Evaluation of parameters influencing the quality of colon preparation with a split-dose regimen of sulfate salts

Introduction

Currently, colorectal cancer (CRC) is the third most common and the second most deadly cancer [1,2]. The burden of CRC may increase in coming years, considering the rising incidence of early onset colorectal cancer [3]. Primary prevention is the principal strategy to reduce CRC [4]. Currently, either colonoscopy or noninvasive methods, such as the fecal immunochemical test, are used for CRC screening.

Bowel preparation is a key performance indicator for the quality of colonoscopy [5] and is closely related to the efficacy of the examination [6]. Inadequate bowel preparation is associated with a lower adenoma detection rate [7,8], longer lasting examinations and more adverse events. Even nowadays, inadequate bowel preparation still occurs in approximately 10-20% of cases [9].

Polyethylene glycol (PEG)-based solutions have been widely used for decades as bowel-cleansing agents, with good efficacy and without serious complications. However, the patients may encounter various difficulties during preparation intake, leading to a suboptimal result. One of the major barriers towards widely accepted colonoscopy is the large volume of conventional (PEG) preparation solutions (4 L) and their poor taste. These factors often cause patients to either postpone or avoid colonoscopy, or to be less compliant in completing their preparation, resulting in a poor outcome.

Oral sulfate salts (OSS) represent a valid alternative for colonoscopy preparation with well-documented efficacy and safety. Several randomized controlled trials have compared OSS to PEG solutions, revealing comparable efficacy. Most of the studies have been conducted in Asia (Korea [10-12], Japan [13]) and the USA [14]. Although data from Europe [15] are rather limited, the European Society of Gastrointestinal Endoscopy include OSS in the recommended preparation regimens [16].

Factors that influence bowel preparation have been studied mainly with PEG solutions, while studies that specifically examine these factors with OSS are limited. The tolerance profile of OSS solutions has been shown to be better than that of PEG-based solutions [17].

Based on our department’s experience, we have been empirically administering OSS for colonoscopy preparation for the last 5 years, because we observed a better tolerance and compliance compared with the PEG solutions administered previously. The aim of our study was to evaluate the efficacy of OSS preparation and to prospectively investigate factors that may influence bowel cleaning in a split-dose regimen.

Patients and methods

In this non-interventional single-center study we prospectively collected data from patients receiving OSS before colonoscopy for different indications in routine clinical practice. The study was registered in ClinicalTrials.gov under the identifier NCT05107505 and was approved by the hospital ethics committee.

Patients were consecutively recruited from the gastroenterology department of Alexandra General Hospital in Athens from July 1, 2021, to June 30, 2022. Written informed consent was obtained from the patient or their legally acceptable representative before their entry into the study.

Results

A total of 446 patients were initially recruited for the study. Eleven were excluded (2 patients had incomplete preparation intake of the preparation solution, 7 patients had a total BBPS<6, and 2 patients had a partial BBPS<2).

Of the 446 patients who completed the laxative intake, 435 (97.5%) achieved successful (optimal or good) preparation. Among the 435 patients included in the analysis, 257 (59.08%) were female. Mean age was 61.98±11.58 years and median body mass index (BMI) was 26.1 kg/m² (IQR 23.8-29.4 kg/m²). Their indication for colonoscopy was colorectal cancer screening: 155 (35.63%), post-polypectomy surveillance: 102 (23.45%), or symptom investigation: 178 (40.92%). In addition, 172 (39.54%) patients had a history of abdominal surgery, while 115 (26.44%) had undergone digestive surgery. Patients’ baseline characteristics are shown in Table 1.

Table 1 Descriptive characteristics of 435 patients who underwent colonoscopy. Values are mean±standard deviation or n (%)

The recommended diet was followed by 421 (96.78%) patients, while 379 (87.13%) patients used bisacodyl (10 mg). In our cohort, the median liquid intake was 3 L (IQR 2-3.5), with 183 (42.26%) consuming <3 L, 138 (31.87%) =3 L, and 112 (25.87%) >3 L. The median time between the end of the second laxative dose and colonoscopy initiation was 5:15 h (IQR 4:30-6:00, min 2:20, max 12:20).

Cecal intubation was achieved in 425 (97.7%) of the cases. Senior endoscopists conducted colonoscopies in 204 (46.9%) patients. Bubbles were found in 173 (39.77%) patients (Table 2).

Table 2 Clinical features of the colonic preparation and colonoscopy for 435 patients who underwent colonoscopy. Values are n (%) or median (interquartile range)

Optimal colon preparation (BBPS=9) prior to the colonoscopy was achieved in 279 (64.14%) patients. The cleansing level was excellent (segmental BBPS=3) in 289 (66.44%) patients in the ascending colon, 346 (79.54%) in the transverse and 387 (88.97%) in the descending colon, indicating more frequent suboptimal preparation for the ascending colon compared to the transverse and the left colon (P<0.001) (Table 3). Concerning the administration of OSS preparation, we observed that optimal preparation decreases according to time, for every hour from the end of the second dose until the initiation of colonoscopy. Moreover, almost 2/3 of the patients maintained their optimal preparation until the 6^th h after the end of the second laxative dose (Fig. 1; for details see Supplementary Table 1).

Table 3 Segmental and overall proportions (95%CI) of BBPS for the 435 patients who underwent colonoscopy. Optimal score was defined as grade=3 for each colon segment, and 9 for the entire colon. Suboptimal segmental score was defined as grade=2 while for the entire colon 6, 7 or 8

Figure 1 Time-related effects on optimal bowel preparation in 435 patients undergoing colonoscopy: LOWESS regression

Multivariate analysis revealed that higher BMI (odds ratio [OR] 1.05, 95% confidence interval [CI] 1-1.1) and a longer time between the second laxative dose and initiation of the examination (OR 1.25, 95%CI 1.08-1.45) were associated with a higher percentage of suboptimal bowel preparation. For every extra hour between the second laxative dose and the initiation of colonoscopy, a 25% increase (95%CI 8-45%) in the odds of having suboptimal bowel preparation is expected. These results were adjusted for age (for details see Supplementary Tables 2 and 3).

The abovementioned influence of time is reflected in the delays of later appointments by the increase in the time between the end of the second laxative dose and the exam initiation for the third [5:30 hh: mm (IQR 4:45-6:30) P=0.001], the fourth [5:45 hh: mm (IQR 5:15-7:00), P<0.001], and fifth [5:30 hh: mm (IQR 4:30-6:30), P=0.003] scheduled examinations compared to the first [4:50 hh: mm (IQR 4:25-5:30)]. The second [5:00 hh: mm (IQR 4:20-5:30, P=0.38)] and paradoxically the sixth [5:15 hh: mm (IQR 4:30-6:00, P=0.14) appointments did not differ to the first appointment concerning the delays (Table 4).

Table 4 Time between second laxative dose and colonoscopy initiation according to the appointment order (h: mm) for 435 patients who underwent colonoscopy

The overall adenoma detection rate (ADR) for the entire colon was 48.74 (95%CI 43.95-53.54). The ADR for the ascending colon was 34.94 (95%CI 30.46-39.63) and for the rest of the colon it was 29.89 (95%CI 25.62-34.43). Age was the only statistically significant predictor for the total ADR and for the ADR in each segment of the colon. The level of bowel cleanliness did not affect the overall ADR, although an indication was detected for a BBPS score of 7 (OR 1.71, 95%CI 0.94-3.09, P=0.08; Table 5) (for details see Supplementary Tables 4 and 5).

Table 5 Logistic regression for the effect of BBPS on ADR for each segment of the colon. Data are adjusted for age

A total of 61 patients (14,02%) reported minor adverse events. The most frequent adverse event was nausea, which was experienced by 8.96% (39/435) of the patients. Vomiting was reported by 3.21% (14/435) of the patients, mainly during the second dose of the OSS. Also reported were abdominal pain in 1.15% (n=5), proctalgia in 0.46% (n=2) and dizziness in 0.46% (n=2).

Discussion

In this study, using a prospective collection of real-world data we observed that OSS is a safe and effective bowel preparation regimen for patients undergoing colonoscopy. Our findings suggest that split-dose OSS, when correctly administered, can significantly improve the quality of bowel preparation, with a successful result (total BBSP≥6 and partial BBSP≥2) in most patients (98%). However, aiming at the optimum result (BBPS=9), the efficacy of OSS was negatively influenced by the patient’s BMI and the timing of the second dose intake. These results reflect in a more realistic manner the everyday practice of many participants.

According to Voiosu et al [15], selecting the appropriate preparation is essential for achieving the optimal level of cleansing. Medication selection may be more important than other patient-related factors. This is probably related to compliance, which depends to a great extent on the taste and the volume of the preparation regimen needed. Several studies have shown that sulfate solutions have similar efficacy to PEG-based solutions [13,14,17,19]. Factors that influence bowel preparation have been studied with PEG solutions, while equivalent studies for OSS are scarce. These studies with PEG solutions have found that patient-related factors, such as age or diabetes mellitus, can influence bowel preparation (Zad et al [20]]. In another study, Cheng et al found that male sex and obesity were also significant factors related to inadequate preparation with PEG [21].

The benefits of a split-dose regimen for PEG preparations are well established [22,23]. Nowadays, it is considered the cornerstone of an efficient and effective bowel preparation for colonoscopy [24]. Studies examining split-dose administration of OSS come to the same conclusions, though there are far fewer data compared to PEG preparations. Palma et al have shown that split-dose is superior to a same-day administration of OSS preparation for achieving excellent bowel preparation cleaning (63.3% vs. 44.6%) [18]. Rex et al [14] demonstrated the superiority of split-dose OSS over PEG-based solutions administered the day before. With the OSS split dose, excellent preparation was achieved in 71.4%, compared to only 34.4% with PEG. Regula et al [15] found in 1281 patients that excellent bowel preparation with OSS was achieved in a total of 43.6%, 74.2% of whom had received a split dose. However, it should be noted that there was heterogeneity among the centers regarding the administration of split-dose preparation, with the Netherlands having the highest level of split-dose administration at 92.4%, compared to the Czech Republic at 39.8%. On the basis of the abovementioned data, we chose to use a split-dose regimen for all our patients over the last 5 years.

Bubbles in the colon lumen during colonoscopy were found in 39.77% of patients, a percentage that is comparable to the incidence of bubbles in studies using PEG solutions. Several studies have shown that simethicone can reduce the formation of bubbles and improve bowel preparation quality [25,26]. We dealt with the bubbles issue using a water pump, with or without the additional use of simethicone, and had very good results, although its use was not systematically recorded.

In a recent European study, Theunissen et al [27] demonstrated that the BBPS was higher for standard preparation combined with bisacodyl, whereas the use of bisacodyl was associated with higher patient discomfort. Ischemic colitis has been described as a rare complication of bisacodyl used as an adjunct in bowel preparation, both with PEG solutions and with OSS [28,29]. In our study, 87.13% of the patients used bisacodyl (10 mg) 2 days before colonoscopy without reporting any complications.

The time interval between the second dose of the preparation and colonoscopy is crucial for successful preparation. According to the European Society of Gastrointestinal Endoscopy’s guidelines [16], the second dose of the preparation in a split-dose regimen should be started at least 5 h before the colonoscopy and completed at least 2 h before colonoscopy.

A meta-analysis, including 29 randomized trials involving 4040 patients with same-day evening-before preparation, and 3679 with split-dose preparation, demonstrated that the benefits of split-dose preparation were evident if the 5-h rule was observed, if the time from preparation to colonoscopy was less than 5 h, and especially when colonoscopy was performed within 3 h after the completion of the second dose [30].

In our study, higher BMI (OR 1.05, 95%CI 1-1.1) and a greater time between the second laxative dose and examination (OR 1.25, 95%CI 1.08-1.45) were correlated with a higher rate of suboptimal bowel preparation.

Five hours after the completion of the second dose of OSS, almost 70% of patients had excellent bowel preparation (Table 4). Furthermore, our study found that, even in patients with long delays exceeding 7 h, and despite the fact that many suboptimal preparations were observed, as predicted by our model, very few unacceptable preparations were recorded and excluded from the analysis. This may be attributed to the use of the water pump. It is to be noted that our study included only morning colonoscopies. We examined the indirect effect of the scheduled time of colonoscopy and found a trend of “poorer” preparation for patients who had the 3^rd-5^th slot in the program. This trend probably represents accumulated delays in everyday practice.

An overall ADR of 48.7% was achieved in our cohort, while segmental ADR was 34.94% in the ascending colon and 29.89% in the left colon. Bowel cleanliness, i.e., optimal vs. suboptimal bowel preparation, did not affect overall or segmental ADR, even though there was an indication for level 7 of BBPS (OR 1.71, 95%CI 0.94-3.09, P=0.08). In a recent meta-analysis of randomized controlled trials by Chen et al [31], the OSS preparation regimen had a higher ADR compared with PEG regimens (44.60% vs. 38.14%). Regarding the influence of bowel preparation on ADR, it has been reported by Calderwood et al [32] that good bowel preparation is better than excellent, since the ADR was higher for BBPS 6-8 than for BBPS 9.

Several adverse events have been previously reported with OSS, such as abdominal distension, abdominal pain, nausea, vomiting and headache. According to a large real-world study by Anastassopoulos et al [29], 299,417 patients were evaluated in European centers, for different colonoscopy preparation regimens, among which 33,465 received OSS. The study found that OSS was safer than non-OSS regimens for bowel preparation. In our study, no major complications were reported, while neither electrolyte abnormalities nor hospitalizations were observed. In the study conducted by Regula et al [15] in Europe, nausea was the most frequent adverse event, more frequently in women. In our study, the most frequent adverse event was nausea, which was reported in 9% of the total population (62.9% among those reporting complications).

In some studies diabetes negatively affects bowel preparation [33,34]. In our study we did not find any correlation between diabetes and suboptimal preparation. However, it should be stressed that we did not calculate the duration of diabetes mellitus, or the level of HbA1c, and we did not record the exact medications taken by the patients or whether patients were receiving insulin or not. While our study provides valuable insights into the efficacy and safety of OSS for bowel preparation before colonoscopy, there are several limitations that should be taken into consideration.

The study was conducted in a single center, which limits the generalizability of our findings. However, with our preparation, using diet and split doses of OSS, almost all patients had good acceptability, and with very few exceptions global BBPS was ≥6 and partial BBPS≥2, indicating a rather well-accepted and efficacious regimen. Although we did not use a validated score to assess patient satisfaction, especially in those with previous PEG experience, no major complaints were reported. This was a non-comparative study, meaning that we did not compare the efficacy and safety of OSS to other bowel preparation agents or non-split-dose regimens, since this was beyond the scope of our study’s aims. Finally, no laboratory values were recorded in our study, which limits our ability to draw conclusions about the impact of OSS on various laboratory parameters. However, it should be noted that no major clinical side-effects were observed, which may suggest that no significant laboratory disturbances were likely to have occurred.

In conclusion, this study shows that OSS offers a safe and convenient alternative to PEG solutions for bowel preparation. BMI may influence the quality of bowel cleaning but the main parameter, which depends on the schedule of the daily endoscopic schedule and may be taken into consideration, is the time between the end of the second dose of the split regimen and colonoscopy initiation. However, it is important to notice that proper bowel preparation is crucial for the quality of colonoscopy, and factors such as BMI and preparation time should be considered when using OSS.