Drubach, Laura A.*; Kourmouzi, Vasiliki*; Cao, Xinhua*; Zurakowski, David†; Fahey, Frederic H.*
*Department of Radiology
†Department of Anesthesia, Children's Hospital Boston, Harvard Medical School, Boston, MA.
Address correspondence and reprint requests to Laura A. Drubach, MD, Department of Radiology, Children's Hospital Boston, 300 Longwood Ave, Boston, MA 02115 (e-mail: email@example.com).
Received 1 August, 2011
Accepted 6 February, 2012
The authors report no conflicts of interest.
Objectives: The geometric mean is the recommended method for calculating gastric emptying in adults to correct for variable attenuation during the study. In children, it has been reported that a single posterior acquisition is sufficient. Our objective was to evaluate the relation between the values of gastric emptying in children obtained by posterior view only and by the geometric mean of conjugate anterior and posterior views.
Methods: The gastric residual of a standard meal was calculated in 81 children of different ages. The calculation of the gastric residual was performed with different methods, posterior only and geometric mean of anterior and posterior acquisitions. The variation between these 2 methods was evaluated in different age groups and different body weights.
Results: There was a high correlation (r = 0.942) between the values using posterior as compared with geometric mean for all of the patients. For children younger than 8 years and weighing <30 kg, there was no significant difference between the 2 methods for either liquid or solid meals (P = 0.89 and P = 0.11 for age; P = 0.95 and P = 0.80 for weight). For children older than 8 years and weighing >30 kg, there was no significant difference between the 2 methods for liquids (P = 0.57 for age; P = 0.69 for weight), but there was a significant difference with solids (P < 0.0001 for both age and weight).
Conclusions: In children younger than 8 years and weighing <30 kg, acquisition of a single posterior image is sufficient for calculating gastric emptying at 1 hour postingestion for either liquid or solid meals. In children older than 8 years and weighing >30 kg, acquisition of both anterior and posterior images with geometric mean calculation is recommended when a solid meal is used. If a liquid meal is used in patients older than 8 years, the posterior only may be adequate.
It is well established that solid-phase gastric emptying studies in adults must be performed with conjugate anterior and posterior views to determine a geometric mean and account for the differential soft tissue attenuation that occurs as food moves from the posterior gastric fundus to the more anterior gastric body and antrum during emptying (1–3). Use of a single left anterior oblique (LAO) view has been described as an alternative (4), although it is not the preferred method of imaging (3). It has been reported that in the pediatric population, as opposed to the recommendations in adults, a single posterior acquisition is sufficient, without the need to perform either conjugate anterior-posterior views or the LAO acquisition (5). Using only a posterior acquisition is desirable in younger children because imaging with a detector under the imaging table is often better tolerated than positioning a detector over the child's chest and abdomen, as is required for both the conjugate anterior-posterior and the LAO techniques. Having a large detector positioned anteriorly in close proximity to the child can cause significant patient anxiety and limits the technologist's access to the patient. Use of a single posterior detector allows the child to have better contact with a parent or technologist, facilitates immobilization, and allows better access to the patient in cases in which oxygen or suctioning is needed.
The objective of the present study was to determine the adequacy of gastric emptying studies performed in children using a single posterior view. We evaluated the correlation between the results of gastric emptying studies performed at 1 hour postingestion in children and young adults of different ages and body weight as determined using either a single posterior view or the geometric mean of conjugate posterior-anterior views.
The gastric emptying studies of 81 consecutive children and young adults performed between August 2010 and April 2011 were reviewed. Approval of our institutional research committee was obtained for this investigation. After a fasting period of at least 6 hours, the study was performed by mixing 99mTc sulfur colloid (15 μCi/kg [0.55 MBq/kg], minimum 200 μCi [7.4 MBq], and maximum of 500 μCi [37 MBq]) with the meal to be administered. Both solid-phase and liquid-phase meals were evaluated. Solid-phase meals consisted of labeled eggs or labeled cheese according to standard published protocols (6). Liquid-phase meals consisted of formula or milk. The solid-phase radioactive meal was administered orally. Liquid-phase meals were administered either orally or by nasogastric or gastrostomy tube. If the meal was given via nasogastric tube, the tube was removed after meal administration and before image acquisition. The patient was positioned supine on the imaging table and anterior and posterior views were acquired concurrently with either an ultra-high or a high-resolution collimator for 60 minutes (1 minute/view). Regions of interest (ROIs) were drawn manually about the stomach on the immediate and 1-hour images. The gastric residual at 1 hour postingestion was calculated by decay correcting the values and expressing the ratio as a percentage. The gastric residual was obtained using 2 different methods: using the posterior view only (post only) and calculating the geometric mean from the conjugate anterior-posterior views (geom mean). The same ROI was used for both the post only and geom mean methods. The correlation and mean difference between these 2 methods were determined overall and stratified according to patient's age and body weight. The gastric residual at 1 hour estimated using the 2 methods was compared using a paired t test. A 2-tailed P value <0.05 was considered significant. Power analysis indicated that the sample size of >80 patients provided 90% power in capturing a mean difference in percent residual between post only and geom mean methods within 10% using a paired t test and 80% power for detecting differences between the 2 methods stratified by patient's age, body weight, and type of meal (version 7.0, nQuery Advisor, Statistical Solutions, Saugus, MA). Statistical analysis was performed using the SPSS software package (version 19.0, SPSS Inc, Chicago, IL).
Eighty-one gastric emptying studies were performed in 81 patients (28 male and 53 female patients, age range 3 months–27 years; mean age 12.2 ± 5.8 years). Sixty-five patients had a solid meal study and 16 patients a liquid meal study. There were 17 patients younger than 8 years, 8 received a liquid meal (all 3 years old or younger) and 9 had a solid study. There were 64 patients older than 8 years: 56 received a solid meal and 8 had a liquid meal. There were 23 patients weighing <30 kg, 10 received a liquid meal, and 13 received a solid meal. There were 58 patients weighing >30 kg, 6 received a liquid meal, and 52 received a solid meal.
Considering both solid- and liquid-phase meals, there was a high positive correlation (r = 0.942; P < 0.001) in the gastric residual values between post only compared with geom mean for all of the patients (Fig. 1). Judging from the 45-degree line of identity, the evidence indicates a trend that the post only method underestimates the percent residual measurements obtained from the geom mean, particularly for solid meals. When evaluating the difference between the 2 methods and plotting this against patient age, it is clear that for older children (ie, older than 8 years) the methods show considerably less agreement, with the post only method generally yielding lower values for percent residual than the geom mean method (P = 0.02; Fig. 2A). The average difference throughout the age range was 5% lower for the post only method. Similarly, the agreement between the 2 methods according to weight suggests less intermethod agreement for heavier patients and better agreement for patients weighing <30 kg (P = 0.002; Fig. 2B). A summary of the percent residual according to method based on age and weight subgroups stratified by meal type is presented in Table 1. It is clear that for liquid meals, the 2 methods agree closely, showing no significant differences for any age or weight subgroup; however, for solid meals, the post only method demonstrates statistically significant lower percent residual values for patients older than 8 years (P < 0.001) and those weighing >30 kg (P < 0.001).
The radionuclide gastric emptying study is widely used in children because it provides a noninvasive and physiological method for measuring gastric emptying with minimal radiation exposure to the patient. Knowledge of gastric emptying is also clinically important because gastric distension is associated with a delay in the emptying of stomach contents and this can be a predisposing factor for the presence of gastroesophageal reflux (7).
Children older than 3 years and children with severe esophagitis and decreased lower esophageal sphincter pressure are especially at risk for abnormal gastric emptying (8). It has been suggested that delayed gastric emptying may be associated with abnormal function of the fundus and antral motor activity, as part of a generalized foregut motor dysfunction (8,9).
Most of the validation of the techniques used for gastric emptying evaluation has been performed in adults. A generally accepted standardized protocol for performing gastric emptying evaluation exists in adults (3). The image acquisition is performed using both anterior and posterior planar imaging, allowing calculation of the geometric mean for gastric ROIs, to correct for varying attenuation present on anterior and posterior images.
The situation in children is more complicated because no generally accepted approach to gastric emptying evaluation exists. This limits comparison of results at different institutions and has also led to the use of some techniques that may not be valid. The varying developmental stages in children make the adoption of a single protocol, as is used in adults, more difficult. Furthermore, many of the approaches presently used in children have not been fully validated. The challenges in developing clinical protocols for children include accommodating the nature of meals that can be ingested at different ages and the positioning of patients needed to allow for comfortable immobilization at varying ages. In infants, the meal used to assess gastric emptying is usually milk or formula. In older children, cooked eggs labeled with 99mTc sulfur colloid are commonly used to evaluate the gastric emptying of solids, although alternative meals using cheese have been shown to represent a good alternative for those children unable to tolerate eggs (6).
The goal of this investigation was to collect data that will allow standardization of image acquisition in children. The calculation of geom mean values recommended in the adult protocol requires posterior image acquisition with a camera positioned under the imaging table and anterior imaging with a camera that is positioned close to the patient's chest and face. This anterior camera positioning can be problematic in young children, who may experience anxiety and in whom adequate immobilization may become difficult. The presence of a large camera detector in close proximity to a small child may also limit access for medical attention. In children, therefore, the use of a single posterior image for gastric emptying calculation is highly desirable. It has been shown that there is no difference in the value obtained with anterior imaging only or with geom mean or LAO view when using glucose solution as the meal (10). Before this evaluation, however, it had not been established whether a single image acquisition could be reliably used with either semisolid meals such as formula, which are commonly used in young children, or with solid meals.
Our results show good agreement between the post only and the geom mean acquisition methods in children that were younger than 8 years and weighing <30 kg, for both liquid and solid meals. When performing solid meal gastric emptying in children older than 8 years and with a weight of >30 kg, the single posterior image underestimated the results provided by the geometric mean method and therefore is not an acceptable technique. The need for conjugate anterior and posterior imaging in older or heavier children likely reflects the same attenuation correction issues present in adults. Smaller children likely present limited soft tissue attenuation, allowing the use of a single posterior image. We found good agreement between the 2 methods in children older than 8 years and heavier than 30 kg when a liquid meal was used, suggesting that if a liquid meal is used a post only acquisition would be sufficient. We do recognize that this subgroup of patients was small; therefore, some caution is required in the interpretation of these findings. A larger number of patients with a liquid meal should be further explored to validate this interpretation.
In summary, gastric emptying studies performed in children younger than 8 years and weighing <30 kg can be successfully performed with post only image acquisition, regardless of whether a liquid phase or solid phase meal is used. In older or larger patients, conjugate anterior and posterior imaging must be performed for solid-phase gastric emptying.
1. Maurer AH, Knight LC, Charkes ND, et al. Comparison of left anterior oblique and geometric mean gastric emptying. J Nucl Med 1991; 32:2176–2180.
2. Ziessman HA, Fahey FH, Atkins FB, et al. Standardization and quantification of radionuclide solid gastric-emptying studies. J Nucl Med 2004; 45:760–764.
3. Donohoe KJ, Maurer AH, Ziessman HA, et al. Procedure guideline for gastric emptying and motility. J Nucl Med 1999; 40:1236–1239.
4. Fahey FH, Ziessman HA, Collen MJ, et al. Left anterior oblique projection and peak-to-scatter ratio for attenuation compensation of gastric emptying studies. J Nucl Med 1989; 30:233–239.
5. Heyman S. Gastric emptying in children. J Nucl Med 1998; 39:865–869.
6. Drubach LA, Kourmouzi V, Fahey FH. Cheese is a reliable alternative meal for solid-phase gastric emptying study. Nucl Med Commun 2010;31:430–3.
7. Holloway RH, Hongo M, Berger K, et al. Gastric distention: a mechanism for postprandial gastroesophageal reflux. Gastroenterology 1985; 89:779–784.
8. Di Lorenzo C, Piepsz A, Ham H, et al. Gastric emptying with gastro-oesophageal reflux. Arch Dis Child 1987; 62:449–453.
9. Behar J, Ramsby G. Gastric emptying and antral motility in reflux esophagitis. Effect of oral metoclopramide. Gastroenterology 1978; 74:253–256.
10. Phillips WT, McMahan CA, Lasher JC, et al. Anterior, posterior, left anterior oblique, and geometric mean views in gastric emptying studies using a glucose solution. Eur J Nucl Med 1995; 22:154–157.
© 2012 by European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology,