1.Study design
This study was a randomized, controlled, single-blind, prospective trial comparing two dietary regimens prior to colonoscopy: a 1-day low-residue diet(LRD) vs. a 3-day LRD the day before colonoscopy. The study protocol was approved by the Ethics Committee of the third Hospital of Hebei Medical University, and was registered at Chinese Clinical Trial Registry(ChiCTR1900025843).
2.Subjects:
A cohort of patients who underwent colonoscopy from April to August 2019 at the Third Hospital of Hebei Medical University,in Shijiazhuang city,Hebei province was selected for our work. All recruited patients signed an appropriate informed consent to be allowed to participate in this study. Both experimental protocol and informed consent were submitted to the Ethics Committee Formal enrollment and study were conducted solely upon approval.The following inclusion criteria were utilized: (i) patients’ age should be between 18 and 80 years old, (ii) patients should be ready for colonoscopy, and (iii) patients should be undergoing colorectal cancer screening or having non-specific gastrointestinal symptoms.The following exclusion criteria were utilized: (i) patients who were unable to cooperate with the examination; (ii) patients who underwent gastrointestinal surgery; (iii) patients with severe diseases in heart, brain, kidney, liver and other organs; (iv) patients who could not tolerate the colonoscopy procedure; (v) patients with contraindications to colonoscopy, including bowel obstruction or perforation, severe intestinal infection or toxic megacolon, consciousness disorder and unstable vital signs; (vi) patients whose colonoscopy failed to reach ileocecum due to reasons other than bowel preparation (such as intestinal stricture caused by tumor); (vii) pregnant and lactating patients; (viii) patients who suffered a stroke or dementia; (ix) patients with anxiety or depressive psychosis; and (x) PEG allergic patients.A total of 240 patients were recruited. Upon signing an informed consent, eligible patients were randomly divided into two groups, according to the period of supplementation with a low-residue diet (one or three-day treatment).Each subject obtained a sealed opaque envelope whose contents were determined by a random number table prepared by an independent researcher. Patients were allowed to eat rice porridge, noodles, steamed bread, bread, tofu, egg soup, chicken, some peeled, seeded fruits and cooked vegetablesbut devoid of untreated vegetables, fruits and coarse grains.
- Bowel preparation schedules:
Patients were instructed to adopt a low residue diet one day or three days before colonoscopy, according to their respective grouping, and then fasted for about 11 hours (10pm to 9am on the day of the procedure).Patients were not allowed to discuss grouping in front of the enteroscopy operators. Eventual issues were directly treated with nursing personnel. Control patients were administered with three liters of PEG electrolyte solution betwenn 9-11am on the day of colonoscopy, followed by oral administration of 30 ml simethicone. Colonoscopy was performed between 3-5pm on the same day.Endoscopic diagnosis and treatment were conducted using an Olympus CF H-290 colonoscope. Endoscopists were unaware of the patient grouping.Colonoscopy experience in at least 1,000 cases was used as a threshold to select the professionals involved in this procedure.
- Outcome indicators:
The following measures were taken in consideration for our dietary evaluation:
(I) Patient profiling (age, gender, weight, education level, constipation history, colonoscopy history, and abdominal symptoms);
(II) Colonoscopy profiling (percentage of arrival at the cecum, intubation time, withdrawal time, and the detection rate of colon polyps).
(III) Bowel-cleansing quality score: assessed according to Boston bowel preparation scale (BBPS) [11]. For this, the BBPS divides the colon into three regions: the right side (cecum and ascending colon), the transverse section (including the hepatic flexure and splenic flexure), and the left side (descending colon, sigmoid colon, and rectum). A 4-point classification scale was utilized as previously reported [12]. The three colon regions were scored accordingly, and the quality of bowel preparation was assessed by its total score (i.e. 4-point scale multiplied by 3). These total scores were mainly divided into 5 grades: excellent (8-9), good (6-7), medium (4-5), poor (2-3) and extremely poor (0-1). Grades over 4 (medium) were considered as qualified for further analyses [12].
(IV) Patient tolerance and satisfaction: patients were submitted to a questionnaire, before the examination, on the day of colonoscopy. According to the digital analogue scale method, the research group designed a hunger-comfort scale [1], which evaluated the hunger-comfort condition of the patients, where “1” indicated comfort and “10” indicated hunger. The diet acceptance of the patients before colonoscopy was graded in 5 distinct levels: very simple acceptance, simple acceptance, tolerable, unacceptable, and difficult to accept [1]. The willingness to repeat the diet protocol for bowel preparation was evaluated accordingly [13]. Eventual occurrence of adverse reactions, such as nausea, vomiting, abdominal distension and discomfort, was carefully annotated.
- Sample size calculation
The primary focus of this study was to evaluate the quality of bowel preparation by BBPS scoring (ranging from 0 to 9). Our previous studies in healthy and with abdominal symptoms polulation have showed that related values are well spread across the 10-point with a mean of 6.5 points, and standard deviation of 1.6 points. We considered a 1-point average difference between groups minimally clinically important. With an s.d. of 1.6 and 42 subjects per arm, the independent t-test would provide 80% power to detect a between group difference of 1 point at a two-sided a of 0.05. Assuming a dropout rate of 20% ,we thus aimed at enrolling 51 subjects in each arm to achieve an effective sample size of at least 42 per group.
- Statistical analysis:
SPSS 19.0 software (SPSS Inc., Chicago, Illinois, USA) was used for our statistical data analysis. Data measurement was represented by mean plus/minus standard deviation. Homogeneity of variance was analyzed and, if the variance was equal, a t-test was applied for pairwise comparison between independent samples. If variance was unequal, Kruskal Wallis test was used, and the difference was statistically significant when P<0.05. χ2test was used for data counting, and statistical significance was set at P<0.05.