Abstract
Introduction
Fetal growth restriction (FGR) occurs in a significant proportion of pregnancies complicated by gastroschisis. Due to abdominal organ herniation, fetal abdominal circumference often measures less than expected when using Hadlock biometric measurements derived from non‐anomalous fetuses. Therefore, growth restriction may be overdiagnosed in fetuses with gastroschisis. Our objectives were (1) to create second‐ and third‐trimester fetal growth curves (nomograms) specific to gastroschisis and (2) to evaluate antenatal and delivery outcomes for fetuses diagnosed with fetal growth restriction by non‐anomalous Hadlock growth curves compared to gastroschisis‐specific nomograms.
Methods
This was a single‐institution, retrospective cohort study of pregnancies complicated by isolated gastroschisis who delivered between 2011 and 2020. Using ultrasound biometric data, we generated growth curves specific to fetuses with gastroschisis, defining the 10th, 50th, and 90th percentiles between 15 and 37 weeks’ gestation. To utilize the same formula most commonly incorporated into ultrasound software programs, estimated fetal weight in grams was calculated using the Hadlock formula. This generated new anomaly‐specific nomograms and percentiles based on gastroschisis‐specific biometric data. We compared antepartum and delivery outcomes for fetuses diagnosed with FGR by non‐anomalous Hadlock growth curves to those diagnosed with FGR by gastroschisis nomograms using a non‐parametric model and ANOVA chi‐square. Perinatal outcomes were analyzed with t‐test, chi‐square, or Fisher's exact test.
Results
Five biometric nomograms were generated for gastroschisis fetuses, including estimated fetal weight, biparietal diameter, head circumference, abdominal circumference, and femur length. Estimated fetal weight curves for gastroschisis fetuses were downshifted significantly compared to 1984 Hadlock growth curves (p < 0.001). The gastroschisis‐specific 50th percentile curve virtually overlays the non‐anomalous Hadlock 10th percentile curve for estimated fetal weight, with similar findings for abdominal circumference and femur length. Based on the revised growth curve, 60% of gastroschisis fetuses diagnosed with FGR by non‐anomalous Hadlock did not meet criteria for FGR by gastroschisis‐specific nomograms. When comparing fetuses diagnosed with FGR by the revised gastroschisis curve compared to non‐anomalous Hadlock, the new gastroschisis‐specific curve was more strongly associated with 6 out of 19 clinically relevant outcomes, including: betamethasone administration (52% vs. 45%, p < 0.001), abnormal umbilical artery Doppler measurement (56% vs. 42%, p = 0.008), admission or induction prior to spontaneous labor (96% vs. 82%, p < 0.001), worsening bowel findings as the indication for delivery (39% vs. 28%, p = 0.03), increased cesarean delivery (54% vs. 49%, p = 0.001), and decreased vaginal delivery (46% vs. 51%, p = 0.001).
Conclusion
Fetuses with gastroschisis have a growth curve that is distinctly different from that defined by the non‐anomalous Hadlock curve. The 1984 Hadlock curves potentially overdiagnose FGR compared to gastroschisis‐specific curves. FGR defined by gastroschisis‐specific growth curves is strongly associated with many of the common perinatal outcomes seen in pregnancies complicated by this anomaly. Given the known risks of prematurity as well as the financial and labor costs of intensive antenatal surveillance, using gastroschisis‐specific growth curves may assist in improved risk stratification and delivery planning among cases of gastroschisis.
Keywords: fetal growth restriction, gastroschisis, prenatal diagnosis, ultrasound
1. INTRODUCTION
Gastroschisis is an abdominal wall defect causing in utero herniation of bowel and other organs, such as the liver, stomach, or bladder [1]. The defect is most commonly to the right of the umbilical cord and is suspected to occur due to a vascular malformation or accident during abdominal wall formation. It occurs in 2 to 5 out of 10,000 live births with increasing prevalence reported [2, 3]. While diagnosis can be made in the late first trimester after completion of physiologic gut herniation, identification of this anomaly is typically made by ultrasound in the second trimester [1, 4]. Gastroschisis requires postnatal repair and can be associated with prolonged postnatal hospitalizations, loss of non‐viable bowel, as well as long‐term pediatric bowel complications [5, 6]. Risk factors for fetal gastroschisis include younger maternal age, use of tobacco or other vasoactive substances, exposure to agricultural pesticides, disadvantaged socioeconomic status, and poor nutritional status [7, 8, 9, 10, 11].
While gastroschisis is often an isolated anomaly without associated genetic mutations, it is highly associated with fetal growth restriction (FGR) [5]. FGR is diagnosed when the estimated fetal weight (EFW) measures less than the 10th percentile and, more recently, abdominal circumference less than the 10th percentile has been added to FGR diagnostic criteria by the Society of Maternal Fetal Medicine (SMFM) [12]. Importantly, these growth centiles are derived from non‐anomalous fetuses and EFW relies heavily on the abdominal circumference (AC), which is likely underestimated in gastroschisis fetuses due to extra‐abdominal organ herniation. While multiple growth curves exist, Hadlock remains the best predictor of neonatal birthweight among neonates with gastroschisis, even when compared to formulas that do not include AC [13, 14, 15]. There have been several previous smaller cohorts evaluating gastroschisis‐specific growth; however, these previous studies have not consistently used Hadlock as the reference comparator [4, 14, 16]. Evaluating the relationship of fetal growth to perinatal outcomes, there are conflicting data regarding FGR and its association with adverse outcomes in pregnancies complicated by gastroschisis [5, 17, 18].
The aims of this study were (1) to develop gastroschisis‐specific fetal growth curves (nomograms) and (2) to determine the potential utility of these revised anomaly‐specific fetal growth curves by evaluating antepartum and delivery outcomes for fetuses with FGR based on non‐anomalous Hadlock growth curves compared to anomaly‐specific growth curves. We hypothesize that gastroschisis‐specific growth curves demonstrate a significantly smaller EFW than Hadlock growth curves across gestation. Additionally, among fetuses diagnosed with gastroschisis and FGR, we hypothesize that revised anomaly‐specific growth curves will be more strongly associated with adverse antenatal and delivery outcomes compared to FGR diagnosed by non‐anomalous Hadlock curves.
2. METHODS
This single‐center, retrospective cohort study included all patients with isolated fetal gastroschisis who delivered at the University of Washington between January 2011 and July 2020. Exclusion criteria were maternal age < 18 years, fetal anomalies in addition to gastroschisis, multiple gestation, or maternal incarceration. All pregnancies complicated by gastroschisis at our institution's Prenatal Diagnosis Program experience standardized management following our Gastroschisis Care Pathway, which includes ultrasounds every 3–4 weeks from diagnosis through 32 weeks’ gestation assessing fetal growth, amniotic fluid, gastroschisis surveillance (internal and external bowel diameter and wall thickness, stomach cross‐section measurement, the presence of other organ herniation), and superior mesenteric artery (SMA) and vein (SMV) patency. At 32 weeks’ gestation, weekly ultrasound monitoring is initiated as defined above with fetal growth assessed every 3 weeks from 32 weeks’ gestation until delivery. Patients begin twice weekly non‐stress tests (NST) starting at 32 weeks’ gestation. Biophysical profiles are not a routine part of the pathway, however, are added to weekly surveillance for the indications of non‐reactive or suspicious NST. During the time period of this study, umbilical artery Doppler (UAD) was performed for estimated fetal weight < 10th percentile, with UAD frequency determined by the Maternal Fetal Medicine provider. The timing of this study data set was just prior to the addition of AC < 10%ile to the definition of FGR by SMFM; thus, UAD was not performed for the isolated indication of AC < 10%ile in this cohort. The goal gestational age for delivery of gastroschisis fetuses at our institution is 37 weeks.
All ultrasounds performed through our prenatal diagnosis program for the evaluation of fetal gastroschisis with biometric data were included for evaluation. The following ultrasound parameters were abstracted: biparietal diameter (BPD), head circumference (HC), AC, femur length (FL), EFW, amniotic fluid index, and umbilical artery doppler. EFW for all patients was calculated within our radiology software using Hadlock's 1984 formula [19]. The 1991 and 1984 Hadlock formulas are commonly available and frequently used in many obstetrical imaging software packages. For generalizability, the intent of this study was to more accurately define and evaluate the utility of growth centiles for gastroschisis fetuses, not to create a new EFW formula. Maternal demographics were abstracted from the electronic medical record (EMR) including age, race, parity, body mass index, and the presence of underlying maternal health conditions such as hypertension and diabetes. Antenatal and delivery outcomes were also abstracted and included gestational week at which NST were initiated, NST abnormality leading to admission or delivery, administration of betamethasone, gestational age at delivery, mode of delivery, whether the patient was induced or admitted for delivery prior to spontaneous labor, and delivery indication. Abstracted neonatal outcomes included birth weight and birth weight percentile, umbilical artery blood gas pH and base deficit (BD), and 1‐min and 5‐min Apgar scores. Neonates with gastroschisis were universally admitted to the neonatal intensive care unit.
2.1. Gastroschisis fetal growth curves
Gastroschisis‐specific normative growth curves for BPD, HC, AC, FL, and EFW were constructed using generalized additive models (GAMs), which allow for both linear and non‐linear associations with gestational age. The estimated fetal weight was calculated using the Hadlock 1984 EFW formula. The gastroschisis‐specific curves were compared to established non‐anomalous Hadlock reference curves (1984 for EFW, 1991 for biometric measures) using analysis of variance (ANOVA) to test for statistical differences between growth trajectories [19, 20]. Sonographer variability and machine calibration were not incorporated in the analysis. Inconsistent biometric data were confirmed by review of the original ultrasound image. If a single biometric measurement was missing from an individual ultrasound study, the available biometry was utilized in the data set.
2.2. Fetal growth restriction and perinatal outcomes
FGR was defined as EFW < 10th percentile. Fetuses were categorized as FGR by non‐anomalous Hadlock growth curves and FGR by gastroschisis‐specific curves. Maternal and perinatal outcomes were compared across FGR classifications. Categorical variables were compared using chi‐square or Fisher's exact test, and continuous variables using independent t‐tests. Logistic regression was used to evaluate the association between postnatal small for gestational age (SGA) and abnormal UAD findings, with odds ratios (ORs) and 95% confidence intervals (CIs) reported. A p value < 0.05 was considered statistically significant.
Analyses were performed using SPSS version 26.0 (IBM Corp.) and R version 4.0.2 (https://www.R‐project.org/). This study was approved by the University of Washington IRB STUDY #00008387.
3. RESULTS
A total of 116 patients with 413 ultrasounds met inclusion criteria. Maternal demographics are shown in Table 1. Mean maternal age was 24 years, with the majority being nulliparous, Caucasian, non‐obese, and lacking underlying medical conditions. Figure 1 shows the number of ultrasound scans by gestational age, with the majority of scans occurring during the third trimester.
TABLE 1.
Demographics.
| Entire cohort (n = 116) | Hadlock < 10% (n = 67) | Gastroschisis < 10% (n = 28) | |
|---|---|---|---|
| Maternal age (years) | 24 (5) | 23 (5) | 23 (7) |
| Race | |||
| American Indian/Alaska Native | 2 (2%) | 0 | 0 |
| Asian | 1 (1%) | 0 | 0 |
| Native Hawaiian/ Pacific Islander | 1 (1%) | 1 (2%) | 0 |
| Black | 5 (4%) | 2 (3%) | 1 (4%) |
| Hispanic | 20 (17%) | 12 (18%) | 7 (25%) |
| Non‐Hispanic White | 59 (51%) | 37 (57%) | 15 (54%) |
| More than one race | 3 (3%) | 2 (3%) | 0 |
| Unknown/Not reported | 25 (22%) | 12 (18%) | 5 (18%) |
| Parity | |||
| 0 | 65 (56%) | 39 (58%) | 15 (54%) |
| 1 | 32 (28%) | 17 (61%) | 9 (69%) |
| ≥ 2 | 19 (16%) | 11 (16%) | 4 (31%) |
| BMI (kg/m2) | 25.9 (4.8) | 25.1 (4.4) | 24.8 (4.6) |
| Underlying maternal conditions | |||
| Hypertensive disorders | 2 (2) | 0 | 0 |
| Diabetes | 8 (7) | 1 (2%) | 0 |
Note: Data presented as n (%) or mean (SD).
FIGURE 1.
Number of Ultrasound Scans by Gestational Age.
3.1. Gastroschisis fetal growth curves
Figure 2 shows the revised gastroschisis EFW growth curve and the Hadlock 1984 EFW growth curve, delineating the 10th, 50th, and 90th percentiles for each. These curves were statistically distinct (p < 0.001). The revised gastroschisis normative EFW curve is downshifted for all gestational ages compared to the Hadlock non‐anomalous EFW curve, such that the 90th percentile gastroschisis EFW curve overlaps the 50th percentile non‐anomalous Hadlock EFW curve. The gastroschisis EFW 50th percentile curve virtually overlays the non‐anomalous Hadlock 10th percentile EFW curve.
FIGURE 2.
Estimated fetal weight: Hadlock and gastroschisis curves.
Individual biometric parameters (BPD, HC, AC, and FL) were also compared to non‐anomalous Hadlock 50th percentile data, although the 10th and 90th percentiles were not available in the Hadlock 1991 publication for comparison (Figures 3, 4, 5, 6). Similar to the EFW curves, the gastroschisis curves for both AC and FL are downshifted, such that the 90th percentile AC and FL curves overlap the non‐anomalous Hadlock 50th percentile curves. BPD and HC were conserved with similar 50th percentile curves for both non‐anomalous Hadlock and revised gastroschisis curves. Tables 2 and 3 show the 10th, 50th, and 90th percentiles by gestational age for EFW and biometric parameters.
FIGURE 3.
Biparietal diameter: Hadlock and gastroschisis curves.
FIGURE 4.
Head circumference: Hadlock and gastroschisis curves.
FIGURE 5.
Abdominal circumference: Hadlock and gastroschisis curves.
FIGURE 6.
Femur length: Hadlock and gastroschisis curves.
TABLE 2.
Gastroschisis estimated fetal weight percentiles by gestational age.
| Gastroschisis EFW (g) | |||
|---|---|---|---|
| Weeks | 10%ile | 50%ile | 90%ile |
| 15 | 103 | 118 | 132 |
| 16 | 145 | 145 | 145 |
| 17 | 160 | 186 | 212 |
| 18 | 186 | 213 | 267 |
| 19 | 216 | 260 | 324 |
| 20 | 274 | 326 | 362 |
| 21 | 338 | 373 | 428 |
| 22 | 356 | 445 | 501 |
| 23 | 420 | 515 | 609 |
| 24 | 490 | 570 | 667 |
| 25 | 588 | 709 | 888 |
| 26 | 596 | 729 | 907 |
| 27 | 720 | 899 | 1135 |
| 28 | 781 | 1025 | 1245 |
| 29 | 958 | 1040 | 1379 |
| 30 | 1065 | 1349 | 1645 |
| 31 | 1177 | 1411 | 1846 |
| 32 | 1379 | 1603 | 1872 |
| 33 | 1496 | 1817 | 2286 |
| 34 | 1618 | 1956 | 2673 |
| 35 | 1789 | 2110 | 2580 |
| 36 | 1950 | 2514 | 2716 |
| 37 | 2537 | 2683 | 3273 |
TABLE 3.
Gastroschisis biometry percentiles by gestational age.
| BPD (mm) | HC (mm) | AC (mm) | FL (mm) | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Weeks | 10%ile | 50%ile | 90%ile | 10%ile | 50%ile | 90%ile | 10%ile | 50%ile | 90%ile | 10%ile | 50%ile | 90%ile |
| 15 | 27 | 30 | 33 | 103 | 111 | 119 | 92 | 94 | 96 | 13 | 16 | 19 |
| 16 | 33 | 34 | 36 | 126 | 129 | 137 | 97 | 104 | 114 | 17 | 19 | 24 |
| 17 | 37 | 38 | 40 | 137 | 143 | 149 | 107 | 113 | 119 | 21 | 24 | 26 |
| 18 | 37 | 39 | 43 | 138 | 148 | 159 | 113 | 124 | 135 | 25 | 26 | 27 |
| 19 | 40 | 45 | 47 | 149 | 162 | 179 | 119 | 133 | 145 | 27 | 29 | 32 |
| 20 | 46 | 47 | 49 | 169 | 176 | 181 | 138 | 147 | 153 | 29 | 32 | 35 |
| 21 | 46 | 50 | 52 | 174 | 184 | 190 | 145 | 153 | 168 | 31 | 34 | 36 |
| 22 | 48 | 52 | 56 | 183 | 196 | 205 | 150 | 161 | 173 | 34 | 37 | 39 |
| 23 | 50 | 54 | 59 | 188 | 206 | 216 | 158 | 174 | 192 | 36 | 39 | 42 |
| 24 | 53 | 57 | 61 | 206 | 218 | 227 | 158 | 181 | 192 | 38 | 41 | 46 |
| 25 | 57 | 61 | 66 | 219 | 231 | 245 | 176 | 197 | 214 | 41 | 44 | 49 |
| 26 | 58 | 63 | 67 | 221 | 238 | 249 | 182 | 193 | 225 | 41 | 44 | 48 |
| 27 | 60 | 66 | 71 | 240 | 247 | 265 | 193 | 213 | 239 | 44 | 48 | 53 |
| 28 | 63 | 69 | 73 | 245 | 259 | 278 | 199 | 221 | 242 | 47 | 51 | 55 |
| 29 | 67 | 72 | 76 | 254 | 268 | 288 | 207 | 225 | 254 | 48 | 52 | 56 |
| 30 | 70 | 75 | 80 | 265 | 281 | 294 | 217 | 7248 | 266 | 50 | 55 | 59 |
| 31 | 71 | 77 | 82 | 272 | 288 | 301 | 227 | 250 | 280 | 54 | 56 | 61 |
| 32 | 75 | 80 | 83 | 277 | 296 | 309 | 241 | 260 | 283 | 54 | 59 | 63 |
| 33 | 76 | 81 | 88 | 286 | 300 | 320 | 250 | 276 | 301 | 57 | 60 | 64 |
| 34 | 78 | 83 | 88 | 292 | 306 | 329 | 260 | 285 | 326 | 58 | 64 | 67 |
| 35 | 81 | 86 | 91 | 297 | 317 | 329 | 268 | 284 | 322 | 60 | 65 | 68 |
| 36 | 78 | 87 | 92 | 306 | 317 | 330 | 277 | 307 | 328 | 63 | 66 | 70 |
| 37 | 88 | 88 | 89 | 322 | 324 | 334 | 307 | 310 | 339 | 67 | 72 | 74 |
Of the 116 fetuses, 67 (58%) met criteria for FGR on at least one ultrasound based on non‐anomalous 1984 Hadlock, while 28 (24%) met FGR criteria based on the revised gastroschisis growth curve on at least one ultrasound. A total of 27 of the 116 patients (23%) met criteria for FGR based on both non‐anomalous Hadlock and revised gastroschisis growth curves. A total of 40 patients (34%) met FGR criteria by non‐anomalous Hadlock but did not meet criteria for FGR using the revised growth curves. Thus, of those diagnosed with FGR by non‐anomalous Hadlock, 60% (40 out of 67) did not meet FGR criteria when using gastroschisis‐specific nomograms.
3.2. Fetal growth restriction and perinatal outcomes
Table 4 shows antepartum and delivery outcomes for fetuses with FGR defined by non‐anomalous Hadlock compared to FGR defined by revised gastroschisis curves. The gastroschisis curve was more strongly associated with 6 out of 19 observed outcomes including betamethasone administration (52% vs. 45%, p < 0.001), abnormal UAD measurement (56% vs. 42%, p = 0.008), admission or induction prior to spontaneous labor (96% vs. 82%, p < 0.001), worsening bowel findings as the indication for delivery (39% vs. 28%, p = 0.03), increased cesarean delivery (54% vs. 49%, p = 0.001), and decreased vaginal delivery (46% vs. 51%, p = 0.001). Defining FGR based on the non‐anomalous Hadlock curve demonstrated improved association with only one antenatal outcome, which was a higher likelihood that an NST abnormality directly led to delivery (43% vs. 41%, p = 0.04). Median birth weight for FGR fetuses defined by the gastroschisis curve confirmed a postnatal diagnosis of SGA more frequently (median 6%, interquartile range 3%–51%) compared to FGR fetuses defined by non‐anomalous Hadlock (median 13%, interquartile range 4%–51%), although this did not reach statistical significance (p = 0.06).
TABLE 4.
Antepartum and immediate delivery outcomes.
| Hadlock < 10% (n = 67) | Gastroschisis < 10% (n = 28) | p value | |
|---|---|---|---|
| Week of NST start | 30.8 (2.5) | 30.1 (2.8) | 0.23 |
| NST abnormality that led to delivery | 28 (43%) | 11 (41%) | 0.04 |
| Received betamethasone | 29 (45%) | 14 (52%) | 0.001 |
| Abnormal umbilical artery Doppler | 21 (42%) | 13 (56%) | 0.008 |
| Gestational age at delivery (weeks) | 36.3 (1.4) | 36.2 (1.2) | 0.74 |
| Mode of delivery | |||
| Vaginal | 34 (51%) | 13 (46%) | 0.001 |
| Cesarean | 33 (49%) | 15 (54%) | 0.001 |
| Induced or admitted for delivery prior to spontaneous labor | 55 (82%) | 27 (96%) | <0.001 |
| Indication for delivery | |||
| Achieved goal gestational age | 12 (22%) | 5 (19%) | 0.06 |
| Abnormal fetal monitoring | 18 (33%) | 7 (27%) | 0.06 |
| Worsening bowel findings on ultrasound | 15 (28%) | 10 (39%) | 0.03 |
| Worsening FGR or abnormal Doppler | 3 (6%) | 1 (4%) | 0.99 |
| Oligohydramnios | 6 (11%) | 3 (12%) | 0.46 |
| Umbilical artery blood gas | |||
| pH | 7.30 (0.05) | 7.29 (0.06) | 0.41 |
| Base deficit | 1.93 (2.18) | 2.23 (2.60) | 0.57 |
| 5‐min Apgar score | 8.0 (1.4) | 8.0 (1.5) | 1.0 |
| 5‐min Apgar score < 7 | 9 (13%) | 2 (7%) | 0.19 |
| Birth weight (g) | 2299 (406) | 2190 (470) | 0.26 |
| Birth weight percentile (%) | 13 [4, 51] | 6 [3, 51] | 0.06 |
Note: Data presented as mean n (%), mean (SD), or median [interquartile range].
A total of 34 out of the 116 fetuses (29%) met criteria for SGA by postnatal birth weight < 10th percentile of anticipated newborn birth weight. This was comparable to the overall antenatal anomaly‐specific FGR rate of 24%, defined as at least one ultrasound during the second and third trimester with EFW < 10%ile by the revised gastroschisis curve. Umbilical artery Dopplers were measured in 57 of the 116 fetuses (49%). Of these, 21 had abnormal UAD (37%). In logistic regression analysis, neonates with a postnatal diagnosis of SGA had fourfold higher odds of having at least one ultrasound with abnormal UAD (OR 4.0, 95% CI 1.28, 12.53). There were no cases of intrauterine fetal demise (IUFD) within our cohort.
4. DISCUSSION
Fetuses with gastroschisis have a growth curve that is distinctly different from that defined by non‐anomalous Hadlock curves, demonstrating a 50th percentile anomaly‐specific EFW growth curve that is less than non‐anomalous Hadlock 50th percentile for all gestational ages. This potentially leads to overdiagnosis of FGR for gastroschisis fetuses when compared to normative growth of non‐anomalous fetuses. Additionally, the gastroschisis‐specific growth curve was more strongly associated with adverse antenatal and delivery outcomes among growth‐restricted fetuses compared to fetuses diagnosed with FGR using non‐anomalous Hadlock growth curves.
Previous studies have proposed revised gastroschisis growth curves; however, none have compared these curves to non‐anomalous Hadlock growth curves, considered the standard of care in the United States [21]. Additionally, data are lacking evaluating perinatal outcomes for gastroschisis fetuses diagnosed with fetal growth restriction. Our study adds new contributions to the literature in two ways: (1) creation of a gastroschisis‐specific normative growth curve using the Hadlock EFW formula, which is readily available in most ultrasound software packages; (2) demonstration of potential clinical utility of revised gastroschisis‐specific growth curves.
The down‐shifted curves in this study were most evident for EFW, AC, and FL, while head measurements of BPD and HC appeared to be preserved. There are three prior studies that have evaluated in utero growth patterns specific to gastroschisis fetuses with a similar intent to create unique growth curves. Previous findings by Barbieri et al. and Netta et al., who also created growth curves based on retrospective sonographic data, support a pattern of fetal growth deficit among fetuses with gastroschisis as compared to non‐anomalous fetuses; however, both studies found symmetric growth deficits rather than the asymmetric findings of this study [14, 16]. Both studies included fewer patients, 72 and 40, respectively, than the current study, and neither used Hadlock as the comparison growth curve. Horton et al. generated gastroschisis‐specific growth curves through ultrasound review and compared to standard curves throughout gestation [4]. This study also described delivery outcomes among the entire cohort of fetuses with gastroschisis (n = 71). None of the previous three studies compared perinatal outcomes between standard and revised groups to determine the utility of the revised EFW growth curves. In a slightly different approach, McKenna et al. in 2023 sought to identify which ultrasound EFW formula for gastroschisis fetuses was most accurate at predicting postnatal SGA neonates [15]. They calculated EFW in 170 gastroschisis fetuses using eight different formulas: Hadlock I–IV, Honarvar, Shepard, Siemer, and Warsof. They found that Hadlock II–IV had the highest accuracy and sensitivity in predicting postnatal SGA neonates.
In addition to the creation of growth curves, there has been increased attention in the literature to interval growth of fetuses with gastroschisis [22, 23]. Zhang‐Rutledge et al. (n = 95) performed a study on interval growth throughout gestation and, similar to the current study, found that in utero growth of fetuses with gastroschisis fell on the Hadlock 10% percentile curve [23]. Sherwin et al. evaluated growth patterns with an individualized growth assessment (IGA) to determine if there was a specific pattern that could distinguish pathological growth restriction among gastroschisis fetuses. This revealed two patterns of growth restriction: persistent and in utero recovery [22]. Asymmetric fetal growth with decreased abdominal circumference in non‐anomalous fetuses is typically attributed to placental insufficiency and increased risk of perinatal morbidity and mortality due to poor placental perfusion. However, the etiology of a smaller AC for gastroschisis is largely anatomical, with the majority of bowel external to the abdomen through the abdominal wall defect. Thus, further prospective evaluation of gastroschisis‐specific growth patterns is essential to better define asymmetric growth that is within normal limits for this anomaly versus in utero growth patterns that suggest a placental component to abnormal fetal growth.
Our data support many of the antenatal and delivery outcomes that have been reported for pregnancies complicated by gastroschisis. There are several studies that describe mean delivery gestational age of approximately 36 weeks, which is confirmed in our study population [24, 25]. A meta‐analysis of 38 studies (65,077 participants) evaluating mode of delivery revealed a vaginal delivery rate of 46% and a cesarean delivery rate of 54%, also consistent with our findings [26]. Birth weight for both simple and complex gastroschisis has been reported as approximately 2400 g, which is slightly larger than neonates in our cohort [24, 25, 27]. Horton et al. described indications for delivery with their cohort, which in general had similar rates of labor induction, although much higher rates of induction for FGR (23% vs. 4% using their revised curve) and oligohydramnios (20% vs. 12% using their revised curve) [4].
The revised anomaly‐specific growth curve was more strongly associated with adverse antenatal and delivery outcomes among fetuses with EFW < 10th percentile when compared to those diagnosed with FGR by non‐anomalous Hadlock. Conversely, FGR defined by non‐anomalous Hadlock demonstrated improved association with only one outcome: a marginal increase in NST abnormalities directly leading to delivery (41% to 43%). The revised curve also trended toward better identification of neonates that went on to be diagnosed with SGA based on postnatal birth weight percentiles. Although the association with SGA was not statistically significant, our limited sample size may preclude us from detecting a statistically significant difference. Future studies to prospectively validate the gastroschisis‐specific nomogram should include evaluation of timing of NST initiation and its impact on neonatal outcomes. Providers managing gastroschisis routinely monitor these pregnancies closely; thus, it is unclear whether a new nomogram in clinical practice would significantly change the utilization of antenatal surveillance.
Increased antenatal surveillance can be costly both in regard to insurance charges, clinic staffing, and labor, as well as time for the patient. Our study demonstrated that 60% of fetuses diagnosed with FGR by Hadlock would not be classified as growth‐restricted fetuses using the revised anomaly‐specific growth curve. Using anomaly‐specific criteria for FGR may provide opportunities for improved cost‐effective patient care due to decreased need for antenatal surveillance and Doppler ultrasounds specifically for a diagnosis of FGR. Reducing unnecessary interventions is crucial for appropriately allocating resources, with anomaly‐specific growth curves providing potential benefits in this area. Prospective cost‐benefit studies would be needed to evaluate this potential benefit. Decreased resource utilization needs to be balanced against the potential for missed fetal NST or Doppler abnormalities due to fewer FGR diagnoses.
Significantly, there were no cases of fetal demise in our cohort while obtaining a mean delivery gestational age of 36 weeks. Our 0% fetal demise rate is lower than that described in the literature [26, 28]. This may be due to our rigorous Gastroschisis Care Pathway, and further inquiry into outcomes prior to and after adoption of this pathway is a potential future direction. Further prospective studies are also needed to evaluate whether gastroschisis‐specific growth curves improve the prediction of longer‐term postnatal outcomes such as length of neonatal hospital stay, type of surgery, time to full enteral feeds, and long‐term bowel dysfunction.
The strengths of this study include that all patients were part of a single prenatal diagnosis program and enrolled in a standardized antenatal Gastroschisis Care Pathway. The majority of patients had multiple ultrasounds to evaluate longitudinal data. Detailed review of medical records with complete antenatal and delivery outcomes were available for all patients. We chose to calculate EFW using the Hadlock 1984 formula, with creation of new normative gastroschisis‐specific percentiles given this formula is easily accessible in most ultrasound software packages in the United States. For generalizability, the intent of this study was to more accurately define and evaluate the utility of growth centiles for gastroschisis fetuses, not to create a new EFW formula.
Limitations include the retrospective nature of the study as well as the unknown relationship to long‐term neonatal and post‐surgical outcomes. Abnormal antenatal testing (abnormal UAD or non‐reassuring NSTs) is an indication for interventions independent of the type of growth curve used, which may be a potential confounder for our adverse antenatal and delivery outcomes. Additionally, our Gastroschisis Care Pathway may differ from other practice patterns and may therefore limit generalizability. We defined FGR as an EFW < 10th percentile as our study was performed prior to the SMFM recommendation to also include AC < 10th percentile in the diagnosis of FGR [12]. We elected not to apply the updated SMFM FGR definition retrospectively as clinical management was based on existing definitions at the time. Prospective assessment of this gastroschisis‐specific growth curve with external validation outside of our institution's care pathway is necessary to better evaluate and confirm potential clinical utility, cost‐effectiveness, and to address potential overfitting due to the single‐center study design.
In summary, growth curves for fetuses with gastroschisis are significantly different than those for non‐anomalous fetuses described by Hadlock. Nomograms for gastroschisis fetuses demonstrate that fetal growth restriction may be overdiagnosed when using non‐anomalous Hadlock curves. Fetal growth restriction defined using a gastroschisis‐specific nomogram was more strongly associated with adverse outcomes than fetal growth restriction defined by Hadlock in our cohort. This provides opportunities for potential improvements in anomaly‐specific antenatal care, exploration of more cost‐effective antenatal surveillance, and future research opportunities to externally validate these findings.
CONFLICT OF INTEREST STATEMENT
The authors declare no conflicts of interest.
ETHICS STATEMENT
IRB approval was obtained for this study. Individual patient consent n/a per IRB exempt policy.
DATA AVAILABILITY STATEMENT
Upon request to the authors.
REFERENCES
- 1. Oakes, M. C. , Porto M., and Chung J. H.. 2018. “Advances in Prenatal and Perinatal Diagnosis and Management of Gastroschisis.” Seminars in Pediatric Surgery 27: 289–99. [DOI] [PubMed] [Google Scholar]
- 2. D'Antonio, F. , Virgone C., Rizzo G., Khalil A., Baud D., Cohen‐Overbeek T. E., Kuleva M., et al. 2015. “Prenatal Risk Factors and Outcomes in Gastroschisis: A Meta‐Analysis.” Pediatrics 136: e159–69. [DOI] [PubMed] [Google Scholar]
- 3. Jones, A. M. , Isenburg J., Salemi J. L., Arnold K. E., Mai C. T., Aggarwal D., Arias W., et al. 2016. “Increasing Prevalence of Gastroschisis–14 States, 1995–2012.” MMWR Morbidity and Mortality Weekly Report 65: 23–6. [DOI] [PubMed] [Google Scholar]
- 4. Horton, A. L. , Powell M. S., and Wolfe H. M.. 2010. “Intrauterine Growth Patterns in Fetal Gastroschisis.” American Journal of Perinatology 27: 211–7. [DOI] [PubMed] [Google Scholar]
- 5. Nicholas, S. S. , Stamilio D. M., Dicke J. M., Gray D. L., Macones G. A., and Odibo A. O.. 2009. “Predicting Adverse Neonatal Outcomes in Fetuses with Abdominal Wall Defects Using Prenatal Risk Factors.” American Journal of Obstetrics and Gynecology 201: 383.e1–6. [DOI] [PubMed] [Google Scholar]
- 6. Andrade, W. S. , Brizot M. L., Rodrigues A. S., Tannuri A. C., Krebs V. L., Nishie E. N., Francisco R. P. V., and Zugaib M., 2018. “Sonographic Markers in the Prediction of Fetal Complex Gastroschisis.” Fetal Diagnosis and Therapy 43: 45–52. [DOI] [PubMed] [Google Scholar]
- 7. Loane, M. , Dolk H., and Bradbury I.. 2007. “Increasing Prevalence of Gastroschisis in Europe 1980–2002: A Phenomenon Restricted to Younger Mothers?.” Paediatric and Perinatal Epidemiology 21: 363–9. [DOI] [PubMed] [Google Scholar]
- 8. Mastroiacovo, P. 2008. “Risk Factors for Gastroschisis.” BMJ 336: 1386–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Draper, E. S. , Rankin J., Tonks A. M., Abrams K. R., Field D. J., Clarke M., and Kurinczuk J. J., 2008. “Recreational Drug Use: A Major Risk Factor for Gastroschisis?.” American Journal of Epidemiology 167: 485–91. [DOI] [PubMed] [Google Scholar]
- 10. Lam, P. K. , and Torfs C. P.. 2006. “Interaction between Maternal Smoking and Malnutrition in Infant Risk of Gastroschisis.” Birth Defects Research. Part A, Clinical and Molecular Teratology 76: 182–6. [DOI] [PubMed] [Google Scholar]
- 11. Waller, S. A. , Paul K., Peterson S. E., and Hitti J. E.. 2010. “Agricultural‐Related Chemical Exposures, Season of Conception, and Risk of Gastroschisis in Washington State.” American Journal of Obstetrics and Gynecology 202: 241.e1–6. [DOI] [PubMed] [Google Scholar]
- 12. Society for Maternal‐Fetal Medicine (SMFM) , Martins, J. G. , Biggio J. R., and Abuhamad A.. 2020. “Society for Maternal‐Fetal Medicine Consult Series #52: Diagnosis and Management of Fetal Growth Restriction: (Replaces Clinical Guideline Number 3, April 2012).” American Journal of Obstetrics and Gynecology 223: B217. [DOI] [PubMed] [Google Scholar]
- 13. Zaki, M. N. , Lusk L. A., Overcash R. T., Rao R., Truong Y. N., Liebowitz M., and Porto M., 2018. “Predicting Birth Weight in Fetuses with Gastroschisis.” Journal of Perinatology 38: 122–6. [DOI] [PubMed] [Google Scholar]
- 14. Barbieri, M. M. , Bennini J. R., Nomura M. L., Morais S. S., and Surita F. G.. 2017. “Fetal Growth Standards in Gastroschisis: Reference Values for Ultrasound Measurements.” Prenatal Diagnosis 37: 1327–34. [DOI] [PubMed] [Google Scholar]
- 15. McKenna, M. , McKenna D., Zhou M., Sonek J., and Wiegand S.. 2023. “Prediction of Neonatal Growth Restriction in Fetuses with Gastroschisis by Early Third Trimester Ultrasonography Utilizing Contemporary Birth Weight Percentiles.” Journal of Ultrasound in Medicine 42(5): 997–1005. 10.1002/jum.16108. [DOI] [PubMed] [Google Scholar]
- 16. Netta, D. A. , Wilson R. D., Visintainer P., Johnson M. P., Hedrick H. L., Flake A. W., and Adzick N. S., 2007. “Gastroschisis: Growth Patterns and a Proposed Prenatal Surveillance Protocol.” Fetal Diagnosis and Therapy 22: 352–7. [DOI] [PubMed] [Google Scholar]
- 17. Overcash, R. T. , DeUgarte D. A., Stephenson M. L., Gutkin R. M., Norton M. E., Parmar S., Porto M., Poulain F. R., and Schrimmer D. B., 2014. “Factors Associated with Gastroschisis Outcomes.” Obstetrics and Gynecology 124: 551–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Girsen, A. I. , Do S., Davis A. S., Hintz S. R., Desai A. K., Mansour T., Merritt T. A., Oshiro B. T., El‐Sayed Y. Y., and Blumenfeld Y. J., 2015. “Peripartum and Neonatal Outcomes of Small‐For‐Gestational‐Age Infants with Gastroschisis.” Prenatal Diagnosis 35: 477–82. [DOI] [PubMed] [Google Scholar]
- 19. Hadlock, F. P. , Harrist R. B., Carpenter R. J., Deter R. L., and Park S. K.. 1984. “Sonographic Estimation of Fetal Weight. The Value of Femur Length in Addition to Head and Abdomen Measurements.” Radiology 150: 535–40. [DOI] [PubMed] [Google Scholar]
- 20. Hadlock, F. P. , Harrist R. B., and Martinez‐Poyer J.. 1991. “In Utero Analysis of Fetal Growth: A Sonographic Weight Standard.” Radiology 181: 129–33. [DOI] [PubMed] [Google Scholar]
- 21. Blue, N. R. , Savabi M., Beddow M. E., Katukuri V. R., Fritts C. M., Izquierdo L. A., and Chao C. R., 2019. “The Hadlock Method is Superior to Newer Methods for the Prediction of the Birth Weight Percentile.” Journal of Ultrasound in Medicine 38: 587–96. [DOI] [PubMed] [Google Scholar]
- 22. Sherwin, E. , Deter R., Joudi N., Trepman P., Lee W., El‐Sayed Y. Y., Girsen A. I., Datoc I., Hintz S. R., and Blumenfeld Y. J.. 2021. “Individualized Growth Assessment in Pregnancies Complicated by Fetal Gastroschisis.” Journal of Maternal‐Fetal & Neonatal Medicine 35(90): 1–11. [DOI] [PubMed] [Google Scholar]
- 23. Zhang‐Rutledge, K. , Jacobs M., Patberg E., Field N., Holliman K., Strobel K. M., Murphy A., et al. 2021. “Interval Growth Across Gestation in Pregnancies with Fetal Gastroschisis.” American Journal of Obstetrics & Gynecology MFM 3: 100415. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Melov, S. J. , Tsang I., Cohen R., Badawi N., Walker K., Soundappan S. S. V., and Alahakoon T. I., 2018. “Complexity of Gastroschisis Predicts Outcome: Epidemiology and Experience in an Australian Tertiary Centre.” BMC Pregnancy Childbirth 18: 222. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Lausman, A. Y. , Langer J. C., Tai M., Seaward P. G. R., Windrim R. C., Kelly E. N., and Ryan G., 2007. “Gastroschisis: What Is the Average Gestational Age of Spontaneous Delivery?.” Journal of Pediatric Surgery 42: 1816–21. [DOI] [PubMed] [Google Scholar]
- 26. South, A. P. , Stutey K. M., and Meinzen‐Derr J.. 2013. “Metaanalysis of the Prevalence of Intrauterine Fetal Death in Gastroschisis.” American Journal of Obstetrics and Gynecology 209: 114.e1–13. [DOI] [PubMed] [Google Scholar]
- 27. Erdoğan, D. , Azılı M. N., Cavuşoğlu Y. H., Tuncer I. S., Karaman I., Karaman A., and Ozgüner I. F.. 2012. “11‐Year Experience with Gastroschisis: Factors Affecting Mortality and Morbidity.” Iranian Journal of Pediatrics 22(3): 339–43. [PMC free article] [PubMed] [Google Scholar]
- 28. Meyer, M. R. , Shaffer B. L., Doss A. E., Cahill A. G., Snowden J. M., and Caughey A. B.. 2015. “Prospective Risk of Fetal Death with Gastroschisis.” Journal of Maternal‐Fetal & Neonatal Medicine 28: 2126–9. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Upon request to the authors.