Accepted: 13 April 2017

DOI: 10.1111/pan.13172

RESEARCH REPORT

Ultrasound assessment of gastric emptying after breakfast in

healthy preschool children

€ mpelmann1 | Robert Su
Anne E. Su € mpelmann2 | Michael Lorenz3 | Ilona Eberwien4 |
Nils Dennhardt2 | Dietmar Boethig5 | Sebastian G. Russo1

1
Department of Anesthesiology, University
of Goettingen Medical Center, Goettingen,
Germany
2
Clinic for Anesthesiology and Intensive
Care Medicine, Hanover Medical School,
Hanover, Germany
3
Department of General, Visceral and
Pediatric Surgery, University Medical
Center Goettingen, Goettingen, Germany
4
Kindergarten, University Medical Center
Goettingen, Goettingen, Germany
5
Clinic for Cardiac, Thoracic, Transplant and
Vascular Surgery, Hanover Medical School,
Hanover, Germany
Correspondence
Prof Dr Sebastian G. Russo, Department of
Anesthesiology, University of Goettingen
Medical Center, Goettingen, Germany.
Email: s.russo@medizin.uni-goettingen.de
Section Editor: Brian Anderson
Summary
Background: In current guidelines, 6 hours of fasting is recommended for solids to limit
the risk of pulmonary aspiration during anesthesia in children. Ultrasonography has
recently been introduced to evaluate gastric volumes in children in the context of pre-
anesthetic fasting. Therefore, in this study, we firstly evaluated the precision of ultrasound
assessment of gastric volume in an experimental setting and secondly studied gastric
emptying times after a normal breakfast in healthy preschool children using ultrasound.
Methods: In a preliminary experiment, a pear-shaped elastic balloon was filled and
emptied in 50 mL steps from 0 to 500 mL with water. After each step, the balloon
antral area was measured using ultrasonography. Thereafter, gastric emptying was
examined in healthy preschool children after normal breakfast by sonographic mea-
surements of the gastric antral area in right lateral decubitus position at two consec-
utive timepoints. Correlation coefficients (Pearson, 95% CI) between the balloon
antral area and the balloon volume or gastric antral area and fasting time were cal-
culated and gastric emptying time was extrapolated by linear regression. Data are
presented as mean (range).
Results: In the balloon experiment, the balloon volume correlated significantly with
the balloon antral area (63 measurements, r=.96, P<.0001, 95% CI 0.93 to 0.97). In the
preschool child measurements, a total of 30 children (age 47 (36-66) months) were
included. The gastric antral area correlated significantly with fasting time (r=!.69,
P<.0001, 95% CI !0.8 to !0.51). The first gastric antral area after breakfast was sig-
nificantly higher when compared to the second gastric antral area before lunch (10.4
" 3.7 (1.7-17.8) vs 5.5 " 2.6 (1.4-11.8) cm2; mean difference !5.04, 95% CI !6.3 to
!3.8, P<.0001). The calculated mean gastric emptying time was 236 minutes.
Conclusion: The results of the balloon experiment showed a high correlation between
balloon antral area and balloon volume. In the preschool child measurements, gastric
antral area correlated with fasting time, and the mean gastric emptying time was lower
than 4 hours after breakfast. These results support a more liberal perioperative fasting
regimen after a light meal or breakfast in routine pediatric anesthesia.

KEYWORDS
adverse events, child, research, ultrasound

Anne E. S€
umpelmann and Robert S€
umpelmann are contributed equally to the study and share responsibility as first author.

816 | © 2017 John Wiley & Sons Ltd wileyonlinelibrary.com/journal/pan Pediatric Anesthesia. 2017;27:816–820.
 €
SUMPELMANN ET AL.
1 | INTRODUCTION
In recent decades, recommendations for preoperative fasting times
decreased from ‘NPO (nil per os) from midnight’ to lower limits of
6 hours for solids, 4 hours for breast milk, and 2 hours for clear flu-
1-3
ids (6-4-2-rule ). Nevertheless, prolonged preoperative fasting is
still common in pediatric anesthesia and may lead to significant
patient discomfort with uncooperative behavior, hunger, thirst, lipol-
ysis, dehydration, and hypotension.4-7 Recent studies showed that
gastric emptying depends on food type,8 and furthermore that more
liberal fasting regimes did not affect the aspiration risk during com-
petent pediatric anesthesia, but improve the perioperative experi-
ence of parents and children.9 Ultrasonography has recently been
introduced to evaluate gastric volumes in children in the context of
preanesthetic fasting.10 Therefore, in this study, we firstly evaluated
the precision of ultrasound assessment of gastric volume in an
experimental setting, and secondly studied gastric emptying times
after a normal breakfast in healthy preschool children using ultra-
sound.
2 | METHODS
2.1 | Balloon experiment
On a pear-shaped elastic balloon, the half distance between the
maximum and minimum diameter was marked and defined as balloon
antrum. The balloon was then vented in a water bath, filled with a
syringe over a stopcock in 50 mL steps from 0 to 500 mL with
water and was emptied in 50 mL steps thereafter. After each step,
two orthogonal diameters (D1 and D2) of the balloon antrum were
measured in the water bath using a conventional ultrasound probe
(Sono Site Edge ultrasound device, Sonosite Inc., Bothell, USA), and
the balloon antral area (BAA) was calculated (BAA = p 9 D1 9 D2/
4). This was repeated three times. The balloon volume (BV) was cal-
culated as gastric volume (GV) using the formulas of Schmitz (GV
[ml kg!1] = 0.0093 9 gastric antral area (GAA) [mm2] !0.96) 11
and
2
Spencer (GV [ml]) = !7.8 + (3.5 9 GAA [cm ]) + 0.127 9 age
[months]).10 As BAA includes no gastric wall, GAA was set as
BAA + 1 cm2. Age was set as 47 months and weight as 16 kg.
2.2 | Measurements in children
This part was approved by the local ethics committee (No. 15/11/
15) and registered in the German registry of clinical studies (DRKS-
ID: DRKS00010158). After written informed consent was obtained
by the parents, healthy children in the Kindergarten of the University
of Goettingen Medical Center were asked to volunteer and were
informed in an age-appropriate manner. After fasting overnight, the
children had a normal Kindergarten breakfast in the morning as per
their wishes, including bread, fruits, cereals, milk, tea, or water. After
breakfast, they were told not to drink or eat until lunch, and they
were then examined in groups of two or three, firstly soon after
breakfast, and secondly later before lunch. The time between the
| 817
What is already known
• In current guidelines, 6 hours fasting is recommended for
solids to limit the risk of pulmonary aspiration during
anesthesia in children.
What this article adds
• The mean gastric emptying time after a normal breakfast
was lower than 4 hours in healthy preschool children.
These results support a more liberal fasting regimen after
a light meal or breakfast in routine pediatric anesthesia.
end of breakfast and the ultrasound examination was defined as
fasting time (FT). All ultrasound examinations were performed by AS,
who had been instructed and supervised by experienced pediatri-
cians. The children were positioned on their right side (RLD—right
lateral decubitus position), and the gastric antrum was scanned using
a conventional ultrasound probe (Sono Site Edge ultrasound device,
Sonosite Inc., Bothell, USA) in a sagittal plane between the liver and
the pancreas as described by van de Putte et al.12 Two orthogonal
diameters (D1 and D2) of the gastric antrum including the gastric wall
were measured and the gastral antral area (GAA) was calculated as
described above for the balloon antral area. The gastric volume (GV)
was calculated using the Schmitz formula as described above for the
balloon volume.11 The mean gastric emptying time (GET) was
defined as calculated fasting time for GAA=1 cm2 and GV=0 mL
using the linear regression equations found for FT vs GAA or GV,
respectively. During the sonographic examinations, the children were
allowed to watch children’s movies on a tablet computer, and after
the second examination, they were rewarded with a small gift.
2.3 | Data analysis
All recorded data were analyzed using MS Excel (Excel 2010;
Microsoft, Seattle, USA), GraphPad Prism (Prism 7; Graph Pad Soft-
ware Inc., San Diego, USA), and MedCalc (MedCalc Statistical Soft-
ware version 17.4, MedCalc Software bvba, Ostend, Belgium)
software tools. According to a post hoc power analysis, a sample
size of 24 would result in a 97.5% power to detect the described
fasting time effect on gastric antral area with an error probability
of 5%. Demographic data and GAA were presented as mean "
standard deviation (range) and difference in mean and 95% confi-
dence interval (95% CI), respectively. Correlation (Pearson, 95% CI)
and linear regression analysis were performed to detect depen-
dency of BAA and BV in the balloon experiments and GAA, GV,
and FT in the children’s measurements. Bland-Altman analysis was
used to compare the measured and calculated BV in the balloon
experiments. Normal distribution was checked with the D’Agostino-
Pearson’s test, and a paired t-test was used to compare GAA after
breakfast and before lunch, all with a predefined significance level
of a=0.05.
818 | €
SUMPELMANN ET AL.

3 | RESULTS

3.1 | Balloon experiment

A total of 63 measurements of balloon antral areas (BAA) were per-
formed with syringe-filled balloon volumes (BV) from 0 to 500 mL.
Measuring BAA was possible in all cases (see Figure 1). BV corre-
lated significantly with BAA (r=.96, P<.0001, 95% CI 0.93 to 0.97;
Figure 2), and the calculated BV using the Schmitz and Spencer for-
mulas correlated significantly with BV (for both r=.96, P<.0001, 95%
CI 0.93 to 0.97). The Bland-Altman plots of the measured and the
calculated BV are presented in Figure 3.

3.2 | Measurements in children

A total of 30 children (17 boys, 13 girls) were included in this
F I G U R E 2 Scatter plots of balloon volume (BV) and balloon
study. Their age was 47 " 8.2 (36-66) months, their body weight
antral area (BAA) measured by ultrasonography
was 16 " 1.7 (13.2- 19.2) kg and their body height was 104 "
6.3 (96-120) cm. Measurement of gastric antral area (GAA) was
possible in 95% of the cases, while in 5% air hinders the visual-
ization (Figure 4). The first measurement was performed 51 " 31
(5-140) minutes and the second one 146 " 33 (40-220) minutes
after breakfast. GAA correlated significantly with the fasting time
(r=!.69, P<.0001, 95% CI !0.8 to !0.51; Figure 5). The first GAA
after breakfast was significantly higher when compared to the sec-
ond GAA before lunch (10.4 " 3.7 (1.7-17.8) vs 5.5 " 2.6 (1.4-
11.8) cm2; mean difference !5.04, 95% CI !6.3 to !3.8,
P<.0001). The calculated gastric volume (GV) moreover correlated
significantly with the fasting time (r=!.69; P<.0001, 95% CI !0.8
to !0.51; Figure 5). The calculated mean gastric emptying time
was 236 min for GAA = 1 cm2 and 232 min for GV = 0 mL (re-
gression equations GAA = !0.05 9 FT + 12.8 and GV= !0.77 9
F I G U R E 3 Bland-Altman plots of calculated minus measured
FT + 178.9). balloon volume (BV) against mean of calculated and measured BV;
dashed lines indicate bias and dotted lines indicate the 95% limits of
agreement; (A) calculated according to Schmitz et al.,11 (B) calculated
according to Spencer et al.10

4 | DISCUSSION

F I G U R E 1 Sonographic image of the cross-sectional area of an
elastic pear-shaped balloon filled with water
The main findings of this study were firstly that in the balloon exper-
iment, the balloon antral area correlated with the balloon volume,
but the results of the calculated balloon volume were imprecise, and
secondly that in the children’s measurements, the gastric antral area
correlated with fasting time and that the calculated mean gastric
emptying time after a normal breakfast was lower than 4 hours in
healthy preschool children.
Various techniques are available to study gastric emptying, eg,
scintigraphy, magnetic resonance imaging, and ultrasonography.13
Although ultrasonography is noninvasive and mostly easily available,
the operator needs adequate training to visualize gastric structures,
and air may hinder the view on parts of the stomach. Most fre-
quently, ultrasonography was used for indirect assessment of gastric
volume by measurement of the cross-sectional area of the gastric
 €
SUMPELMANN ET AL.
F I G U R E 4 Ultrasonographic image of the gastric antrum (GA)
below the liver (L) of a preschool child
antrum. In the beginning, this technique was evaluated in adults,14,15
10,11,16
and later on also in pediatric patients. All studies found a sig-
nificant correlation between the gastric antral area and the gastric
volume suctioned via a gastric tube or endoscope or measured using
magnetic resonance imaging. In the review of van de Putte et al.,12
sonographic examination of the gastral area was recommended in a
right lateral decubitus position, because this leads to a downward
shift of gastric content into the gastric antrum and to an upward
shift of air outside the field of view.
Our balloon experiment also showed a significant correlation
between sonographically measured balloon antral area and balloon
volume, but with higher correlation coefficients when compared to
previous clinical studies with children.10,11,16 We hypothesized that
this was possibly caused by a more regular expansion of an elastic
balloon compared to the more irregular shape of a child’s stomach.
In accordance with Schmitz et al.,11 the Bland-Altman analysis of
F I G U R E 5 Scatter plots of preschool
children’s fasting time and gastric antrum
(GA) or gastric volume (GV), respectively,
measured by ultrasonography (GAA) and
calculated according to Schmitz et al. (GV) 11
| 819
the measured and calculated BV showed a high bias with both for-
mulas and a progressive underestimation of large values with the
Spencer formula. This is possibly due to the fact that the Spencer
formula is derived from a fasted pediatric cohort with a low range
of residual gastric volume and the Schmitz formula from nonfasted
children with a wider range of gastric volume. From our point of
view, a volume deduction of a complex three-dimensional body
from a two-dimensional cross-sectional area of the body is gene-
rally questionable. Despite the imprecision of the calculated abso-
lute gastric volume values, all studies showed that gastric emptying
is accompanied by a continuous decrease in gastric antral area and
that the residual gastric volume is minimal when the gastric antral
area is close to 1 cm2. As the fixed time interval between breakfast
and lunch in the Kindergarten was lower than 4 hours, we mea-
sured the gastric antral area at two consecutive timepoints after
breakfast to achieve an indirect measure for the calculation of the
gastric emptying time.
The mean gastric emptying time of about 4 hours after a Kinder-
garten breakfast found in our study is significantly shorter than the
recommended fasting time of 6 hours for solids in current guide-
lines.3 In accordance, Schmitz et al.17 found low residual gastric con-
tents 4 hours after breakfast using magnetic resonance imaging in
school-age children. A recent meta-analysis of 49 studies with
1457 individuals from 28 weeks of gestation to adults showed that
meal type and not age was the significant covariate of gastric empty-
ing with mean gastric residence times of 45 minutes for aqueous
solutions, 57 minutes for breast milk, 64 minutes for formula milk,
87 minutes for semisolid food, and 98 minutes for solid test meals.8
However, all studies found considerable interindividual variations in
gastric emptying time, and significant residual gastric volumes are
possible even after prolonged fasting. In the Kindergarten, we
noticed again that during daytime intervals between meals in pre-
school children were normally lower than 6 hours, and that addition-
ally the children also frequently ate or drank something in between.
As a consequence, in some institutions, shorter fasting times of
4 hours are routinely used for light meals (eg, toast and clear fluid)
as suggested by Schmitz et al.17 Experience has shown that this
practice is safe, but safety studies under clinical conditions with a
high sample size are still needed.
820 | €
In a previous study, we were able to show that an optimized
fasting management was effective to reduce fasting times close to
current guidelines and to improve the metabolic and hemodynamic
condition during induction of anesthesia in young children.18 Unfor-
tunately, in daily routine, keeping fasting times short is often more
difficult than expected. Predicting the timing of an operation is often
inaccurate, and the surgical schedule is frequently subject to changes
due to emergencies or cancellations. In this context and in light of
increasing evidence of low residual gastric contents 4 hours after
light meals and of the low perioperative aspiration risk during com-
petent pediatric anesthesia,19,20 current fasting guidelines should be
critically reassessed, favoring more liberal modifications to avoid
unnecessary fasting and to facilitate the perioperative processes.16
The presented study has some limitations. Due to the observa-
tional study design, the Kindergarten breakfast could not be stan-
dardized. As a consequence, the type and volume of food and
fluid intake varied interindividually, which may affect gastric emp-
tying. In addition, fasting after breakfast could not be controlled
and some children may have secretly nibbled something in
between. The maximum fasting time was limited because the time
interval between lunch and breakfast was fixed. The time intervals
between the sonographic measurements were inconsistent because
the children were sometimes busy with other things or difficult to
find.
In conclusion, the current study showed that the mean gastric
emptying time was lower than 4 hours after normal breakfast in
healthy preschool children. These results support a more liberal peri-
operative fasting regimen after light meals or breakfast in routine
pediatric anesthesia.
ACKNOWLEDGMENTS
We thank the Kindergarten children and parents for participating in
this study.
ETHICAL APPROVAL
This study was approved by the local ethics committee (No. 15/11/
15) and registered in the German registry of clinical studies (DRKS-
ID: DRKS00010158).
REFERENCES
1. Cote" CJ. NPO after midnight for children–a reappraisal. Anesthesiol-
ogy. 1990;72:589-592.
2. Maltby JR. Fasting from midnight–the history behind the dogma. Best
Pract Res Clin Anaesthesiol. 2006;20:363-378.
3. Smith I, Kranke P, Murat I, et al., European Society of A. Periopera-
tive fasting in adults and children: guidelines from the European
Society of Anaesthesiology. Eur J Anaesthesiol. 2011;28:556-569.
4. Engelhardt T, Wilson G, Horne L, Weiss M, Schmitz A. Are you hun-
gry? Are you thirsty?–fasting times in elective outpatient pediatric
patients. Paediatr Anaesth. 2011;21:964-968.
SUMPELMANN ET AL.
5. Dennhardt N, Beck C, Huber D, et al. Impact of preoperative fasting
times on blood glucose concentration, ketone bodies and acid-base
balance in children younger than 36 months: a prospective observa-
tional study. Eur J Anaesthesiol. 2015;32:857-861.
6. Brady M, Kinn S, Ness V, O’Rourke K, Randhawa N, Stuart P. Preop-
erative fasting for preventing perioperative complications in children.
Cochrane Database Syst Rev. 2009;(4):CD005285.
7. Maekawa N, Mikawa K, Yaku H, Nishina K, Obara H. Effects of 2-,
4- and 12-hour fasting intervals on preoperative gastric fluid pH and
volume, and plasma glucose and lipid homeostasis in children. Acta
Anaesthesiol Scand. 1993;37:783-787.
8. Bonner JJ, Vajjah P, Abduljalil K, et al. Does age affect gastric emp-
tying time? A model-based meta-analysis of data from premature
neonates through to adults. Biopharm Drug Dispos. 2015;36:245-257.
9. Andersson H, Zaren B, Frykholm P. Low incidence of pulmonary
aspiration in children allowed intake of clear fluids until called to the
operating suite. Paediatr Anaesth. 2015;25:770-777.
10. Spencer AO, Walker AM, Yeung AK, et al. Ultrasound assessment of
gastric volume in the fasted pediatric patient undergoing upper gas-
trointestinal endoscopy: development of a predictive model using endo-
scopically suctioned volumes. Paediatr Anaesth. 2015;25:301-308.
11. Schmitz A, Schmidt AR, Buehler PK, et al. Gastric ultrasound as a
preoperative bedside test for residual gastric contents volume in
children. Paediatr Anaesth. 2016;26:1157-1164.
12. Van de Putte P, Perlas A. Ultrasound assessment of gastric content
and volume. Br J Anaesth. 2014;113:12-22.
13. Hellstrom PM, Gryback P, Jacobsson H. The physiology of gastric
emptying. Best Pract Res Clin Anaesthesiol. 2006;20:397-407.
14. Bouvet L, Mazoit JX, Chassard D, Allaouchiche B, Boselli E, Ben-
hamou D. Clinical assessment of the ultrasonographic measurement
of antral area for estimating preoperative gastric content and vol-
ume. Anesthesiology. 2011;114:1086-1092.
15. Perlas A, Davis L, Khan M, Mitsakakis N, Chan VW. Gastric sonogra-
phy in the fasted surgical patient: a prospective descriptive study.
Anesth Analg. 2011;113:93-97.
16. Schmitz A, Thomas S, Melanie F, et al. Ultrasonographic gastric
antral area and gastric contents volume in children. Paediatr Anaesth.
2012;22:144-149.
17. Schmitz A, Kellenberger CJ, Liamlahi R, Fruehauf M, Klaghofer R,
Weiss M. Residual gastric contents volume does not differ following
4 or 6 h fasting after a light breakfast - a magnetic resonance imag-
ing investigation in healthy non-anaesthetised school-age children.
Acta Anaesthesiol Scand. 2012;56:589-594.
18. Dennhardt N, Beck C, Huber D, et al. Optimized preoperative fast-
ing times decrease ketone body concentration and stabilize mean
arterial blood pressure during induction of anesthesia in children
younger than 36 months: a prospective observational cohort study.
Paediatr Anaesth. 2016;26:838-843.
19. Walker RW. Pulmonary aspiration in pediatric anesthetic practice in
the UK: a prospective survey of specialist pediatric centers over a
one-year period. Paediatr Anaesth. 2013;23:702-711.
20. Neuhaus D, Schmitz A, Gerber A, Weiss M. Controlled rapid
sequence induction and intubation - an analysis of 1001 children.
Paediatr Anaesth. 2013;23:734-740.
How to cite this article: Su
€mpelmann AE, Su
€mpelmann R,
Lorenz M, et al. Ultrasound assessment of gastric emptying
after breakfast in healthy preschool children. Pediatr Anesth.
2017;27:816–820. https://doi.org/10.1111/pan.13172