Gastric Ultrasound Assessing Gastric Emptying of Preoperative Carbohydrate Drinks: A Randomized Controlled Noninferiority Study : Anesthesia & Analgesia

Secondary Logo

Journal Logo

Original Research Articles: Original Clinical Research Report

Gastric Ultrasound Assessing Gastric Emptying of Preoperative Carbohydrate Drinks: A Randomized Controlled Noninferiority Study

Cho, Eun-Ah PhD*,†; Huh, Jin PhD*; Lee, Sung Hyun MD, PhD; Ryu, Kyoung-Ho PhD; Shim, Jae-Geum MD; Cha, Yun-Byeong MD; Kim, Mi Sung PhD; Song, Taejong PhD§

Author Information
doi: 10.1213/ANE.0000000000005411



  • Question: Does gastric emptying assessed with gastric ultrasound delayed by the preoperative carbohydrate drink ingested up to 2 hours before surgery, compared to the midnight fasting?
  • Findings: Gastric antral cross-sectional area and residual gastric volume were not different between the carbohydrate drinking group and the midnight fasting group.
  • Meaning: Carbohydrate drink ingested 2 hours before surgery does not increase the residual gastric volume compared to the midnight fasting.

Recently, updated fasting guidelines emphasizing minimal fasting time have been published. Although, many anesthesiologists are aware of these guidelines, the standard fasting policy have still not been revised in many hospitals.1 However, prolonged fasting induces a catabolic state, in which protein and fat break down to compensate for the depletion of hepatic glycogen storage to produce endogenous glucose via glycogenolysis and gluconeogenesis.2–4 Insulin resistance contributes to the main pathogenic features of postoperative metabolic response.2,3 Along with the surgical stress, preoperative fasting exacerbates postoperative insulin resistance and stimulates inflammatory responses.2,5 Insulin resistance as well as reduced insulin sensitivity decrease postoperative bowel function and increase complications, such as fatigue and surgical site infection.3 Therefore, for decades, efforts have been made to reduce preoperative fasting time.

Recently, carbohydrate beverages have been used preoperatively. It has been well demonstrated that preoperative carbohydrate drinks reduce anxiety, thirst,6 postoperative nausea and vomiting,7 and attenuate insulin resistance.8,9 Because carbohydrate drinks consist of polymers, which have lower osmolality than monomers, it is believed that carbohydrate drinks do not delay gastric emptying.10 Some studies support that gastric emptying time is not prolonged after preoperative carbohydrate drinking, and carbohydrates can be administered up to 2 hours before anesthesia.7,10–12 However, these studies have evaluated gastric fluid volume by insertion of a nasogastric tube, gamma camera, or by marker dilution, all of which are rather invasive techniques, expose patients to radiation, or are not feasible in awake patients because of discomfort.

Gastric ultrasound is noninvasive, easy to perform at bedside with minimal discomfort, and avoids radiation exposure.13 Therefore, it can be performed with fewer limitations in place and time. Furthermore, gastric ultrasound provides reliable quantitative and qualitative information about the nature of gastric contents before anesthesia.14 Thus, it is a useful tool for assessing the risk of pulmonary aspiration, which has greatly altered clinical management during anesthesia.13 The aim of our study was to demonstrate that the mean gastric antral cross-sectional area (CSA) after preoperative ingestion of carbohydrate drinks is noninferior to that of the fasted patients undergoing elective surgery.


Study Participants

We conducted this prospective, double-blinded, randomized placebo-controlled study at Kangbuk Samsung Hospital, Seoul, Republic of Korea, after approval from our hospital’s Ethics Committee (Kangbuk Samsung Medical Center institutional review board No. 2019-04-025-004, Chairperson: Kwan Joong Joo MD, PhD). Written informed consent was obtained from all participants. This trial was registered before patient enrolment at (NCT 03955926, principal investigator: Eun-Ah Cho, date of registration: March 4, 2020). Female patients scheduled for elective laparoscopic benign gynecologic surgery, aged between 18 and 70 years, and with American Society of Anesthesiologists (ASA) physical status I or II were included. Exclusion criteria included pregnancy, lactation, coexisting disease that delays gastric emptying (eg, obesity, diabetes, hiatal hernia, gastroesophageal reflux disease, ileus, or enteral tube feeding), psychiatric or mental disorders, alcoholism, or drug abuse.

Randomization and Blinding

Participants were randomly allocated to the 1 of the 2 groups in a 1:1 ratio. A study coordinator, who was only involved in randomization and allocation, performed computer-generated randomization and allocation of the patients using random block with stratification ( Randomized numbers were sealed in opaque envelopes in order and kept concealed until the study day. The investigator (T.S.) called the study coordinator the day before surgery to confirm patient allocation. Assessments of study outcomes were performed by investigators (E.-A.C. and J.-G.S.) blinded to group allocation.

Study Protocol

The day before surgery, all participants were allowed to eat and drink freely until midnight. For the nil per os (NPO) group, participants fasted from midnight until surgery. Participants in the NO-NPO group drank 400 mL of a carbohydrate beverage (NO-NPO, Daesang WelLife Co, Ltd, Seoul, Korea; 12.8% carbohydrates, 50 kcal/100 mL, 290 mOsm/kg) at midnight before surgery, and 100–200 mL every hour freely up to 2 hours before surgery. All participants received intramuscular 0.005 mg/kg glycopyrrolate injection 1 hour before anesthetic induction.

Ultrasonographic Assessment of Stomach

Ultrasound examinations were conducted in the waiting area before entering the operation room. All ultrasound examinations were performed by 2 staff anesthesiologists (E.-A.C. and J.G.S.), with previous experience in gastric ultrasound examination—more than 100 times for E.-A.C. and 50 times for J.G.S. They were blinded to each other’s measurements. A curvilinear array low-frequency transducer (1.6–4.6 MHz) and a LogiQ E (GE Healthcare, Piscataway, NJ) were used at a standard abdominal setting.

Ultrasonographic assessment was first performed in the semisitting position (45° head up) followed by the right lateral dependent position (RLDP). The gastric antrum was identified in the epigastric area in the parasagittal plane, using the left lobe of the liver, pancreas, inferior vena cava, and superior mesenteric vein as landmarks. Usually, the gastric antrum is found between the left lobe of the liver anteriorly and the head or neck of the pancreas posteriorly, slightly toward the right from the midline. The inferior vena cava is positioned posterior to the pancreas. The probe was tilted clockwise or counterclockwise to obtain the smallest, round-shaped, cross-sectional view of the antrum. Both qualitative and quantitative assessments were performed. For the qualitative assessment, the gastric antrum was defined as “empty” when it was flat without any fluid. When fluid was detected, the gastric antrum was defined as “fluid.” Based on the qualitative assessments, the gastric antrum was graded using the semiquantitative 3-point grading system: grade 0, the antrum appears empty in both semisitting position and RLDP; grade 1, gastric fluid is only visible in RLDP; and grade 2, gastric fluid is visible in both semisitting position and RLDP.15

For the quantitative assessments, 2 perpendicular diameters, the longest diameter (LD) and the shortest diameter (SD), were measured (Figure 1). These measurements included the outermost serosa layer, which appears echogenic in ultrasound. Measurements were performed between peristalsis. A CSA of the gastric antrum was calculated using the formula of the area of an ellipse: CSA = (LD × SD × π)/4 as described in our previous study.16 Gastric volume (GV) was calculated based on the following formula from Perlas et al:17

Figure 1.:
Representative figure of gastric CSA measurement, showing CSA measurements in the 2 perpendicular diameters. CSA indicates cross-sectional area; IVC, inferior vena cava; L, liver; P, pancreas.

After dividing GV by weight (kg), GV/kg and the incidence of GV/kg >1.5 mL/kg was assessed.18

Assessments of Ultrasound Image Quality

The ultrasound images were captured and saved as separate files. The ultrasound images were reviewed by an expert radiologist with more than 19 years of experience in abdominal ultrasound (approximately 850 examinations per year). The sonographer blinded by the group allocation, scored images with 5-point scale image quality score (IQS) as used in the previous study: 1, not interpretable; 2, barely interpretable; 3, adequate for interpretation, but of poor quality; 4, interpretable and of average quality; 5, interpretable and of superior quality.19

Assessments of Other Outcomes

The degrees of hunger, thirst, and anxiety were assessed using the visual analog scale, from 0 (not at all) to 10 (very much) in the waiting room right after gastric ultrasound assessment. Postoperative outcomes including length of hospital stay, postoperative blood glucose, time to first gas passage, time to initiate solid food, and time to able to walk.

Statistical Analysis

The primary end point was the antral CSA assessed by gastric ultrasound in RLDP in different fasting policies. The purpose of this study was to show that the risk of pulmonary aspiration is not increased with carbohydrate drink. Therefore, we hypothesized that the gastric antral CSA of the patients randomized to NO-NPO is noninferior to that of patients randomized to NPO. To assess for noninferiority, noninferiority delta of 2.8 was chosen for the following reasons. In our previous study, the baseline CSA after 8 hours of fasting was 5.0 ± 1.5 cm2, and it was statistically not different from 7.2 ± 2.9 cm2 which measured after postprandial 5 hours.16 Considering that the CSA in grade 2 stomach was 11.6 ± 3.2 cm2,15 we thought that CSA that can be admitted to be “not different” from the baseline value would be <8.4 cm2 (11.6 − 3.2 cm2). Based on these reports, we chose the average value between 2.2 cm2 (7.2 − 5.0 cm2) and 3.4 cm2 (8.4 − 5.0 cm2), as a noninferiority delta (2.8 cm2).

According to our preliminary data (unpublished data), the gastric residual volumes in surgical patients were 0.008 mL/kg in the fasted patients (n = 6, over 8 hours of fasting including solid and clear fluid), and 0.28 mL/kg in the patients who drank carbohydrate drinks 2 hours before surgery (n = 6). These values corresponded to the estimated GV/kg of grade 0, and one of the previous study by Perlas et al, respectively. Therefore, the values of CSA in RLDP of 3.6 ± 1.0 for Perlas grade 0 stomach, and 5.6 ± 1.4 for Perlas grade 1 stomach were incorporated into the sample size estimation.15 Given the Δ of 2.8, we calculated the sample size to test the null hypothesis H0: μ1μ0 > 2.8 versus the alternative hypothesis H1: μ1μ0 ≤ 2.8, where, μ1μ0 is the mean difference of CSA value between the NO-NPO group and the NPO group (noninferiority test). A total of 32 participants in each group, 64 in total, were included, based on 80% power and α = .05 significance level, considering a 5% drop-out rate.

Data are presented as mean ± standard deviation, median (interquartile range), or numbers (%) as appropriate. The difference between the randomized groups on baseline variables was assessed using standardized mean difference (SMD). We assessed the effect of NPO versus NO-NPO on the primary outcome, CSA in RLDP, using a 2-sample t test. Noninferiority was claimed when the upper limit of the 2-sided 95% confidence interval was below the noninferiority margin. CSA in RLDP, GV, GV/kg, Perlas grade, GV/kg > 1.5 mL/kg, IQS in supine, and RLDP were presented as point estimate and corresponding 95% confidence interval (CI). We assessed the treatment effect on normally distributed continuous variables using the 2-sample t test, and Perlas grade and IQS using the Wilcoxon rank-sum test. Randomized groups were compared on the incidence of GV/kg > 1.5 mL/kg using a χ2 test. We assessed the interrater reliability of CSA in supine, CSA in RLDP, GV, and GV/kg, which were measured by 2 different examiners, using the intraclass correlation coefficient (ICC): 2-way random-effect model, single measurement type.20 The ICC was assessed by the classification by Landis and Koch,21 and ICC ≥0.7 was interpreted to have strong agreement between the 2 examiners. All statistical analyses were performed with Statistical Package for the Social Sciences (SPSS) v 20.0 software (SPSS, Inc, Chicago, IL).


Seventy-five patients were assessed for eligibility from July 2019 to February 2020. Eleven patients were excluded owing to the following reasons: 1 patient did not meet the inclusion criteria and 10 patients declined to participate in this study. Sixty-four individuals were enrolled and randomly allocated to the NPO or the NO-NPO group. All individuals followed fasting instructions and showed no adverse side effects. Sixty-four individuals were included in the statistical analysis (Figure 2).

Figure 2.:
CONSORT diagram of the study. CONSORT indicates CONsolidated Standards Of Reporting Trials; NPO, nil per os.

The baseline characteristics of the study population are presented in Table 1. Fasting time for solid food was 15 hours for both groups. Fasting time for liquids was 12 hours for the NPO group and 2 hours for the NO-NPO group. Gastric ultrasound measurements between the 2 examiners were not different with respect to gastric CSA, GV, and GV/kg (Supplemental Digital Content 1, Table 1, In each participant, Perlas grades observed by the 2 examiners were in concordance.

Table 1. - Baseline Characteristics of the Patients
NPO (n = 32) NO-NPO (n = 32) Standardized mean differencea
Age, y 38 ± 10 40 ± 12 −0.181
Height, cm 162.6 ± 5.6 160.0 ± 5.1 0.485
Weight, kg 56.9 ± 7.4 58.1 ± 8.7 −0.149
BMI, kg/m2 21.7 ± 2.9 22.6 ± 3.0 −0.305
ASA physical status, I/II 25 (78.1)/7 (21.9) 27 (84.4)/5 (15.6) 0.162
Smoking 0 (0) 0 (0) -
Fasting hours for solids, h 15 (13–18) 15 (13–17) 0.118
Fasting hours for liquids, h 12 (10–15) 2 (2–2) 4.045
Type of surgery
 Adnexal surgery 19 (59.4) 17 (53.1) 0.127
 Myomectomy 7 (21.9) 8 (25.0) −0.073
 Hysterectomy 6 (18.8) 7 (21.9) −0.077
Data are presented as mean ± standard deviation, median (interquartile range), or number (%).
Abbreviations: ASA, American Society of Anesthesiologists; BMI, body mass index; NPO, nil per os.
aImbalance defined as standardized mean difference larger than ±0.2.

The main measurements of gastric ultrasound are presented in Table 2. The ultrasound images were all interpretable in both groups, represented by all IQSs >3. The difference in means (95% CI) of CSA in RLDP between the 2 study groups was 0.04 (−1.56 to 1.64). The upper confidence limit of the difference in means was below the noninferiority margin of 2.8 (Figure 3).

Table 2. - Gastric Antral CSA, Perlas Grade, Gastric Volume, and Incidence of Risk Stomach Assessed by Gastric Ultrasound
NPO (n = 32) NO-NPO (n = 32) P value Difference (95% CI)
Primary outcome
 CSA, RLDP, cm2 6.25 ± 3.79 6.21 ± 2.48 .959a,b 0.04 (−1.56 to 1.64)c
Secondary outcomes
 CSA, supine, cm2 4.17 ± 2.34 4.28 ± 1.23 .828b −1.08 (−1.04 to 0.82)
 GV, mL 70 ± 56 66 ± 36 .756b 3.66 (−20 to 27)
 GV/kg, mL/kg 1.25 ± 1.00 1.17 ± 0.67 .694b 0.08 (−0.34 to 0.51)
 Perlas grade 1 (0–1) 0.5 (0–1) .871d 0 (0-0)
 GV/kg > 1.5 mL/kg 10 (31.3) 7 (21.9) .768e 3.1 (−17.65 to 23.85)
 IQS, supine 4 (3–5) 4 (3–5) .495d 0 (−1 to 0)
 IQS, RLDP 4 (4–5) 4 (4–4) .538d 0 (0-0)
Data are presented as mean ± standard deviation, median (interquartile range), or number (%).
Abbreviations: CI, confidence interval; CSA, cross-sectional area; GV, gastric volume; IQS, image quality score; NPO, nil per os; RLDP, right lateral decubitus position.
aP value for superiority.
bData are compared using student t test.
cUpper confidence interval limit is less than the noninferiority delta of 2.8 cm2, so noninferiority is claimed.
dData are compared using Wilcoxon rank-sum test.
eData are compared using χ2 test.

Figure 3.:
Noninferiority diagram of the mean difference of gastric antral cross-sectional area in right lateral decubitus position, between the NPO group and the NO-NPO group. The dashed line represents the noninferiority margin (Δ). The error bars indicate the 95% CI of the cross-sectional area difference (NO-NPO group − NPO group). The diagram describes noninferiority (equivalence) between the 2 groups. CI indicates confidence interval; NPO, nil per os.

The degrees of preoperative hunger, thirst, and anxiety are compared in Supplemental Digital Content 2, Table 2, Hunger and thirst score were lower in the NPO group than in the NO-NPO group. As shown in the Supplemental Digital Content 3, Table 3,, the gas out time and the time able to walk were shorter in the NO-NPO group than in the NPO group.


In this randomized controlled trial, we conducted the study of gastric ultrasound to evaluate gastric emptying of preoperative carbohydrate drink ingested up to 2 hours before anesthesia, compared to midnight fasting. All gastric ultrasound images were interpretable and we found no statistical difference in gastric antral CSA, GV, GV/kg, Perlas grade, and incidence of risk stomach between the 2 fasting strategies. This is consistent with prior studies showing that drinking carbohydrate 2 hours before anesthesia induction can be safely performed without increasing the risk of pulmonary aspiration.10,11 We also demonstrated that gastric ultrasound, which can be performed at bedside without risk, may be useful for the detection of gastric emptying and the reduction of the risk of pulmonary aspiration and fasting time.

Gastric emptying of carbohydrate beverages has been previously investigated with various methods, including GV assessment with naso- or orogastric tube drainage, gamma camera with radiotracer, and dye dilution.10–12,22 However, these methods are rather invasive, time-consuming, complicated, or not feasible at bedside. Moreover, naso- or orogastric tube insertion is uncomfortable for awake patients. Ultrasonography is used to evaluate gastric content.18 Previous studies have demonstrated that gastric ultrasonography provides useful information for patients undergoing surgery to determine GV and “risk stomach,” which may cause pulmonary aspiration, by observing solid particles or large gastric fluid volume.18,23 Ultrasonography is a noninvasive, simple, and safe tool without risk of exposure to radiation. Furthermore, ultrasonographic gastric assessment can be easily learned and applied by anesthesiologists with relatively high reliability.24,25 Therefore, assessment of gastric emptying of carbohydrate drink with gastric ultrasound has been spotlighted by many researchers.26–29 In 1 study, the investigators compared gastric CSA after ingestion of 400 mL of carbohydrate drink and 400 mL of 10% glucose solution and described that carbohydrate drink was emptied after 60 minutes of ingestion in healthy volunteers.26 Wang et al27 conducted gastric ultrasound in patients scheduled for the endoscopic submucosal dissection under general anesthesia. The study showed that while Perlas grade was increased in the carbohydrate group (710 mL at midnight followed by 355 mL 2 hours before operation) compared to the midnight NPO group, the suctioned residual gastric fluid was similar in both groups.27 In the observational studies, the investigators demonstrated that carbohydrate drinks were emptied after 2 hours of ingestion with gastric ultrasound in patients undergoing surgeries under general anesthesia.28,29 However, in these observational studies, they compared the gastric CSAs, before and after ingestion of carbohydrate drinks, within the same subjects. Our study is a randomized controlled study that compares the gastric CSA between the NPO and the NO-NPO group. We aimed to show that ingestion of carbohydrate drinks before anesthesia would not increase gastric contents compared to midnight fasting by performing a noninferiority test. Our results demonstrated that there were no differences in the gastric CSA and the semiquantitative 3-point grades in 2 different fasting strategies, suggesting that gastric emptying of carbohydrate drinks is not delayed in NO-NPO, compared to NPO patients, consistent with previous studies.10–12,22

We found that the reference CSA value measured after sufficient fasting time was not completely consistent between different studies. For example, in the present study, the CSAs of grade 0 stomachs were 4.40 ± 1.26 cm2 in the NPO group and 4.81 ± 1.07 cm2 in the NO-NPO group, consistent with a previous study, demonstrating that CSA of 4 cm2 in RLDP corresponds to the empty stomach.30 However, CSA was 3.1 cm2 for grade 0 stomach in gravid women scheduled for cesarean delivery in another study.31 Furthermore, Perlas et al15 reported that CSA in RLDP was 3.6 ± 1.0 cm2 for Perlas grade 0. In another study, median CSA after fasting was much, 2.60 cm2.32 We believe that these differences may be due to different characteristics between study cohorts, such as sex distribution, height, weight, and race. Therefore, it is not recommended to use a specific cutoff value to discriminate between nonfasting and fasting conditions.30,32 We think that discrimination between grade 2 and grade 0 would be preferred to predict risk stomach.

In this study, there showed trends toward statistical significance for hunger, thirst, and anxiety, which are consistent with the previous studies.10,27,33 However, because this study was not powered to assess hunger, thirst, and anxiety, compartment of these variables might be likely to result in a type II error. Furthermore, the following factors might cause to underestimate the benefits of carbohydrate drink on hunger, thirst, and anxiety. First, the patients in both groups had their last meal in the evening before the day of surgery. Therefore, fasting time for solid food was approximately 15 hours, which is comparable between the 2 groups. Moreover, as surgery started at 11:00 a.m. (data not shown), the fasting time included sleeping, during which humans are fasting physiologically. Similarly, the fasting time for liquids (12 hours) in the NPO group included night time sleeping. Moreover, we administered intramuscular glycopyrrolate as premedication according to our hospital policy. Since most patients felt dryness of mouth, and expressed this as “thirst,” we assume that this would result in higher-than-expected scores for thirst in the NO-NPO group.

There are some limitations in our study. First, baseline gastric ultrasound assessment was not performed, because NO-NPO patients were maintained in nonfasting conditions at all times. However, we believe that this would not affect our results, because single gastric assessment may also effectively estimate gastric emptying.32 Second, GV was calculated using the formula developed from a prior study performed in a different patient population. Therefore, GV may be different from the actual GV in our patients. For example, we could not use the formula by Bouvet et al,23 because GV values calculated according to that formula were negative. Therefore, care should be taken to use the definite GV values from our results. Third, this study was conducted in a small, limited female population, within normal range of body mass indexes, and without comorbidities that might delay gastric emptying. Therefore, our results are not generalizable to broader population such as men, patients with comorbidities that affects gastric emptying, morbid obesity, and patients who had had foregut surgeries. Therefore, this study needs to be validated in various patient populations with more complex underlying diseases in the further studies. Fourth, the timing when the patients in the carbohydrate group ingested carbohydrate drink was not recorded in our study. In the present study, the patients were asked hourly to freely decide whether to drink 100–200 mL of carbohydrate or not. However, the residual GV might be affected by the amount and the timing of carbohydrate drink. Therefore, care should be taken when applying our results in cases where different carbohydrate regimens are used.

In conclusion, our study suggests that carbohydrate drinks ingested up to 2 hours before anesthetic induction do not delay gastric emptying compared to midnight fasting, as evaluated by gastric ultrasound. Hence, gastric ultrasound is a useful tool which can assess gastric emptying when minimal fasting time is preferred.


Name: Eun-Ah Cho, PhD.

Contribution: This author helped create the study design, collect the data, and write the manuscript.

Name: Jin Huh, PhD.

Contribution: This author helped with the study idea, study design, interpretation of the data, and writing the manuscript.

Name: Sung Hyun Lee, MD, PhD.

Contribution: This author helped collect, interpret, and analyze the data.

Name: Kyoung-Ho Ryu, PhD.

Contribution: This author helped design the study, collect the data, and analyze the data.

Name: Jae-Geum Shim, MD.

Contribution: This author helped create the study design, collect the data, and write the manuscript.

Name: Yun-Byeong Cha, MD.

Contribution: This author helped collect the data and write the manuscript.

Name: Mi Sung Kim, PhD.

Contribution: This author helped interpret and analyze the data, write and correct the manuscript.

Name: Taejong Song, PhD.

Contribution: This author helped with the study idea, study design, data collection, and manuscript writing.

This manuscript was handled by: Tong J. Gan, MD.


    1. Breuer JP, Bosse G, Seifert S, et al. Pre-operative fasting: a nationwide survey of German anaesthesia departments. Acta Anaesthesiol Scand. 2010;54:313–320.
    2. Nygren J, Thorell A, Ljungqvist O. Preoperative oral carbohydrate nutrition: an update. Curr Opin Clin Nutr Metab Care. 2001;4:255–259.
    3. Carli F. Physiologic considerations of Enhanced Recovery After Surgery (ERAS) programs: implications of the stress response. Can J Anaesth. 2015;62:110–119.
    4. Nygren J, Soop M, Thorell A, Sree Nair K, Ljungqvist O. Preoperative oral carbohydrates and postoperative insulin resistance. Clin Nutr. 1999;18:117–120.
    5. Svanfeldt M, Thorell A, Hausel J, et al. Randomized clinical trial of the effect of preoperative oral carbohydrate treatment on postoperative whole-body protein and glucose kinetics. Br J Surg. 2007;94:1342–1350.
    6. Breuer JP, von Dossow V, von Heymann C, et al. Preoperative oral carbohydrate administration to ASA III-IV patients undergoing elective cardiac surgery. Anesth Analg. 2006;103:1099–1108.
    7. Tudor-Drobjewski BA, Marhofer P, Kimberger O, Huber WD, Roth G, Triffterer L. Randomised controlled trial comparing preoperative carbohydrate loading with standard fasting in paediatric anaesthesia. Br J Anaesth. 2018;121:656–661.
    8. Singh BN, Dahiya D, Bagaria D, et al. Effects of preoperative carbohydrates drinks on immediate postoperative outcome after day care laparoscopic cholecystectomy. Surg Endosc. 2015;29:3267–3272.
    9. Bilku DK, Dennison AR, Hall TC, Metcalfe MS, Garcea G. Role of preoperative carbohydrate loading: a systematic review. Ann R Coll Surg Engl. 2014;96:15–22.
    10. Nygren J, Thorell A, Jacobsson H, et al. Preoperative gastric emptying. Effects of anxiety and oral carbohydrate administration. Ann Surg. 1995;222:728–734.
    11. Yagci G, Can MF, Ozturk E, et al. Effects of preoperative carbohydrate loading on glucose metabolism and gastric contents in patients undergoing moderate surgery: a randomized, controlled trial. Nutrition. 2008;24:212–216.
    12. Hausel J, Nygren J, Lagerkranser M, et al. A carbohydrate-rich drink reduces preoperative discomfort in elective surgery patients. Anesth Analg. 2001;93:1344–1350.
    13. Arzola C, Perlas A, Siddiqui NT, Downey K, Ye XY, Carvalho JCA. Gastric ultrasound in the third trimester of pregnancy: a randomised controlled trial to develop a predictive model of volume assessment. Anaesthesia. 2018;73:295–303.
    14. Alakkad H, Kruisselbrink R, Chin KJ, et al. Point-of-care ultrasound defines gastric content and changes the anesthetic management of elective surgical patients who have not followed fasting instructions: a prospective case series. Can J Anaesth. 2015;62:1188–1195.
    15. Perlas A, Davis L, Khan M, Mitsakakis N, Chan VW. Gastric sonography in the fasted surgical patient: a prospective descriptive study. Anesth Analg. 2011;113:93–97.
    16. Cho EA, Kim MS, Cha YB, Lee MS, Song T. Evaluation of gastric emptying time of a rice-based meal using serial sonography. Biomed Res Int. 2019;2019:5917085.
    17. Perlas A, Mitsakakis N, Liu L, et al. Validation of a mathematical model for ultrasound assessment of gastric volume by gastroscopic examination. Anesth Analg. 2013;116:357–363.
    18. Van de Putte P, Perlas A. Ultrasound assessment of gastric content and volume. Br J Anaesth. 2014;113:12–22.
    19. Dang C, Dickman E, Tessaro MO, et al. Does oral radiocontrast affect image quality of abdominal sonography? Am J Emerg Med. 2018;36:684–686.
    20. Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater reliability. Psychol Bull. 1979;86:420–428.
    21. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33:159–174.
    22. Kaska M, Grosmanová T, Havel E, et al. The impact and safety of preoperative oral or intravenous carbohydrate administration versus fasting in colorectal surgery–a randomized controlled trial. Wien Klin Wochenschr. 2010;122:23–30.
    23. Bouvet L, Mazoit JX, Chassard D, Allaouchiche B, Boselli E, Benhamou D. Clinical assessment of the ultrasonographic measurement of antral area for estimating preoperative gastric content and volume. Anesthesiology. 2011;114:1086–1092.
    24. Kruisselbrink R, Arzola C, Endersby R, Tse C, Chan V, Perlas A. Intra- and interrater reliability of ultrasound assessment of gastric volume. Anesthesiology. 2014;121:46–51.
    25. Arzola C, Carvalho JC, Cubillos J, Ye XY, Perlas A. Anesthesiologists’ learning curves for bedside qualitative ultrasound assessment of gastric content: a cohort study. Can J Anaesth. 2013;60:771–779.
    26. Jian W-L, Zhang Y-L, Xu J-M, et al. Effects of a carbohydrate loading on gastric emptying and fasting discomfort: an ultrasonography study. Int J Clin Exp Med. 2017;10:788–794.
    27. Wang Y, Zhu Z, Li H, et al. Effects of preoperative oral carbohydrates on patients undergoing ESD surgery under general anesthesia: a randomized control study. Medicine (Baltimore). 2019;98:e15669.
    28. Popivanov P, Irwin R, Walsh M, Leonard M, Tan T. Gastric emptying of carbohydrate drinks in term parturients before elective caesarean delivery: an observational study. Int J Obstet Anesth. 2020;41:29–34.
    29. Song IK, Kim HJ, Lee JH, Kim EH, Kim JT, Kim HS. Ultrasound assessment of gastric volume in children after drinking carbohydrate-containing fluids. Br J Anaesth. 2016;116:513–517.
    30. Perlas A, Chan VW, Lupu CM, Mitsakakis N, Hanbidge A. Ultrasound assessment of gastric content and volume. Anesthesiology. 2009;111:82–89.
    31. Arzola C, Perlas A, Siddiqui NT, Carvalho JC. Bedside gastric ultrasonography in term pregnant women before elective cesarean delivery: a prospective cohort study. Anesth Analg. 2015;121:752–758.
    32. Bouvet L, Miquel A, Chassard D, Boselli E, Allaouchiche B, Benhamou D. Could a single standardized ultrasonographic measurement of antral area be of interest for assessing gastric contents? A preliminary report. Eur J Anaesthesiol. 2009;26:1015–1019.
    33. Zhang Y, Min J. Preoperative carbohydrate loading in gynecological patients undergoing combined spinal and epidural anesthesia. J Invest Surg. 2020;33:587–595.

    Supplemental Digital Content

    Copyright © 2021 International Anesthesia Research Society