Pediatric patients with motility disorders experience a significant disease burden and lower quality of life,^1-4 frequently undergo bothersome, costly testing. As an example, on average, pediatric patients with rumination syndrome undergo 8.8 diagnostic tests costing $19 795 United States dollars per patient.⁵ Antroduodenal manometry (ADM) and gastric emptying scintigraphy (GES) are frequently performed in pediatric patients with impaired upper gastrointestinal (GI) function, which contributes to the discomfort and expense of multiple medical procedures.^6-13 It is currently unclear whether ADM and GES provide valuable distinctive information or if one would sufficiently assess upper motility disorders in pediatric patients with many GI symptoms.

Gastric emptying depends on 2 types of motor activity: intermittent muscle contractions and peristalsis. Isolated, intermittent muscle contractions occur in the antrum to mix and break down gastric contents. Peristalsis propels the gastric contents from the antrum through the pylorus into the proximal small intestine with coordinated and propagated pressure waves.¹⁴ ADM and GES can measure the neuromuscular activity of the antrum and duodenum and the rate of gastric transit, respectively.

Current indications for ADM in pediatrics include diagnosing and classifying types of intestinal pseudo-obstruction; evaluating patients with severe nausea, retching, and feeding intolerance; distinguishing between rumination and other vomiting types; and determining gastric and small bowel response to medications.⁶

Current indications for GES include evaluating for delayed gastric emptying (DGE) or gastroparesis in patients with nausea, vomiting, retching, post-prandial fullness, decreased appetite, and early satiety.¹⁵ Standard adult normal values for gastric emptying (< 10% gastric retention of the meal at 4 hours from ingestion) have been established using a standard meal of 2 slices of white toast, 30 g of jelly, 120 mL of water, and 120 g of liquid egg whites cooked with technetium-99m sulfur colloid quantified after 4 hours.^7,16 These adult standard values have also been used to identify DGE in children ingesting a non-standard cheese-based alternative meal with a radioactive isotope or by ingesting at least 50% of a standard meal.¹⁷

Although ADM and GES are often used in the same patients to evaluate GI symptoms, these tests have not been directly compared to each other in children. In a study of healthy adults, ADM identified that 64.3% (9/14) of patients emptied a non-digestible wireless motility capsule from the stomach with a phase III pattern of the migrating motor complex (MMC) and 35.7% (5/14) with isolated distal antral contractions.¹⁸ When comparing GES and ADM to wireless motility capsules in 22 pediatric patients, wireless motility capsules detected 70% more motor abnormalities than ADM and the wireless motility capsules had poorer specificity in detecting gastroparesis than GES.¹⁹ An adult study from 2008 directly compared ADM and GES: compared to healthy volunteers, abnormal ADM pressure waves were associated with slow gastric emptying in 15 adult intensive care unit patients.²⁰ This study is one of very few that evaluated whether DGE in GES correlates with abnormalities in ADM.

Our study seeks to directly compare ADM and GES in children with upper GI symptoms to evaluate the correlation of values from the testing and if distinct conclusions can be drawn from each of these testing modalities.

Patient Population

We performed a retrospective chart review of pediatric patients who underwent ADM and GES from 2011 to 2020 at the GI Motility Center at Nationwide Children’s Hospital in Columbus, Ohio. GES studies were included if patients ingested ≥ 50% of a standard solid meal without vomiting.¹⁷ In order for the ADM studies to be included in the study, all the following had to be fulfilled: manometry sensors were placed in both antrum and small intestine, some antral activity was detected during the fasting phase, a phase III of the MMC was identified during fasting, and sufficient calories were ingested to induce a fed pattern (400 kcal or >10 kcal/kg).²¹ Calories were calculated using the CalorieKing app (CalorieKing Wellness Solutions, La Mesa, CA, USA). We first identified the group of patients with DGE defined as > 10% gastric retention at 4 hours, who satisfied the inclusion criteria. Patients with normal gastric emptying (NGE) studies meeting inclusion criteria were then retrospectively reviewed to match with the DGE cohort by age and date of testing completed. Demographic information, including sex, weight, body mass index (BMI), diagnosis, symptoms, symptom duration, and medications were collected. Our Institutional Review Board approved the study protocol (Protocol No. 00001709).

Gastric Emptying Scintigraphy Analysis

Patients ingested a standard meal labeled with technetium-99m sulfur colloid. The amount of meal ingested was recorded for each GES exam. Nuclear medicine studies were conducted until 4 hours post-ingestion or until gastric emptying was measured at > 90% if before 4 hours. DGE was defined as > 10% of the meal was retained in the stomach at the 4-hour measurement.

Antroduodenal Manometry Analysis

The ADM study protocols were done typically with a 2 hour fasting period, possible use of pro-kinetic medicine for at least 1 hour, followed by ingestion of a meal ordered by the child, and followed by a post-prandially period of about 2 hours. Manometric studies were analyzed using Medical Measurement Systems software (Enschede, Netherlands). Two investigators, reviewed the ADM tracings. Manometric data were analyzed during 2 time periods, fasting and post-prandial. During fasting, investigators recorded the direction of the phase III pattern of the MMC (antegrade or retrograde), if an antral component was present in any phase III, fasting antral motility index (MI = Ln [sum of amplitudes x number of contractions +1]),²² and number of fasting antral contractions. The fasting antral MI was calculated 1 hour before the meal or before any pro-kinetic medication was given. Post-prandial findings were also recorded, including post-prandial antral MI for 1 hour after the meal, the number of post-prandial antral contractions, and the MI ratio (fasting MI/post-prandial MI). These measurements and overall interpreting physician’s impression of the ADM were compared to GES results at the 4-hour measurement.

Statistical Methods

Mann-Whitney U and exact chi-square tests were used to compare continuous and categorical variables based on gastric emptying status. Pearson correlation was used to assess the relationship between continuous outcomes of GES and ADM. A P-value of < 0.05 was considered significant for all inference testing. Statistical analyses were performed using Rprogram, version 4.1.0 (R Development Core Team, Vienna, Austria).

Between 2011 and 2020, 226 patients underwent both ADM and GES testing. Ninety-nine (43.8%) patients were excluded due to vomiting during GES, and 36 (15.9%) were removed for not ingesting > 50% of a standard meal. Twenty-seven (11.9%) were removed for reasons related to the ADM, including no antral activity during fasting, insufficient calories to induce fed pattern, no MMC seen during fasting phase, or computer malfunction/inability to find manometry tracing (Supplementary Figure). In total, 32 had DGE meeting our inclusion criteria. These patients were then compared to 32 symptomatic patients with NGE meeting the inclusion criteria. When comparing patient groups with DGE and NGE, no statistically significant differences were noted in sex, BMI Z-score, or weight Z-score. More patients in the DGE group reported nausea than in the NGE group (27 vs 20, P = 0.049). No other symptoms significantly differed between the 2 groups (Table 1). There was no difference in use of prokinetic medications. The rate of macrolide prokinetic medication between the groups (13 vs 17, P = 0.450). The use of metoclopramide was not different (2 vs 3, P > 0.99 ). The use of domperidone included 2 in the DGE group only (P = 0.490). No patients were on amoxicillin/clavulaunate. The 2 tests, ADM and GES were done an average of 1.47 days apart from each other for the individual patients.

Table 1 . Demographic Summary by Gastric Emptying Group

Variable	Gastric emptying		Total (N = 64)	P-values
	Delayed (n = 32)	Normal (n = 32)
Age in years (at time of ADM)	13.2 ± 4.1	13.5 ± 4.3	160.3 ± 50.0	0.781a
Z score of BMI (on date of ADM)	0.0 ± 1.0	0.5 ± 1.1	0.3 ± 1.1	0.093a
Z score of weight (on date of ADM)	–0.3 ± 1.3	0.3 ± 1.1	0.0 ± 1.2	0.074a
Gender				0.077b
Female	23 (71.9)	16 (50.0)	39 (60.9)
Male	9 (28.1)	16 (50.0)	25 (39.1)
Primary GI diagnosis/indication for ADM study
Gastroparesis	5 (15.6)	8 (25.0)	13 (20.3)	0.523c
Nausea	4 (12.5)	8 (25.0)	12 (18.8)	0.348c
Vomiting	8 (25.0)	7 (21.9)	15 (23.4)	> 0.999c
Pseudo-obstruction	2 (6.3)	2 (6.3)	4 (6.3)	> 0.999c
Rumination	7 (21.9)	2 (6.3)	9 (14.1)	0.152c
Upper GI dysmotility	4 (12.5)	5 (15.6)	9 (14.1)	> 0.999c
Feeding intolerance	2 (6.3)	0 (0.0)	2 (3.1)	0.494c
Symptoms
Nausea	27 (84.4)	20 (62.5)	47 (73.4)	0.049b
Vomiting	20 (62.5)	24 (75.0)	44 (68.8)	0.286b
Abdominal pain	29 (90.6)	24 (75.0)	53 (82.8)	0.183c
Abdominal distention	6 (18.8)	10 (31.3)	16 (25.0)	0.241b
Bloating	8 (25.0)	2 (6.3)	10 (15.6)	0.076c
Early satiety	7 (21.9)	3 (9.4)	10 (15.6)	0.305c
Postprandial fullness	1 (3.1)	2 (6.3)	3 (4.7)	> 0.999c
Weight loss	12 (37.5)	7 (21.9)	19 (29.7)	0.175b
Decreased oral intake/appetite	10 (31.3)	12 (37.5)	22 (34.4)	0.598b
Feeding intolerance	11 (34.4)	10 (31.3)	21 (32.8)	0.793b
Frequent belching	2 (6.3)	1 (3.1)	3 (4.7)	> 0.999c

^aP-values from Mann Whitney U test.

^bP-values from Chi-square test.

^cP-values from Fisher’s exact test.

ADM, antroduodenal manometry; BMI, body mass index; GI, gastrointestinal.

Data are presented as mean ± SD or n (%).

ADM findings differed between the DGE and NGE groups (Table 2). The DGE group had lower post-prandial MI than the NGE group (9.4 vs 11.2, P = 0.005). The mean post-prandial number of antral contractions was significantly lower in the DGE group than in the NGE group (21.8 vs 49.6, P < 0.001). The original ADM interpretation by the managing pediatric neurogastroenterologist was more likely to conclude a normal study in the NGE group than the DGE group (75% vs 41%, P = 0.005). The pediatric neurogastroenterologist interpreting the study diagnosed antral hypomotility more often in the DGE group than the normal group (41% vs 9%, P = 0.008).

Table 2 . Associations Between Antroduodenal Manometry Outcomes and Gastric Emptying

Variable	Gastric emptying		Total (N = 64)	P-value
	Delayed (n = 32)	Normal (n = 32)
Fasting period: antral MI	8.7 ± 2.0	8.9 ± 2.2	8.8 ± 2.1	0.417b
Fasting period: number of antral contractions for MI	14.0 ± 13.4	16.4 ± 12.9	15.2 ± 13.1	0.334b
Post-prandial: antral MI	9.4 ± 2.6	11.2 ± 1.9	10.3 ± 2.4	0.005b
Post-prandial: number of antral contractions for MI	21.8 ± 24.0	49.6 ± 37.0	35.7 ± 34.0	< 0.001b
MI ratioa	1.0 ± 0.6	0.8 ± 0.2	0.9 ± 0.5	0.074b
Fasting period: are all MMCs antegrade or retrograde?				0.612c
Antegrade	29 (90.6)	31 (96.9)	60 (93.8)
Retrograde	3 (9.4)	1 (3.1)	4 (6.3)
Antroduodenal manometry clinician interpretation
Normal	13 (40.6)	24 (75.0)	37 (57.8)	0.005d
Rumination syndrome	4 (12.5)	6 (18.8)	10 (15.6)	0.735c
Antral hypomotility	13 (40.6)	3 (9.4)	16 (25.0)	0.008c
Neuropathic dysmotility: mild	7 (21.9)	6 (18.8)	13 (20.3)	0.755d
Neuropathic dysmotility: moderate	2 (6.3)	1 (3.1)	3 (4.7)	> 0.999c
Neuropathic dysmotility: severe	1 (3.1)	0 (0.0)	1 (1.6)	> 0.999c
Myopathic dysmotility	2 (6.3)	0 (0.0)	2 (3.1)	0.497c
Pseudo-obstruction	1 (3.1)	0 (0.0)	1 (1.6)	> 0.999c
Other	7 (21.9)	4 (12.5)	11 (17.2)	0.504c

^aFasting MI/post-prandial MI.

^bP-values from Mann Whitney U test.

^cP-values from Fisher’s exact test.

^dP-values from the chi-square test

MI, motility index; MMC, migrating motor complex.

Four-hour emptying GES values were compared to categorical ADM outcomes, such as phase III MMC pattern direction and presence of the antral component of phase III (Fig. 1). Patients with an antegrade phase III had a higher average four-hour emptying value than patients with a retrograde phase III (83.9% vs 59.2%, P = 0.022; Fig. 1). Similarly, the 4-hour emptying values for patients with intact antral components during the phase III MMC were higher than patients without an antral component (86.5% vs 70.3%, P = 0.003).

Figure 1. Relationship between gastric emptying scintigraphy (GES) 4-hour values (percentage emptied) and categorical antroduodenal manometry outcomes. (A) Fasting migrating motor complex (MMC) direction. (B) Presence of antral component in MMC pattern. Gastric emptying at 4 hr was significantly faster in pediatric patients (83.9%) with antegrade phase III MMC fasting patterns than those with retrograde phase III patterns (59.2%). Patients who did not have an antral component to their phase III MMC pattern had slower gastric emptying on GES (70.3%) than those with antral component (86.5%).

Comparisons were not significant between continuous 4-hour gastric emptying values and postprandial MI (r = 0.045, P = 0.721), postprandial number of antral contractions (r = 0.032, P = 0.814), or MI ratio (r = –0.13, P = 0.314) (Fig. 2).

Figure 2. Relationship between gastric emptying scintigraphy (GES) 4-hr values and continuous antroduodenal manometry (ADM) outcomes: (A) post-prandial motility index; (B) total post-prandial antral contractions for 1 hour after meal ingestion. No significant correlation between GES and ADM outcomes. The shaded area is 95% confidence interval, each point is an individual patient.

Our study found that pediatric patients with DGE had fewer total antral contractions and a lower antral MI in the post-prandial period than patients with normal gastric emptying. We also found that patients with retrograde phase III of the MMC activity or lack of an antral component in the fasting period were associated with decreased gastric emptying.

Similar to our results relating to patients with DGE, adult studies have also reported that patients with DGE often have both decreased frequency and strength of antral contractions in the post-prandial period.^23,24 This correlation between antral contractions and the degree of gastric emptying was also found in the gastric emptying rates of adult men and women. Specifically, women exhibit lower antral contraction amplitude on ADM and decreased gastric emptying as measured by GES than men.²⁵

Our study also examined continuous GES values to further investigate how the severity of gastric emptying correlates with ADM parameters. Patients who were included in the DGE group had emptied between 88% and 51% of the meal at the 4-hour mark during GES, representing a wide range of values. We examined whether more severely DGE correlated with more severe ADM findings. This study compared the values at the 4-hour mark using GES to ADM measures of interest, including the fasting period antral MI, antral contraction number during the fasting period, post-prandial antral MI, post-prandial antral contraction number, and the antral MI ratio. None of these correlated significantly with those measurements. Thus, it seems that the severity of delayed emptying on GES does not correlate with ADM measurements.

We also assessed whether specific ADM abnormalities correlated with continuous values of gastric emptying. Patients with retrograde phase III MMC contractions in the fasting phase had significantly DGE, suggesting that retrograde phase III MMCs indicate intestinal dysmotility.²² These results suggest that intestinal dysmotility may function as a “roadblock” contributing to a delayed solid gastric emptying. Also, if patients did not have an antral component in their phase III pattern during fasting, gastric emptying was significantly delayed. Even though studies in adults indicate that phase III patterns lack an antral component 30% of the time,²⁶ these data suggest that the complete lack of an antral component in any phase III MMC could be an abnormality correlating with DGE.

Our study differs from previous studies directly comparing ADM and GES in adult patients²⁰ as ours did not include patients in intensive care units. Significant disease and critical illness can affect GI motility patterns in children and adults, likely due to antral pressure wave suppression, proximal and distal stomach activity incoordination, and lower gastric emptying.^27,28 Overall, our study population better represents the pediatric patients commonly referred for motility studies in the outpatient care setting.

Our study of 64 patients represents the largest published comparison between ADM and GES in adults or children. These patients also had a short interval, an average of 1.47 days, between their ADM and GES testing. This very short time between the testing makes us believe the data can be quite comparable. Our cohorts were also similar in age, weight, BMI, and sex, and we established rigorous criteria for exclusion to ensure that studies contained all components needed for comparison. While these criteria were used to ensure the best quality of data to interpret these studies, it is a limitation as it required that we eliminate over 70% of the patients in the initial data set. These patients were removed due to imperfect ADM or GES which is frequently the reality in patient care. This was a single-center, retrospective design, which allowed for selection bias. Also, adult values for gastric emptying results were used which could limit reproducibility in a pediatric population.

Our data raises the question that one may be able to use just the ADM to provide valuable information in the diagnostic evaluation of pediatric patients with upper GI symptoms. There were much fewer patients being excluded for their ADM (n = 27) compared to those excluded based on the GES criteria (n = 135). This could be due to the fact that the GES is designed to use an adult-sized meal with adult diagnostic parameters. About 73% of the excluded GES studies were due to vomiting, which invalidates that test’s results. Vomiting can certainly occur during ADM testing as well, but there are still meaningful data from this test that can be used for diagnostic purposes even when vomiting occurs. Since our study shows that abnormalities in ADM are common in patients with DGE, one may be able to consider using ADM as the diagnostic test of choice for children with refractory upper GI symptoms in the right patient. Especially if both resources are available to the clinician at their center.

In conclusion, we conclude that DGE is associated with several abnormal ADM findings in pediatric patients. Patients with DGE have both decreased post-prandial antral MI and total number of post-prandial antral contractions. Retrograde phase III contractions and phase III MMCs lacking in antral component occur more frequently in patients with delayed gastric emptying. Further prospective research on these diagnostic tests should aim to determine if they predict clinical outcomes.

We would like to acknowledge our Motility nurses that make this testing possible

None.

None.

Raul E Sanchez: participated in data collection, statistical analysis, manuscript preparation, and manuscript edits; Elizabeth Reichard, Adam Bobbey, Neetu Bali Puri, Peter L Lu, Desale Yacob, and Carlo Di Lorenzo: participated in significant manuscript edits, participated in significant manuscript edits; Kent Williams: participated in research conceptualization, statistical analysis, and significant manuscript edits; and Karla K H Vaz: participated in research conceptualization, data collection, and significant manuscript edits.

Note: To access the supplementary figure mentioned in this article, visit the online version of Journal of Neurogastroenterology and Motility at http://www.jnmjournal.org/, and at https://doi.org/10.5056/jnm24057.