A longer duration of inflammatory bowel disease (IBD), particularly ulcerative colitis (UC), is associated with a higher risk of developing colorectal cancer.¹ In one meta-analysis,² the pooled standardized incidence ratio of colorectal cancer in patients with UC was 2.4. Accordingly, guidelines have emphasized the importance of periodic colonoscopic surveillance in UC.³ Furthermore, the Selecting Therapeutic Targets in Inflammatory Bowel Disease program recommends regular endoscopic monitoring to evaluate disease activity even in a clinically remitted state.⁴ Therefore, accurate and high-quality monitoring is crucial for patients with IBD, particularly those with UC. Colonoscopy is the most significant modality for diagnosing and evaluating IBD.⁵ However, patients with IBD have a negative perception of colonoscopy and preparation, lowering adherence to surveillance.^6,7 In addition, patient anxiety decreases the tolerance for colonoscopy and preparation.⁸ Therefore, there is a high possibility of incomplete bowel preparation, which may negatively affect the quality of colonoscopy.

Polyethylene glycol (PEG) is an effective and safe agent that has been widely used in medicine. However, PEG for bowel preparation requires the ingestion of a 4-L solution.⁹ Therefore, low-volume preparations have been developed, and recent studies have shown their comparable efficacy, safety, and compliance with 4-L PEG.^10,11 Another alternative is a 2-L PEG with ascorbic acid solution (PEG+Asc). Studies have shown that 2-L PEG+Asc has a similar or superior effect than 4-L PEG. The lower volume and better taste also increase patient satisfaction and tolerance.¹² Moreover, a 1-L oral sulfate solution (OSS) has been found to have a similar effect to 2-L PEG+Asc or 4-L PEG. Importantly, OSS showed superior safety and compliance compared to 4-L PEG.^13,14 However, the usefulness of OSS preparations in patients with UC has not been clarified.

Therefore, this study aimed to compare the efficacy, safety, tolerability, and acceptability of OSS to 2-L PEG+Asc for bowel preparation in patients with inactive UC.

1. Study design and ethics

This was a prospective, randomized, multicenter, single-blind clinical trial conducted at six tertiary referral hospitals in Korea. The study was reviewed and approved by the institutional research ethics board of each hospital including St. Vincent’s Hospital (IRB number: VC17OEDI0054). The research was carried out in accordance with the Declaration of Helsinki. Written informed consent was obtained from all the participants.

2. Study population

Consecutive outpatients diagnosed with clinically stable UC between September 2017 and October 2019 were recruited. The inclusion criteria were as follows: (1) age >19 years; (2) diagnosed with UC based on clinical, endoscopic, radiographic, laboratory, and pathologic findings; and (3) inactive disease, that is, they had not required additional medications or changes to their medication dosage for the previous 1 year.^15,16 The exclusion criteria were: (1) suspected gastrointestinal obstruction or perforation; (2) previous gastrointestinal resection; and (3) severely compromised medical statuses, such as congestive heart failure of New York Heart Association grade III or IV, severe liver cirrhosis (Child-Pugh score C), and renal failure (creatinine clearance <30 mL/min).

3. Randomization and blinding

At the screening visit, participants were randomly assigned using computer-generated lists of numbers to the OSS or 2-L PEG+Asc group by research nurses who were not involved in the colonoscopy examination. The patients were unblinded; however, colonoscopists and assistant nurses who participated in the procedure were blinded to the treatment arms. To ensure blinding of the investigators to the allocated treatments, the colonoscopists and participants were instructed not to discuss the bowel cleansing methods before or during the procedure.

4. Bowel preparation and colonoscopy protocol

Participants were educated by research nurses about the study medications and their diets. Participants in both groups were instructed to follow a low-residue diet for 3 days prior to colonoscopy and a clear liquid diet for lunch (before 2:00 p.m.) on the day before examination. A split-dose regimen was used for both treatment groups, and participants were instructed to take half the allocated study medication in the evening (at 8:00 p.m.) the day before examination and to take the remaining half early in the morning of the day of examination, ensuring that the ingestion was complete at least 3 hours before examination. In the OSS group, each administration involved one 6-oz (177 mL) bottle of OSS (SUCLEAR^Ⓡ; Pharmbio Korea Co., Ltd., Seoul, Korea) diluted with water in a 16-oz (473 mL) cup followed by two additional 16-oz cups of water over the next 2 hours. Meanwhile, in the 2-L PEG+Asc group, each administration involved 1 L of PEG + Asc solution (250 mL every 15 minutes) followed by at least 500 mL of clear fluid. Two liters of PEG (HAPREP^Ⓡ; Pharmbio Korea Co., Ltd.) divided into 250 mL were administered every 15 minutes to participants in the PEG group.

All participants in both groups were instructed to take 10 mL of simethicone solution (ENDOCOL^Ⓡ; Pharmbio Korea Co., Ltd.) with the last bottle of preparation solution to remove foam from the colonic mucosa. Colonoscopies were scheduled in the morning (09:00 a.m. to 12:30 p.m.) to reduce procedure time-related bias. Conscious sedation using midazolam and pethidine hydrochloride was allowed based on participant preference. The Boston Bowel Preparation Scale (BBPS) score was assessed by colonoscopists who had performed at least 1,000 colonoscopies and who fully understood the scale. All colonoscopies were performed using a high-definition colonoscope.

5. Efficacy of bowel preparation and quality of colonoscopy

The primary endpoint was the proportion of successful bowel preparations in each treatment group. Bowel cleaning was evaluated using the BBPS, after removing the retained fluid and residual debris during the procedure, in three segments (right-side colon, transverse colon, and left-side colon) and given a score of 0 (solid stools) to 3 (no residual stool or mucus). Successful bowel preparation was defined as a score of ≥6. The secondary endpoints included the overall BBPS scores and BBPS scores of each segment. In addition, some quality indicators such as cecal intubation, adenoma detection rate, and polyp detection rate were assessed.

6. Assessment of safety, tolerability, and acceptance

The secondary endpoints included safety, tolerability, and acceptance, which were evaluated on the day of the colonoscopy before the procedure by research nurses who were not involved in the examination. The nurse interviewed each patient about their experience using a standardized questionnaire. All assessments were performed using scoring systems from previous bowel cleansing studies.^17,18

1) Safety

Vital signs were assessed, a complete physical examination was performed, and adverse events were evaluated immediately before examination by direct questioning. Additionally, new symptoms and exacerbations of pre-existing symptoms occurring after treatment (except those included in the evaluation of tolerability) were recorded.

2) Tolerability

Tolerability was assessed based on gastrointestinal symptoms. Patients reported on the occurrence and severity of nausea, bloating, and abdominal pain/cramps. A 4-point ordinal scale (1=no distress; 2=mild distress; 3=moderate distress; 4=severe distress) was used to score tolerability.¹⁷

3) Acceptance

The ease of taking the solution was graded using a 4-point ordinal scale. Willingness to repeat the bowel preparation type was evaluated. Compliance was scored on a 3-grade scale according to the percentage of solution consumed (excellent, intake of the whole solution; good, intake of at least 75% of the solution; poor, intake of <75%).¹⁸

7. Assessment of disease activity and laboratory data

Clinical disease activity of UC was assessed using the partial Mayo score. The score was evaluated on the day of enrollment and within 2 to 4 weeks of follow-up after the colonoscopy to determine the effect of bowel preparation on disease status. The results of blood tests, including hematology and blood chemistry, performed before enrollment and on the day of colonoscopy were collected. Mayo endoscopic subscore is assessed during the colonoscopy for evaluating mucosal change.

8. Statistical analysis

Non-inferiority was defined as a one-sided 97.5% confidence interval if the difference in successful cleansing rate between the two treatment groups was greater than −15.0%. The sample size of 88 patients for each group was determined, assuming success rates of 80% for colon cleansing in both groups, a 15.0% non-inferiority margin, and a significance level of 0.05 with 80% statistical power. The success rate for colon cleansing was based on a previous study.¹⁹ Considering a 10% drop-out rate, we aimed to recruit 196 patients. Continuous variables were presented as the mean±standard deviation, whereas categorical variables were expressed as totals and percentages. Continuous variables were analyzed using the t-test and Wilcoxon rank-sum test, and univariate analyses were performed using the chi-square test. p<0.05 was considered statistically significant. All statistical analyses were performed using IBM SPSS Statistics version 20 (IBM Corp., Armonk, NY, USA).

1. Patient characteristics

A total of 199 patients were enrolled and randomized into the OSS and 2-L PEG+Asc groups (99 and 100 patients, respectively). Among them, 92 and 93 patients in the OSS and 2-L PEG+Asc groups, respectively, completed the study and were evaluated (Fig. 1). The only significant between-group difference in demographic characteristics was patient sex, with the 2-L PEG+Asc group including significantly more men (61% vs 47%, p=0.03). Approximately 85% of the patients were in clinical remission. The patient characteristics are summarized in Table 1.

Table 1 . Baseline Characteristics of the Patients in the Study.

Characteristic	OSS (n=92)	2-L PEG+Asc (n=93)	p-value
Male sex	43 (47)	57 (61)	0.03
Age, yr	47.9±14.7	48.9±15.0	0.53
Body mass index, kg/m2	23.3±3.0	23.6±3.1	0.54
Clinical activity			0.88
Remission (partial Mayo score ≤1)	78 (85)	80 (86)
Mild activity (score 2-3)	14 (15)	13 (14)
Extension			0.29
E1	33 (36)	33 (35)
E2	37 (40)	29 (31)
E3	22 (24)	31 (34)
Current treatment			0.24
No medication	0	3 (3)
5-ASA only	66 (72)	65 (70)
Immunomodulators	15 (16)	19 (20)
Biologics	11 (12)	6 (7)
Other medical conditions
Diabetes mellitus	6 (7)	6 (6)	0.74
Parkinsonism	2 (2)	0	0.15
Other medication
Prokinetics	1 (1)	3 (3)	0.32
Anticholinergics	0	1 (1)	0.32
Anti-constipated drug	2 (2)	2 (2)	0.99
Probiotics	20 (22)	23 (25)	0.76
Tricyclic antidepressants	0	1 (1)	0.32
Previous abdominal surgery*	19 (21)	14 (15)	0.21
Disease duration, yr	7.0±6.4	7.6±6.6	0.53
Interval since last colonoscopy, yr	2.6±1.6	2.4±1.3	0.37
Sedative colonoscopy	84 (91)	88 (95)	0.27

Data are presented as number (%) or mean±SD..

OSS, oral sulfate solution; PEG+Asc, polyethylene glycol with ascorbic acid; 5-ASA, 5-aminosalicylic acid..

*Gastrointestinal surgery (bowel resection, appendectomy, cholecystectomy), obstetric and gynecologic surgery (caesarean section, uterine myomectomy, hysterectomy), urologic surgery (nephrectomy, tumorectomy)..

Figure 1. Flowchart of the study subjects.
OSS, oral sulfate solution; PEG+Asc, polyethylene glycol with ascorbic acid.

2. Quality of bowel preparation

No significant difference was found in the rate of successful preparation (OSS [96.7%] vs 2-L PEG+Asc [97.8%], p=0.64). The mean BBPS scores in the OSS and 2-L PEG+Asc groups were 8.3±1.1 and 8.1±1.1 (p=0.33), respectively. All examinations were performed using cecal intubation. No significant differences were identified in the mean Mayo endoscopic subscore, adenoma detection rate, or polyp detection rate (Table 2).

Table 2 . Efficacy of Bowel Preparation and Quality Indicators According to Both Preparation (OSS vs 2-L PEG+Asc).

Variable	OSS (n=92)	2-L PEG+Asc (n=93)	p-value
Successful preparation (BBPS ≥6)	89 (96.7)	91 (97.8)	0.64
BBPS (total)	8.3±1.1	8.1±1.1	0.33
Right-side colon	2.6±0.5	2.5±0.5	0.62
Transverse colon	2.9±0.4	2.8±0.5	0.17
Left-side colon	2.8±0.4	2.8±0.4	0.62
Mayo endoscopic subscore	1.1±1.0	1.0±0.9	0.52
Cecal intubation, yes	92 (100)	93 (100)	0.94
Intubation time, min	4.5±3.0	4.3±2.9	0.52
Retrieval time, min	9.4±3.1	10.5±5.0	0.06
Adenoma detection rate	11.6 (11)	6.5 (6)	0.20
Polyp detection rate	26.1 (24)	18.3 (17)	0.20

Data are presented as number (%), mean±SD, or % (number)..

OSS, oral sulfate solution; PEG+Asc, polyethylene glycol with ascorbic acid; BBPS, Boston Bowel Preparation Scale..

3. Safety, tolerability, and acceptance

No significant between-group difference was observed in the mean blood pressure and pulse rate immediately before colonoscopy. The rates of symptom occurrence, such as those for headache, dizziness, chills, or epigastric discomfort, were not significantly different (Table 3).

Table 3 . Safety, Tolerability, and Acceptance According to the Preparation (OSS vs 2-L PEG+Asc).

Variable	OSS (n=92)	2-L PEG+Asc (n=93)	p-value
Safety
Systolic blood pressure, mm Hg	122.7±14.0	125.4±16.0	0.24
Diastolic blood pressure, mm Hg	74.7±11.1	75.2±10.7	0.80
Pulse rate, beats/min	82.3±14.1	79.1±15.4	0.16
Newly developed symptom
Headache	4 (4)	4 (4)	0.98
Dizziness	1 (1)	4 (4)	0.10
Chilling	2 (2)	3 (3)	0.66
Epigastric discomfort	1 (1)	0	0.31
Tolerability
Nausea*	1.8±0.8	1.6±0.8	0.06
No or mild	79 (86)	80 (86)	0.98
Bloating*	1.5±0.8	1.4±0.7	0.59
No or mild	83 (90)	83 (89)	0.65
Abdominal pain/cramps*	1.2±0.5	1.2±0.5	0.70
No or mild	88 (96)	88 (95)	0.75
Acceptance
Ease of taking the solution*	1.5±0.7	1.6±0.8	0.41
No or mild	84 (91)	81 (87)	0.36
Willingness to repeat	78 (85)	81 (87)	0.41
Compliance			0.51
Excellent	91 (99)	90 (97)
Fair: intake of at least 75%	1 (1)	2 (2)
Poor: intake of <75%	0	1 (1)

Data are presented as mean±SD or number (%)..

OSS, oral sulfate solution; PEG+Asc, polyethylene glycol with ascorbic acid..

*4-Point ordinal scale (1, no distress; 2, mild distress; 3, moderate distress; 4, severe distress)..

The numeric rating scale scores for tolerability and acceptance are shown in Table 3. No significant differences were observed in tolerability (nausea, bloating, and abdominal pain) and acceptance (ease of taking the solution, willingness to repeat, and compliance) between the two groups. The mean nausea score tended to be higher in the OSS group than in the 2-L PEG+Asc group (1.8±0.8 vs 1.6±0.8, p=0.06). However, the proportion of tolerant patients to the total number of patients with nausea was not significantly different between the OSS and 2-L PEG+Asc groups (86% [79/92] vs 86% [80/93], p=0.98). In total, 99% (91/92) and 97% (90/93) of the patients in the OSS and 2-L PEG+Asc groups, respectively, ingested the entire solution (p=0.51). The rate of willingness to repeat was not significantly different between the groups (85% [78/92] vs 87% [81/93], p=0.41).

4. Disease activity and laboratory findings before and after colonoscopy

The mean partial Mayo scores were 0.6±0.9 and 0.6±1.0 before colonoscopy in OSS and 2-L PEG+Asc groups, respectively, and there were no significant changes after colonoscopy (0.8±1.4, p=0.42 and 0.6±1.0, p=0.94, respectively). No significant change was found in the mean laboratory data in both groups, except for chloride in the OSS group. There was a significant change in the mean serum chloride level in the OSS group (103.7±2.6 mEq/L vs 104.7±3.1 mEq/L, p=0.03); however, the values remained within the normal range (Table 4).

Table 4 . Activity Index and Laboratory Data before and after Colonoscopy in the OSS and 2-L PEG+Asc Groups.

Variable	OSS (n=92)				2-L PEG+Asc (n=93)
	Pre	Post	p-value		Pre	Post	p-value
Activity index
Partial Mayo score	0.6±0.9	0.8±1.4	0.42		0.6±1.0	0.6±1.0	0.94
Laboratory findings
Sodium, mEq/L	139.8±2.5	140.2±2.5	0.34		139.8±2.0	140±2.1	0.60
Potassium, mEq/L	4.2±0.4	4.2±0.4	0.30		4.2±0.4	4.2±0.4	0.83
Chloride, mEq/L	103.7±2.6	104.7±3.1	0.03		105.4±3.9	104.5±2.4	0.08
Magnesium, mg/dL	2.1±0.1	2.1±0.4	0.08		2.15±0.2	2.0±0.2	0.48
Calcium, mg/dL	9.5±0.6	9.5±0.6	0.76		9.5±0.6	9.4±0.6	0.63
Phosphorus, mg/dL	3.5±0.6	3.5±0.6	0.79		3.4±0.7	3.4±0.6	0.77
Urea nitrogen, mg/dL	12.5±3.4	12.7±3.5	0.78		12.6±4.1	12.7±4.0	0.98
Creatinine, mg/dL	0.9±0.3	0.9±0.3	0.63		0.9±0.3	12.6±3.9	0.43
Osmolarity, mOsm/kg	282.1±9.0	283.7±7.9	0.28		282.3±8.7	283.1±7.6	0.54
Hemoglobin, g/dL	14.3±1.6	14.0±1.6	0.35		14.2±1.8	14.1±1.7	0.81
White blood cell, 109/L	6.0±1.9	6.3±2.2	0.35		6.0±1.4	6.3±1.8	0.12
Platelet, ×109/L	266.4±60.6	253.3±56.7	0.13		259.4±64.9	264.7±82.7	0.63
C-reactive protein, mg/dL	0.1±0.6	0.2±0.7	0.47		0.2±0.7	0.2±0.7	0.76

Data are presented as mean±SD..

OSS, oral sulfate solution; PEG+Asc, polyethylene glycol with ascorbic acid..

Data on the efficacy, safety, tolerability, and acceptability of OSS for bowel preparation in patients with UC are rare. In this study, the rate of successful bowel preparation in the OSS group was non-inferior to that of the 2-L PEG+Asc group (96.7% vs 97.8%, respectively). This result was similar to that of previous studies on the general population. In the first report that compared OSS and 2-L PEG, the rate of successful bowel preparation with OSS was 97.2%.²⁰ Furthermore, several subsequent studies showed satisfactorily high rates of successful bowel preparation with OSS.^13,14,21-24 Although the scoring system for evaluating bowel preparation differed among previous studies, the success rate was more than 80% in most studies. The mean BBPS in this study was high (≥8 points), and no significant between-group differences were noted in the scores according to the involved section (right-side colon, transverse colon, and left-side colon). This showed that OSS is non-inferior to 2-L PEG with respect to successful bowel cleansing.

This was the first randomized controlled study using OSS for bowel preparation in patients with UC. There are concerns that hyperosmotic agents such as OSS would induce mucosal changes in UC. Indeed, some studies suggested that sodium phosphate- or sodium picosulfate-based preparations may induce or aggravate mucosal inflammation.^25,26 However, it seemed to have little effect in this study. The mean Mayo endoscopic subscore checked during the test was as low as 1 in both groups, and no significant difference was found. In addition, clinical activity (partial Mayo score) and laboratory data before and after the test showed no clinically significant differences between the groups. A previous study on patients who underwent screening colonoscopy showed that mucosal changes, including erythema and aphthous lesions, were not observed in the OSS group, confirming its safety.²³ In terms of safety, in addition to mucosal changes, vital signs obtained immediately before the colonoscopy were normal in both groups and did not show any differences. Collectively, these results support that OSS may be safely used in patients with clinically inactive UC. However, large-scale investigations are still required to more effectively ensure the safety of OSS in UC.

Several studies have compared pre-colonoscopy bowel preparation in patients with IBD. An Italian study published in 2015 compared 4-L PEG and 2-L PEG+bisacodyl in patients with UC,¹⁹ and a Korean study published in 2017 compared 4-L PEG and 2-L PEG+Asc in patients with UC.¹⁶ Both studies found that 2-L PEG is more acceptable with respect to willingness to repeat. These results were supported by a recent retrospective French study which found that 2-L PEG and sodium picosulfate have better efficacy and tolerability than 4-L PEG.²⁷ However, to the best of our knowledge, there have been no published data on OSS. Our findings provide a novel direction for bowel preparation in patients with UC.

To investigate the tolerability of preparations, we compared the incidence of solution-related gastrointestinal symptoms. The majority of patients (>85%) in the OSS group had no or mild symptomatic discomfort, and the mean symptom score was not significantly different between the groups. In other studies, the symptom score for vomiting was significantly higher with OSS than with 2-L PEG+Asc,²² and the incidence of nausea was significantly higher than that with sodium picosulfate with magnesium citrate.^21,24 However, these studies further concluded that the severity of symptoms in the OSS group was not clinically significant and was mild.

With respect to acceptance, no significant between-group differences were identified in the ease of taking the solution, willingness to repeat, or drug compliance. This could be because, although the volume of the OSS dose was approximately 1 L less than that of 2 L PEG+Asc, the total amount of liquid to be ingested was still approximately 3 L as these preparations require ingestion of 2 L and 1 L of water, respectively. This result is consistent with that of previous studies that found no significant difference in satisfaction between OSS and 2-L PEG.^22,23 Meanwhile, studies that compared OSS and 4-L PEG found significantly superior satisfaction with OSS.^14,28 Therefore, differences may be observed if OSS is compared with 4-L PEG in patients with UC. A recent study in Italy suggested that severe symptoms during preparation are associated with the female sex (odds ratio, 3.64; 95% confidence interval, 1.94 to 6.83).²⁹ In our study, significantly more females were allocated to the OSS group, and this may also affect our results showing no between-group differences in tolerability and acceptance.

The present study had several strengths. First, this was a prospective, randomized, controlled study limited to homogeneous subjects who had stable disease activity of UC, increasing the reliability of the results. Underlying conditions that could affect bowel cleansing were controlled through randomization. No significant difference was noted in disease, drug use, and abdominal surgery, which could affect bowel motility, between the two groups. Second, the time from preparation to colonoscopy was strictly controlled. At the time of enrollment, the unblinded researchers explained how to take the drug with a split dose, and all patients were examined by the blinded researchers in the morning, controlling for the change in the degree of bowel preparation over time. Third, colonoscopies were performed by colonoscopists with at least 10 years of experience. The cecal intubation rate was 100%, the mean insertion time was approximately 4.5 minutes, and the mean withdrawal time was approximately 10 minutes. Fourth, the quality of bowel preparation was objectively assessed using the BBPS, which is a validated scoring system for preparation evaluation and is widely used in similar studies.

However, this study had some limitations. First, we could not investigate the efficacy of bowel preparations under various conditions of UC. Given that the study exclusively enrolled clinically inactive patients with maintenance therapy to minimize the mucosal effects of bowel preparations, the results may have limited generalizability to the overall UC population. Another limitation was the single-blinded design of the study, which may have introduced a bias. However, the researchers ensured that the colonoscopists were blinded to the bowel preparation method.

In conclusion, OSS is an effective bowel preparation for colonoscopy and has comparable tolerability to 2-L PEG+Asc. In addition, OSS does not affect disease activity. Thus, OSS can be used safely for pre-colonoscopy bowel preparation in patients with clinically inactive UC.

This study was an investigator-initiated study funded partly by Pharmbio Korea Co., Ltd., Seoul, Korea.

This study was an investigator-initiated study funded partly by Pharmbio Korea Co., Ltd., Seoul, Korea. Except for that, no potential conflict of interest relevant to this article was reported.

Study concept and design: J.M.L., K.M.L. Data acquisition: J.M.L., K.M.L., H.S.K., J.S.K., H.S.L., S.H.J., J.H.K., D.B.K. Data analysis and interpretation: J.M.L., K.M.L., H.S.K., J.S.K., H.S.L., S.H.J., J.H.K., D.B.K. Drafting of the manuscript: J.M.L. Critical revision of the manuscript for important intellectual content: J.M.L., K.M.L. Statistical analysis: J.M.L. Obtained funding: J.M.L., K.M.L. Administrative, technical, or material support; study supervision: K.M.L. Approval of final manuscript: all authors.