean | 19(61.3) | 41 (31.5) | |
| Medical indication for delivery | 16(51.6) | 93 (58.1) | .04e |
| None | 3 (9.7) | 5(3.1) | |
| PREEC | 6(19.4) | 6 (3.8) | |
| PPROM | 5 (16.1) | 16(10.0) | |
| Nonreassuring fetal status | 1 (3.2) | 9 (6.9) | |
| Other | 0(0.) | 1 (0.7) | |
| Postnatal | .10b | ||
| Diagnosisf | 29 (93.5) | 129 (99.2) | |
| None | 1 (3.2) | 0 | |
| GI | 1 | 0 | |
| Duodenal web Jejunal atresia | 0 | 0 | |
| Duodenal hernia | 0 | 0 | |
| Gastroschisis | 0 | 0 | |
| Omphalocele | 0 | 0 | |
| Heterotaxy | 0 | 0 | |
| Non-GI | 1 (3.2) | 1 (0.8) | |
| Imaging indicated | <.01e | ||
| Yes | 25 (80.6) | 22 (16.9) | |
| Abdominal US | 11 | 11 | |
| X-ray | 21 | 7 | |
| Echo | 3 | 3 | |
| Other | 1 | 2 | |
| No | 6(19.4) | 108(83.1) | |
| Surgery indicated | .04e | ||
| Yes | 2 (6.5) | 0(0) | |
| No | 29 (93.5) | 130 (100) | |
Values are expressed as number (percentage) unless indicated otherwise.
GI, gastrointestinal; LOS, length of stay; NICU, neonatal intensive care unit; PPROM, preterm premature rupture of the membranes; PREEC, preeclampsia; US, ultrasound.
The isolated large stomach group was divided into those who were admitted to the NICU and those who were not, and the postnatal outcomes were analyzed between those 2 subgroups;
Chi-squared analysis;
Student t test analysis;
The preterm cases that were not admitted to the NICU were all late preterm without any postnatal complications;
Fisher exact test analysis;
Only anatomic postnatal diagnosis was considered; most infants admitted to the NICU with no anatomic anomalies presented as having preterm birth, small for gestational age, or respiratory difficulties.
Outside records of previous sonograms were reviewed in all cases of small bowel atresia who did not have their original anatomic US examinations performed at Johns Hopkins to assess their initial presentations (Table 7).
TABLE 7.
Small bowel obstruction case descriptionsa
| Case | Obstruction classification | Gestational age of earliest US from OSH | Gestational age of first enlarged stomach | Gestational age of first double bubble or intestinal dilation | Case type |
|---|---|---|---|---|---|
| 1 | Duodenal web | 18wk0d | 18wk0d | 29wk3d | Isolated |
| 2 | Duodenal atresia | 23wk0d | 31wk0d | 31wk0d | Complex |
| 3 | Duodenal atresia | 16wk1d | 32wk4d | 16wk1d | Complex |
| 4 | Duodenal atresia | 19wk5d | 32wk2d | 32wk2d | Complex |
| 5 | Duodenal atresia | 25wk2d (late to care) | 25wk2d | 25wk2d | Complex |
| 6 | Duodenal atresia | 30wk5d (noted to have previous normal anatomy screening at OSH) | 39wk2d | 30wk5d | Complex |
| 7 | Duodenal atresia | 19wk2d | 31wk5d | 31wk5d | Complex |
| 8 | Duodenal atresia | 16wk0d | 27wk6d | 27wk6d | Complex |
| 9 | Duodenal atresia | 22wk2d | 22wk2d | 22wk2d | Complex |
| 10 | Duodenal atresia | 28wk0d (late to care) | 28wk0d | 28wk0d | Complex |
| 11 | Duodenal atresia | 18wk2d | 28wk3d | 28wk3d | Complex |
| 12 | Duodenal atresia | 27wk3d (late to care) | 28wk3d | 28wk3d | Complex |
| 13 | Duodenal atresia | 19wk0d | 29wk4d | 29wk4d | Complex |
OSH, outside hospital; US, ultrasound.
All cases with postnatal bowel obstruction with transfer of care into the Johns Hopkins Health System from an outside hospital received further follow-up by phone call with the referral center at which they received their initial management for the pregnancy. Records were extracted from the referral centers, and the ultrasound presentations for their prereferral visits were assessed for any anomalies with documentation of ultrasound findings along with gestational ages for both the initial US findings at Johns Hopkins and for the initial US findings at the referring institution.
Statistical analysis
Variables analyzed by means of Pearson’s or chi-squared analysis included maternal race, maternal gravity and parity, previous abortion or miscarriage, maternal infection during pregnancy, substance use disorder during pregnancy, mode of delivery, fetal sex, preterm or term birth, resolution of stomach enlargement before birth, and postnatal infant diagnosis. Fisher exact test was used to analyze for maternal diabetes, maternal body mass index, amniotic fluid level, medical indications for delivery, NICU admission, imaging indications, and postnatal surgical indications. Student t test was used to analyze maternal age, gestational age at delivery, fetal weight, and average length of stay (LOS). Statistical analysis was performed with Stata (StataCorp LLC, College Station, TX). IRB approval was obtained for this retrospective chart review along with a waiver of the Health Insurance Portability and Accountability Act of 1996 privacy authorization.
Results
Of the 57,346 cases with US examinations within the study time frame, 348 had an enlarged fetal stomach. Notably, 107 (31.2%) were excluded owing to delivery at an outside institution (n=97 [90.7%]), termination of pregnancy (n=4 [3.7%]), early pregnancy loss (n=2 [1.9%]), or fetal demise (n=4 [3.7%]). Of the 241 cases (69.3%) that met the inclusion criteria, 161 (66.8%) were classified as an isolated enlarged stomach and 80 (33.2%) were classified as complex (Figure).
FIGURE. Cohort flow chart.
The Johns Hopkins Health System ultrasound database was searched for all cases born between January 12002 and June 2016 flagged with an enlarged stomach on an ultrasound at any given gestational age. Stomach enlargement was confirmed by the authors, and cases were excluded if they did not result in a live birth within the study time frame in the Johns Hopkins Health System, resulting in 241 cases with an enlarged stomach, 80 classified as complex and 161 classified as isolated.
NICU, neonatal intensive care unit.
Demographic information is presented in Table 1. The average maternal age of isolated large stomach cases was greater than in the complex cases. Isolated large stomach cases were more likely to be African American or Asian.
Outcome data are presented in Table 2. Birthweight and gestational age at delivery were greater in the isolated cases. Isolated cases had lower rates of preterm and medical indicated deliveries and NICU admissions and length of hospital stay. Isolated cases were more likely to have normal amniotic fluid levels along with stomach size resolution to within normal limits before birth. They had lower needs for postnatal imaging and surgical procedures.
Of the 161 cases of fetuses with isolated large stomachs, 1 developed a postnatal GI anomaly, that is, a duodenal web (Table 2). In addition, a postnatal diagnosis of any kind was less common in isolated large stomach cases compared with complex.
All complex large stomach cases had additional US findings that were confirmed postnatally. Notably, 56 (70.0%) had an additional GI anomaly (Table 2); 14 (17.5%) had duodenal atresia whereas 4 (5.0%) had jejunal atresia. The additional sonographic bubble sign, dilated bowel, echogenic findings occurring simultaneously with bowel, cardiac defects, abdominal wall or before the first enlarged stomach in defects, diaphragmatic hernia, and hetthe complex cases included a double erotaxy (Table 7).
Of the isolated group, 158 (98.1%) had no anomalies seen after birth (Table 2). Moreover, 2 cases (1.2%) had cardiac findings, that is, a ventricular septal defect and congenital aortic stenosis; 1 isolated large duodenal web.
The isolated cases were stratified by NICU admission and compared; 31 (19.3%) required a NICU admission whereas 130 (80.7%) did not (Table 5, Table 6). There were no considerable differences in maternal demographics (Table 5). The infants who required a NICU stay weighed less than those who did not and were more likely to have had a medical indication for delivery, cesarean delivery, and preterm delivery (Table 6). The preterm infants who were admitted to the NICU were younger and weighed less than the premature infants who were not admitted (Table 6). The number of cases requiring postnatal surgery and imaging was higher in NICU infants.
Isolated cases were also stratified by whether their large stomach persisted or resolved before birth (Table 3). Those who had continued stomach enlargement had higher rates of polyhydramnios and average length of hospital stay than those whose enlarged stomachs resolved. There was no significant difference in postnatal diagnosis (P=.34) or preterm delivery rates (P=.31) between these 2 groups.
Complex cases were stratified into GI- and non-GI–related postnatal diagnoses (Table 4). The GI cases had substantially lower average birthweights, gestational ages at delivery, and rates of resolution of their enlarged stomachs before birth. The GI cases had higher rates of medical indications for delivery, average length of hospital stay, preterm birth rates, NICU admission rates, and surgical needs.
Structured Discussion and Comment
Principle findings
We found that an isolated enlarged fetal stomach occurs at a prevalence of 0.60%. Of the 161 cases with an isolated large stomach, 1 (0.62%) resulted in a GI obstruction, forming a duodenal web. This began with the first observation of an isolated large fetal stomach on US at 18 weeks’ gestation with measurements of 12.4, 8.1, and 21.2 mm in the anterior-posterior, transverse, and longitudinal directions, respectively. At 29 weeks and 3 days of gestation, a double bubble sign and echogenic bowel were also visualized. Notably, 13 of the isolated cases had their first stomach enlargement identified between 16 and 19 weeks’ gestation; 8 of these 13 cases had stomach volumes greater than the case that resulted in a duodenal web at this gestational age, yet did not result in any postnatal diagnosis themselves. Thus, the degree of stomach enlargement does not seem to be a useful parameter in predicting which cases may proceed to develop an intestinal obstruction.
Of the 80 complex cases, 18 (22.5%) had a postnatal bowel obstruction, 4 jejunal and 14 duodenal atresias. All 18 cases began with findings of either dilated bowel or a double bubble sign at the time when their enlarged stomach was noted.
Although our study analyzed all complex large fetal stomachs, the prenatal sonographic diagnoses of diaphragmatic hernia, heterotaxy, and an abdominal wall defect are obvious and are virtually never associated with a previous isolated enlarged stomach. However, small bowel obstructions or malrotation was thought to possibly present initially with isolated fetal stomach enlargement. Thus, the low frequency of formation of a GI obstruction from an isolated large stomach remains our principal finding from this study.
Results
The outcomes that differed markedly between the isolated and complex large stomach groups include birthweight, preterm delivery, polyhydramnios, a medical indication for delivery, NICU admission, length of hospital stay, and need for postnatal imaging or surgical procedures. Outcomes were all more favorable for the isolated group. The frequency at which an isolated large fetal stomach develops into an intestinal obstruction was found to be 0.62%.
Clinical implications
On literature review, only 1 case report was found to depict an isolated large fetal stomach on US before postnatal bowel obstruction.6 In this report, the only finding was an enlarged fetal stomach at 22 weeks after a normal early anatomy scan at 15 weeks’ gestation. A dilated bowel loop was later seen at 32 weeks’ gestation. In this case, the duodenum was stenotic, although the conclusion of the report recommended a consideration of duodenal atresia when abnormal stomach size or position is observed.6 Given an unknown incidence of this phenomenon, we sought to determine more systematically the degree of association between an isolated enlarged stomach and subsequently identified bowel obstruction and found that this association occurs less than 1% of the time.
Only 1 of the small bowel obstruction cases in our study initially presented with an isolated large stomach (Table 7, case number 1). In addition, of the 18 complex small bowel obstruction cases, 3 (17%) presented at <24 weeks’ gestation and 6 (33%) presented at <28 weeks’ gestation, with the finding of a double bubble sign or dilated bowel accompanying an enlarged fetal stomach. Thus, our data indicate that one-third of cases of small bowel obstruction will present with specific signs of GI obstruction earlier than the midthird trimester.
Our study helps address the counseling and follow-up of a finding with a formerly vague prognosis. The limitations in interpretation inherent to certain US findings cause confusion and anxiety for future parents.13 Studies acknowledge that increased routine use of US may lead to more parental anxiety by identifying insignificant findings that would otherwise not have been discovered.13,14 Even if later scans show resolution of the abnormality, parental concern persists, emphasizing the lasting effects of the anxiety engendered by these findings.15 Studies analyzing the effects of false positive soft markers for fetal abnormalities show that mothers who were told of these soft markers had higher rates of third-trimester anxiety and depression, increased problems with emotional attachment and regulation after delivery, and higher instances of infant avoidance and intrusive behaviors.16
Although pregnancies with soft markers and diagnostic anomalies on fetal US are often followed up with additional US examinations, a cost-benefit analysis has not been assessed.15 This includes the direct cost for additional testing and the indirect cost of reduced working ability of mothers-to-be burdened by additional testing. Per the Choosing Wisely campaign, medical practices should have specific evidence to support their use.17 This approach led to the questioning of the significance of other soft markers, including echogenic intracardiac focus and choroid plexus cyst.16,18 Studies looking at the significance of isolated markers led to the Society for Maternal-Fetal Medicine guidelines recommending against following up these markers in women at low risk for aneuploidy, in whom these findings are considered to be a normal variant.19 Our data indicate that an isolated large fetal stomach does not seem to warrant additional follow-up given its probability of resulting in a serious condition like a small bowel obstruction (0.62%) vs a healthy neonate (98.1%).
Research implications
Similar findings were reflected in a retrospective study analyzing 33 cases of enlarged fetal stomachs in the second trimester.20 Of the 19 isolated large stomach cases, no GI anomalies were noted postnatally.20 We saw similar results on a larger scale, determining that an isolated large stomach on US signifies a GI obstruction less than 1% of the time. In addition, our study found 65.8% of isolated large stomachs resolve before birth, whereas the earlier study found that 48% resolved. Although the isolated cases with persistently enlarged fetal stomachs had higher amniotic fluid levels, there was no significant difference in preterm delivery rates (Table 3). Those with persistently enlarged stomachs had a significantly longer hospital stay, which we believe is predominantly caused by the outlier of having 1 case within this group that resulted in a duodenal web with subsequent postnatal surgical needs and a LOS of 46 days. Ultimately, there was no significant difference in postnatal diagnosis between isolated cases whose stomachs resolved before birth and those whose did not. Notably, 15 of the 24 complex cases with a non-GI–related postnatal diagnosis (62.5%) also saw their large stomachs resolve before birth. Thus, a large stomach may be an incidental finding in cases of fetuses both with and without additional congenital abnormalities.
The explanation for our findings may lie in understanding the development of the fetal stomach, which reaches its final shape at 22 weeks’ gestation.21 Studies suggest that gastric growth occurs in distinct phases, with increased and decreased growth rates throughout development occurring in a surprisingly nonlinear fashion.22 The correlation coefficient of the gastric surface area with gestational age (r=0.74) is lower than that of the fetal biparietal diameter with gestational age (r=0.97), emphasizing that fetal gastric growth trajectory does not perfectly align with overall fetal growth throughout development.7 This is also demonstrated by the increased standard deviation in average gastric size as gestational age increases.7,11,23 Gastric area ratio (fetal gastric area divided by the abdominal transverse section) is stable throughout pregnancy regardless of gestational age, and fetuses with duodenal atresia have a gastric area ratio of more than the 95% confidence interval for the predicted value, whereas fetal gastric area was found to be significantly less useful.7 Thus, there may be other forms of stomach measurement that may be better at assessing pathologic enlargement.
Although this study shows that isolated large stomachs have better prognostic outcomes than complex, it does not compare the outcomes of a fetus with an isolated large stomach as the sole abnormality to fetuses without any abnormalities of any kind. It would be beneficial to compare isolated large stomach pregnancies with a cohort of pregnancies with no anomalies and no stomach enlargement to determine whether there were any statistically significant differences in outcomes. This would aid in determining whether there are subsequent consequences of this abnormality outside of the <1% risk of the development of a small bowel obstruction. In addition, the complex group has significant heterogeneity and includes diagnoses that are not known to be associated with an initial isolated enlarged stomach, making prognostic counseling on the complex cases difficult to generalize. We stratified the complex group by comparing those with GI-related postnatal diagnoses with those with non-GI–related postnatal diagnoses. However, given the continued variability in diagnosis within both the GI and non-GI cases, we believed counseling in complex cases is best done on an individual basis that is related specifically to the additional US findings.
Strengths and limitations
The major strength of this study was the large number of cases with ascertainment of a large fetal stomach. However, late access to prenatal care was a limitation. A few women presented for their initial sonogram after signs of the development of intestinal obstruction were evident on US, but who might have presented with an isolated large stomach if scanned earlier. Notably, 3 of 18 complex small bowel atresia cases in our study presented late to care, although one of these cases had a normal anatomy scan at her late initial presentation and developed a double bubble sign a week later with concurrent stomach enlargement (Table 7). The remaining 2 cases presenting late to care had a double bubble sign evident on their first US screening (Table 7).
In addition, US examinations vary in their ability to view every fetal organ owing to uncontrollable factors affecting visibility. Although uncommon for the stomach, one cannot be certain as to whether an enlarged stomach would have been picked up before additional findings of GI obstruction were noted had the most optimal view been available. Finally, there were no cases of pyloric atresia in our cohort. This exceedingly rare phenomenon would be expected to present with an isolated enlarged fetal stomach. Its absence is likely related to its extreme infrequency.
Conclusions
An isolated large fetal stomach has been considered a predecessor to fetal intestinal obstruction. However, with 1 of 161 isolated large stomach cases resulting in an intestinal obstruction, we conclude that the frequency of this outcome is likely <1%. When additional US findings accompanied the initial presentation of an enlarged stomach, a GI obstruction occurred 22.5% of the time. Thus, in the vast majority of cases, an isolated enlarged stomach is an incidental finding for which serial sonographic follow-up may not be indicated.
AJOG MFM at a Glance.
Why was this study conducted?
An enlarged fetal stomach is often regarded as a marker for the evolution of small bowel obstruction and is thus followed up with serial ultrasound (US) examinations. However, the frequency of this phenomenon has not been determined, making prognostic counseling and optimal practice management challenging.
Key findings
Isolated large fetal stomachs reflect intestinal obstructions at a frequency of <1%. This suggests that, when isolated, an enlarged stomach may not warrant serial US examinations.
What does this add to what is known?
This study provides new information concerning the prognosis of isolated fetal stomach enlargement, the counseling for which was previously based on case series and reports.
Acknowledgments
This research was conducted at The Johns Hopkins University School of Medicine, Baltimore, Maryland.
This study was supported by The Johns Hopkins University School of Medicine, which had no involvement in the conduct of research or presentation of the article.
Footnotes
The authors report no conflict of interest.
References
- 1.Lau PE, Cruz S, Cassady CI, et al. Prenatal diagnosis and outcome of fetal gastrointestinal obstruction. J Pediatr Surg 2017;52:722–5. [DOI] [PubMed] [Google Scholar]
- 2.Silva P, Reis F, Alves P, Farinha L, Gomes MS, Câmara P. Fetal bowel dilatation: a sonographic sign of uncertain prognosis. Case Rep Obstet Gynecol 2015;2015:608787. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Lupo PJ, Isenburg JL, Salemi JL, et al. Population-based birth defects data in the United States, 2010e2014: a focus on gastrointestinal defects. Birth Defects Res 2017;109:1504–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Traubici J. The Double Bubble sign. Radiology 2001;220:463–4. [DOI] [PubMed] [Google Scholar]
- 5.Tonni G, Grisolia G, Granese R, et al. Prenatal diagnosis of gastric and small bowel atresia: a case series and review of the literature. J Matern Fetal Neonat Med 2016;29:2753–61. [DOI] [PubMed] [Google Scholar]
- 6.Pariente G, Landau D, Aviram M, Hershkovitz R. Prenatal diagnosis of a rare sonographic appearance of duodenal atresia: report of 2 cases and literature review. J Ultrasound Med 2012;31:1829–33. [DOI] [PubMed] [Google Scholar]
- 7.Sase M, Asada H, Okuda M, Kato H. Fetal gastric size in normal and abnormal pregnancies. Ultrasound Obstet Gynecol 2002;19:467–70. [DOI] [PubMed] [Google Scholar]
- 8.Wilpers AB, Kennedy HP, Wall D, Funk M, Bahtiyar MO. Maternal anxiety related to prenatal diagnoses of fetal anomalies that require surgery. J Obstet Gynecol Neonat Nurs 2017;46:456–64. [DOI] [PubMed] [Google Scholar]
- 9.Kaasen A, Helbig A, Malt UF, Næs T, Skari H, Haugen G. Maternal psychological responses during pregnancy after ultrasonographic detection of structural fetal anomalies: a prospective longitudinal observational study. PLoS One 2017;12:e0174412. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Mulder EJH, Robles de Medina PG, Huizink AC, Van den Bergh BRH, Buitelaar JK, Visser GHA. Prenatal maternal stress: effects on pregnancy and the (unborn) child. Early Hum Dev 2002;70:3–14. [DOI] [PubMed] [Google Scholar]
- 11.Kepkep K, Tuncay YA, Göynümer G, Yetim G. Nomogram of the fetal gastric size development in normal pregnancy. J Perinat Med 2005;33:336–9. [DOI] [PubMed] [Google Scholar]
- 12.Geerts LT, Brand EJ, Theron GB. Routine obstetric ultrasound examinations in South Africa: cost and effect on perinatal outcome–a prospective randomised controlled trial. Br J Obstet Gynaecol 1996;103:501–7. [DOI] [PubMed] [Google Scholar]
- 13.Edvardsson K, Small R, Persson M, Lalos A, Mogren I. ‘Ultrasound is an invaluable third eye, but it can’t see everything’: a qualitative study with obstetricians in Australia. BMC Preg Childbirth 2014;14:363. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Larsson AK, Svalenius EC, Lundqvist A, Dykes AK. Parents’ experiences of an abnormal ultrasound examination—vacillating between emotional confusion and sense of reality. Reprod Health 2010;7:10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Garcia J, Bricker L, Henderson J, et al. Women’s views of pregnancy ultrasound: a systematic review. Birth 2002;29:225–50. [DOI] [PubMed] [Google Scholar]
- 16.Viaux-Savelon S, Dommergues M, Rosenblum O, et al. Prenatal ultrasound screening: false positive soft markers may alter maternal representations and mother-infant interaction. PLoS One 2012;7:e30935. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Corrigan K, Aksenov L, Paul A, Sharif-Askary B, Agarwal S, Kamal A. Characterization of the recommendations in the choosing wisely initiative. Am J Med Qual 2019;34:360–6. [DOI] [PubMed] [Google Scholar]
- 18.Van den Hof MC, Wilson RD ; Diagnostic Imaging Committee, Society of Obstetricians and Gynaecologists of Canada; Genetics Committee, Society of Obstetricians and Gynaecologists of Canada. Fetal soft markers in obstetric ultrasound. J Obstet Gynaecol Can 2005;27:592–636. [DOI] [PubMed] [Google Scholar]
- 19.Norton ME, Biggio JR, Kuller JA, Blackwell SC. The role of ultrasound in women who undergo cell-free DNA screening. Am J Obstet Gynecol 2017;216:B2–7. [DOI] [PubMed] [Google Scholar]
- 20.Richardson J, Boxall S, Stanwell J, Wellesley D, Brackley K. PF.66 Prognostic significance an enlarged fetal stomach in the second trimester. Arch Dis Child Fetal Neonat Ed 2013;98(Suppl1):A22. [Google Scholar]
- 21.Hawass NE, al-Badawi MG, Fatani JA, Meshari AA, Edrees YB. Morphology and growth of the fetal stomach. Invest Radiol 1991;26:998–1004. [DOI] [PubMed] [Google Scholar]
- 22.Nagata S, Koyanagi T, Horimoto N, Satoh S, Nakano H. Chronological development of the fetal stomach assessed using real-time ultrasound. Early Hum Dev 1990;22:15–22. [DOI] [PubMed] [Google Scholar]
- 23.Goldstein I, Reece EA, Yarkoni S, Wan M, Green JL, Hobbins JC. Growth of the fetal stomach in normal pregnancies. Obstet Gynecol 1987;70:641–4. [PubMed] [Google Scholar]