Study design and population
The study design was randomized (1:1), placebo-controlled, single-blind, parallel-group, with gender-stratified sampling at a single center. Inclusion criteria were participants over the age of 18, scheduled for elective gastroscopy, and exclusion criteria were participants with symptoms suggestive of gastric outlet obstruction, scheduled to undergo colonoscopy in the same setting. background and incomplete gastroscopy.
Sample size was calculated using the two means formula for gastric residual volume and the paired difference formula for patient well-being. For objective 1, the sample size was calculated using the formula for comparing two means. The ratio of group A to group B was set at 1 with a standard deviation of 18.46 based on a previous study13. The mean difference was set at 12.5. Therefore, the sample size was 35 in each arm. For objective 2, the sample size was calculated using the pairwise difference comparison formula. The expected standard deviation of the paired differences was set at 2 times the expected mean of the paired differences. Therefore, the sample size was 34 in each arm. Type I error was set at 5% (two-sided), Type 2 error at 20% (to achieve 80% power), and dropout rate was set at 10%. The estimated sample size was 78 subjects with 39 subjects in each arm.
Randomization was performed using stratified permuted blocks, a computer-generated 1:1 allocation sequence stratified by sex. Random block sizes of 6 were used. The sealed envelope method was used to implement the allocation sequence, and the results were hidden from the principal investigator.
Study protocol
Patients scheduled for elective gastroscopy were first informed of the study protocol and recruited after informed consent. The recruitment period was 12 months, from April 2019 to March 2020. Participants were randomized to plain water (group A) or carbohydrate drink (group B). Group A participants received 250ml of drinking water while Group B received a Resource Pack® (Nestle Health Science, Malaysia) drink (237ml with 53.6g carbohydrates and 9g whey protein) mixed in water at the same volume of 250ml. They were asked to consume the drink within 10 minutes of being served. The gastroscopy procedure would be performed after 2 hrs on duty, and in our unit the test was performed under local oral anesthesia (10% lidocaine). After the procedure, participants were separated and prohibited from interactions with other study participants and endoscopists.
To minimize the effect on GRV before measurement, endoscopists were instructed not to use high-volume water rinses, but were allowed to rinse the lens in a limited manner if necessary. Once the IBC was documented, the endoscopists were allowed to resume normal water jet and lens rinsing for proper diagnostic examination. All residual fluid visible in the stomach was aspirated by direct visualization using the gastroscope. Then, the fluid was collected in a suction reservoir bottle (Fig. 1) to record the IBC.
The suction tank bottle for IBC measurement.
Data analysis
Data were analyzed using SPSS version 26 (SPSS Inc., Chicago, IL, USA). Numerical data were presented as the mean (standard deviation, SD) or median (interquartile range, IQR) depending on the distribution. For the comparison of gastric residual volume between arms A and B, the independent T-test was used. To compare participants’ well-being (hunger, thirst, anxiety, fatigue and weakness) between groups, the repeated measures ANCOVA test was used.
Ethical Approval, Trial Enrollment, and CONSORT Statement
Approval was obtained from the Human Research Ethics Board of Universiti Sains Malaysia (USM/JEPeM/19010082 dated 04/15/2019) that compliance with the Declaration of Helsinki, the guidelines of the International Conference on Harmonization (ICH), Good Clinical Practice (GCP) Standards, Council for International Organizations of Medical Sciences (CIOMS) Guidelines, Global Practices for Research and Investigation and Ethics Review, EC/IRB Standard Operating Procedures (SOPs) and local ethical review regulations and standards. This study was registered with Clinicaltrial.gov (NCT03948594 2019-05-14). The study protocol was carried out in accordance with the CONSORT statement and the checklist.
