PDF icon Download PDF
Endoscopy 2010; 42(8): 633-638
DOI: 10.1055/s-0029-1244236
Original article

© Georg Thieme Verlag KG Stuttgart · New York

A randomized single-blind trial of standard diet versus fiber-free diet with polyethylene glycol electrolyte solution for colonoscopy preparation

A.  M.  Soweid1 , A.  A.  Kobeissy1 , F.  R.  Jamali2 , M.  El-Tarchichi1 , A.  Skoury1 , H.  Abdul-Baki1 , L.  El-Zahabi1 , A.  El-Sayyed1 , K.  A.  Barada1 , A.  I.  Sharara1 , F.  Mourad1 , A.  Arabi1
  • 1Department of Internal Medicine, American University of Beirut Medical Center, Beirut, Lebanon
  • 2Department of Surgery, American University of Beirut Medical Center, Beirut, Lebanon
Further Information

A. SoweidMD 

American University of Beirut Medical Center

P.O. Box 11-0236
Riad El Solh 1107 2020
Beirut
Lebanon

Fax: +961-1-366098

Email: as25@aub.edu.lb

Publication History

submitted 19 November 2009

accepted after revision 28 April 2010

Publication Date:
09 July 2010 (online)

Also available ateRef
 PDF Download Permissions and Reprints
Table of Contents

Background and study aims: Colonoscopy preparation usually involves the intake of large volumes of polyethylene glycol electrolyte solution (PEG-ES) in combination with a clear-liquid diet (CLD). Liberalization of the diet might enhance the tolerance to PEG-ES without compromising the quality of the preparation. The primary aims of this study were to evaluate the efficacy and tolerability of PEG-ES given with a CLD compared with a fiber-free diet (FFD) for colonoscopy preparation. The incidence of adverse events among patients in the two diet groups was also assessed as a secondary outcome.

Methods: This was a single-center randomized, prospective, single-blind study. A total of 200 patients undergoing colonoscopy were randomized to either CLD or FFD in addition to PEG-ES.

Results: Patients in the FFD group were able to drink more PEG-ES (mean ± SD, 3.9 ± 0.3 L) compared with those in the CLD group (3.3 ± 0.7 L) (P < 0.01). The quality of the preparation was significantly better in the FFD group, with more patients having satisfactory preparations than those in the CLD group (81.4 % vs. 52.0 %; P < 0.001). Tolerance to the preparation was higher in the FFD group compared with the CLD group, with significantly more patients adhering to the FFD regimen (P < 0.001). There were more adverse events experienced in the CLD group, with odds ratios of 1.9 for nausea (95 % confidence interval [CI] 1.0 – 3.6), 3.8 for vomiting (95 % CI 1.3 – 11.3), and 3.0 for headache (95 % CI 1.5 – 5.9).

Conclusion: FFD given with PEG-ES on the day before colonoscopy is a more effective regimen than the standard CLD regimen, and is better tolerated by patients.

 

Introduction

Colonoscopy is a very commonly performed procedure worldwide. Proper colonic preparation is vital for adequate visualization of the entire colonic mucosa, as the detection of small polyps, neoplasms, and other mucosal abnormalities depends primarily upon the quality of bowel cleansing [1] [2] [3]. Inappropriate bowel preparation has been reported to occur in as many as 25 % of colonoscopies [1]. This may hinder the identification of small pre-cancerous colonic lesions and usually leads to a repeat examination being necessary with its attendant preparation discomfort, procedure risks, and costs [1] [2] [4].

Polyethylene glycol electrolyte solution (PEG-ES) had become the most commonly used agent for colonic cleansing since its introduction in 1980 [5]. It has a minimal effect on the intravascular volume and serum electrolytes compared with other agents such as sodium phosphate oral solution. This makes PEG-ES a safer choice of laxative for elderly patients or patients with cirrhosis, renal insufficiency, and heart failure [6] [7] [8]. However, patients usually complain of the large amount of PEG-ES that has to be ingested and the requirement for a strict clear-liquid diet (CLD) for at least 24 hours prior to the procedure; these requirements make patient compliance with bowel preparation even more problematic [9].

This study was designed to test the hypothesis that if patients had to adhere to less-severe dietary restrictions, their tolerance to and compliance with the preparation would improve. This may lead to improved quality of colon cleansing with a consequent reduction in failed and repeated procedures.

Many published trials have addressed cleansing ability, safety, and tolerability of the multiple purgative regimens; however, only a few studies have examined the effect of diet liberalization on the process of bowel preparation. The main objective of our prospective single-blind randomized trial was to compare the clinical efficacy and patient tolerability of oral PEG-ES given with two different diet regimens before colonoscopy. The incidence of adverse events among patients in the two diet groups was also assessed as a secondary outcome. The first group used the standard CLD on the day before colonoscopy. The diet for the second group consisted of a fiber-free diet (FFD) on the day before colonoscopy.

 

Patients and methods

This was a prospective single-blind, randomized study that was conducted at the American University of Beirut Medical Center (AUBMC). The study protocol was approved by the Institutional Review Board at AUBMC. All patients who agreed to participate in the study signed a written informed consent form.

Patients above the age of 18 years who were scheduled for colonoscopy for various indications were enrolled in the study. Exclusion criteria were severe congestive heart failure (New York Heart Association class IV), end-stage renal disease (estimated glomerular filtration rate < 10 mL/minute), severe metabolic abnormalities, drug abuse history, severe psychiatric diseases, history of prior colonoscopy, and allergy to PEG-ES.

A total of 200 consecutive patients who were scheduled to undergo colonoscopy were randomized to one of two methods of bowel preparation. Randomization was via computer-generated random numbers inside sealed opaque numbered envelopes that contained written instructions regarding the assigned bowel preparation. The first group of patients were instructed to follow the standard CLD preparation by drinking fluids only ([Table 1]).

Table 1 List of foods allowed.
Fiber-free diet* Clear-liquid diet†
Breakfast
30 g of cheese, or two eggs (fried/boiled) + œ cup of milk + Œ loaf of white bread + 1 tbsp olive oil or butter
Lunch
90 g of meat (beef, chicken or fish) + œ cup of cooked white rice + œ cup of ice cream + 2 tbsp olive oil
Dinner
30 g of cheese, or two eggs (fried/boiled) + 1 cup of jello + œ cup milk or yoghurt or pudding + 1 tbsp olive oil 		Water, broths, soft drinks, tea, coffee (without milk or cream), clear fruit juices, and jello
* Patients were allowed to add any item from the clear-liquid diet list any time during the day.
† No red-, purple-, or blue-colored fluids were allowed.

In the second group, patients were allowed to eat three meals of solid food ([Table 1]), which were designed by dieticians at AUBMC to contain negligible amounts of fiber (FFD). The type and intensity of instruction were the same for both groups. All participants were asked to drink 4 L of PEG-ES on the evening before the procedure. Each sachet of PEG-ES (Klean-Prep; Norgine Ltd., Uxbridge, UK) contained 59 g of macrogol 3350, 5.685 g anhydrous sodium sulfate, 1.685 g sodium bicarbonate, 1.465 g sodium chloride, and 0.7425 g potassium chloride. Each sachet was to be dissolved in exactly 1 L of water, with the total being 4 L of PEG-ES.

Immediately before colonoscopy, patients completed a questionnaire regarding their preparation experience with the help of a research fellow, who was blinded to the dietary randomization. Patients were asked about tolerability of the preparation, any associated adverse effects (including bloating, nausea, vomiting, abdominal cramps, fecal incontinence, and sleep disturbances), energy level, and willingness to repeat the preparation in the future.

Colonoscopies were performed under conscious sedation by experienced endoscopists who were blinded to the dietary randomization. At the end of the procedure, the endoscopists were asked to grade the overall quality and efficacy of bowel cleansing as “excellent”, “good”, “fair”, or “poor”, according to a previously described scale ([Table 2]) [10] [11]. The bowel assessment was stratified into “satisfactory” (excellent/good) and “unsatisfactory” (fair/poor) for the purpose of analysis.

Table 2 Colon-cleansing quality grading score by endoscopists.
Grade Description
Excellent No fecal matter or nearly none in the colon; small amounts of clear liquid
Good Small amounts of thin liquid fecal matter seen and suctioned easily, mainly distal to the splenic flexure; all mucosa seen
Fair Moderate amounts of thick liquid to semi-solid fecal matter seen and suctioned, including proximal to splenic flexure; small lesions may be missed; > 90 % of mucosa seen
Poor Large amounts of solid fecal matter found, precluding a satisfactory study; unacceptable preparation; < 90 % of mucosa seen
 

Statistical analysis

The chi-squared test or Fisher’s exact test was used to assess the relationship between categorical variables such as the presence or absence of a specific adverse event and categorical predictors. An independent t-test was used to assess the relationship between continuous variables and diet or other categorical variables. Multivariate analyses were then conducted by building logistic regression models. In the first model, the outcome was the quality of bowel preparation and the predictors were age, sex, diet, quantity of PEG-ES intake, and indication. In the second logistic regression model, the outcome was the adverse event and the independent variables were those found to have significant effect on that outcome in bivariate analyses. Models were built for each adverse event at a time. The models were built with and without interaction and the results of the best fit model were reported. All statistical analyses were performed using the statistical software SPSS 16.0 for Windows. Results were regarded as statistically significant with a two-tailed P-value of 0.05 or less.

 

Results

 

Patient characteristics

Patient characteristics are shown in [Table 3].

Table 3 Patient demographic data and indication for colonoscopy.
Parameters Diet regimen P-value
Clear liquid (n = 98) Fiber-free (n = 102)
Demographic data
 Male/female
 Age range, years
 Mean age (SD), years
55/43
18 – 82
55.5 (16.6)
50/52
18 – 80
56.6 (11.5)
NS
NS
NS
Indications for colonoscopy, n (%)
 Screening
 Change in bowel habits
 Abdominal pain
 Hematochezia
 Inflammatory bowel disease
 Anemia
41 (41.8)
13 (13.3)
17 (17.3)
14 (14.3)
8 (8.2)
5 (5.1)
62 (60.8)
14 (13.7)
10 (9.8)
8 (7.8)
5 (4.9)
3 (2.9)
NS
NS
NS
NS
NS
NS
NS, not significant.

A total of 200 patients (105 men [52.5 %]) who were scheduled for colonoscopy at AUBMC were included in the trial. The age of patients ranged from 18 to 82 years, with a mean of 56.1 ± 14.2 years.

Patients were randomized to either the CLD group (n = 98) or the FFD group (n = 102). No statistically significant difference was noted in terms of age, sex, and indication for colonoscopy between the two groups ([Table 3]).

 

Assessment of bowel preparation

Colon-cleansing quality was reported by the endoscopists as follows: excellent (n = 43; 21.5 %), good (n = 91; 45.5 %), fair (n = 46; 23.0 %), and poor (n = 20; 10.0 %). Thus, the bowel preparation was considered “satisfactory” in 134 patients (67 %) and “unsatisfactory” in 66 patients (33 %).

 

Bivariate analysis – Relationship between the quality of bowel preparation and potential determinants

Indications for colonoscopy were also divided into two groups: one group included all patients undergoing colonoscopy for screening and the other group included patients with other medical problems. The FFD group had 83 satisfactory preparations (81.4 %) compared with 51 (52.0 %) in the CLD group (P < 0.001). The consumption of PEG-ES was significantly greater in the satisfactory group compared with the unsatisfactory group (3.78 ± 0.46 vs. 3.28 ± 0.76; P < 0.001) ([Table 4]).

Table 4 Quality of bowel preparation.
Quality P-value
Satisfactory Unsatisfactory
Age, mean ± SD, years 56.7 ± 13.5 54.8 ± 15.5 NS
Sex, n (%)
 Male
 Female
65 (61.9)
69 (72.6)
40 (38.1)
26 (27.4) 		NS
Diet regimen, n (%)
 Fiber-free
 Clear liquid
83 (81.4)
51 (52.0)
19 (18.6)
47 (48.0) 		< 0.001
Quantity of PEG-ES, mean ± SD, L 3.78 ± 0.46 3.28 ± 0.76 < 0.001
Indication for colonoscopy, n (%)
 Screening
 Medical condition
75 (72.8)
59 (60.8)
28 (27.2)
38 (39.2) 		NS
NS, not significant; PEG-ES, polyethylene glycol electrolyte solution.

There was no significant effect of medical indication (screening vs. medical problems) on the outcome (satisfactory vs. unsatisfactory); 72.8 % of patients who underwent colonoscopy for screening had a satisfactory colonoscopy preparation compared with 60.8 % of those who underwent colonoscopy for other medical indications (P = 0.07).

 

Multivariate analysis

The type of diet and the quantity of PEG-ES intake remained the only independent predictors for the quality of bowel preparation, after adjustment for age, sex, and indication, with an adjusted odds ratio (OR) of 2.24 (95 % confidence interval [CI] 1.09 – 4.58) for the type of diet (FFD vs. CLD), and 2.93 (95 % CI 1.65 – 5.21) for the quantity of PEG-ES consumed.

 

Tolerability and adherence

All patients were instructed to drink 4 L of PEG-ES. A total of 131 patients were able to take the whole 4-L dose, 19 patients (9.5 %) drank 3.5 L, 28 (14.0 %) drank 3 L, eight (4.0 %) drank 2.5 L, and 14 (7.0 %) drank only 2 L. Patients who were assigned to the FFD group were able to drink more PEG-ES (3.9 ± 0.3 L) than patients in the CLD group (3.3 ± 0.7 L) (P < 0.01). No significant difference was noted in the quantity consumed between males and females (3.7 ± 0.6 vs. 3.6 ± 0.6; P = 0.3). The quantity consumed correlated significantly with a satisfactory colonoscopy preparation, independent of the diet type (P < 0.01).

Adherence to the designated preparation regimen was higher in the FFD group as reflected by the quantity of PEG-ES intake, the number of patients who were able to drink more than 3L, and by the ease of adherence to the instructed preparation. A total of 101 patients (99.0 %) in the FFD group drank more than 3L compared with 77 (78.6 %) in the CLD group (P < 0.001). In the FFD group, 83 patients (81.4 %) reported the ingestion of the solution to be easy compared with 61 (62.2 %) in the CLD group (P = 0.003). When asked about their willingness to take the preparation again, 84 patients (82.4 %) who followed the FFD were willing to repeat their experience compared with 66 patients (67.3 %) in the CLD group (P = 0.013). Additionally, significantly more episodes of weakness during the preparation were noted in the CLD group than in the FFD group.

 

Adverse experiences

No serious adverse events were encountered during or after the colonoscopy preparation process. The incidence of adverse events among patients in the two diet groups is shown in [Table 5].

Table 5 Adverse events.
Adverse event Diet regimen P-value
Clear liquid Fiber-free
n % n %
Nausea 43 43.9 31 30.4 0.048
Vomiting 15 15.3 5 4.9 0.014
Bloating 45 45.9 35 34.3 NS
Cramps 39 39.8 34 33.3 NS
Headache 35 35.7 17 16.7 < 0.01
Weakness 52 53.1 27 26.5 < 0.01
Sleeping difficulties 31 31.6 19 18.6 0.03
Hunger 51 52.0 22 21.6 < 0.01
Interference with daily life activities 16 17.6 10 10.1 NS
NS, not significant.
 

Bivariate analysis – Relationship between adverse events and sex/age

Female patients reported significantly more adverse events than male patients, in terms of nausea, vomiting, bloating, headache, and weakness (P < 0.05). Patients who developed nausea, headache, and hunger were younger than those who did not: nausea 51 ± 14 vs. 59 ± 4 years (P < 0.01); headache 52 ± 15 vs. 57 ± 12 (P = 0.02); and hunger 53 ± 16 vs. 58 ± 13 (P = 0.03).

There was no significant relationship between the incidence of adverse events and the indication for colonoscopy.

 

Bivariate analysis – Relationship between diet and sex/age

There was no significant difference in the sex distribution of each diet group. Also there was no significant difference in the mean age between the two diet groups ([Table 3]).

 

Bivariate analysis – Relationship between adverse events and diet

Patients in the FFD group reported significantly fewer adverse events in terms of nausea, vomiting, headache, weakness, sleeping difficulties, and hunger (P < 0.05) ([Table 5]). There was no significant difference between the two groups in bloating, cramps, or interference with daily activities.

 

Multivariate analysis

There was no significant interaction between sex and diet type on any adverse event. The CLD regimen was significantly associated with the incidence of adverse events, independent of age and sex. The adjusted odds ratio for nausea was 1.9 (95 % CI 1.0 – 3.6), 3.8 for vomiting (95 % CI 1.3 – 11.3), and 3.0 for headache (95 % CI 1.5 – 5.9). Age had a negative relationship with nausea, vomiting, headache, and hunger but not with bloating, cramps, weakness, sleeping difficulties, and interference with daily life activities. Females were more likely than males to report adverse events, except for hunger ([Table 6]).

Table 6 Logistic regression; outcome (adverse events) adjusted for age, sex, and diet.
Age Sex
OR (95 % CI) 		Diet
OR (95 % CI)
β (SE) P-value
Nausea – 0.039 (0.011) 0.001 0.374 (0.200 – 0.698) 1.929 (1.037 – 3.589)
Vomiting – 0.041 (0.017) 0.014 0.269 (0.092 – 0.786) 3.776 (1.264 – 11.28)
Bloating – 0.009 (0.01) 0.375 0.484 (0.27 – 0.868) 1.725 (0.962 – 3.094)
Cramps – 0.013 (0.01) 0.221 0.875 (0.489 – 1.564) 1.317 (0.736 – 2.356)
Headache – 0.025 (0.012) 0.031 0.447 (0.227 – 0.880) 2.979 (1.496 – 5.933)
Weakness – 0.008 (0.011) 0.458 0.503 (0.274 – 0.922) 3.387 (1.842 – 6.229)
Sleeping difficulties + 0.003 (0.012) 0.778 0.594 (0.308 – 1.144) 2.123 (1.093 – 4.121)
Hunger – 0.024 (0.011) 0.032 1.302 (0.701 – 2.418) 3.897 (2.086 – 7.280)
Interference with daily life activities – 0.019 (0.014) 0.174 0.988 (0.425 – 2.292) 1.815 (0.769 – 4.283)
β, beta estimate; CI, confidence interval; OR, odds ratio; SE, standard error.
 

Discussion

Colonoscopy remains the standard diagnostic and therapeutic tool for examining the large intestine. However, a successful and complete procedure depends mainly on a successful bowel preparation.

The large volume of PEG-ES (4 L) that is usually prescribed remains a major obstacle for completion of the bowel preparation because of the multiple associated adverse effects, including abdominal pain, cramping, bloating, nausea, and vomiting. Patients also have to stay on a diet of only clear liquids for at least 24 hours prior to colonoscopy [12] [13]. For these reasons, the whole cleansing process is extremely uncomfortable for most of the patients, resulting in decreased compliance and often to an incomplete preparation and a poorly cleansed colon. In order to decrease the adverse effects associated with ingesting such a large quantity of liquid (PEG-ES and CLD), different types and forms of preparation solutions have been proposed, with mixed results (e. g. fleet phosphosoda, a lower volume or a split regimen of PEG-ES, with or without adjuncts of flavors, magnesium sulfate, bisacodyl, cisapride, low-residue kits) [9] [13] [14] [15] [16] [17] [18]. Only a few published studies have addressed the impact of diet liberalization on the day before colonoscopy on the quality of bowel cleansing and patient tolerance. Raymer et al. conducted an open-label trial on L-glucose (24 g in 250 mL of water), as an osmotic laxative for bowel cleansing in healthy volunteers. He allowed them to eat a light lunch no later than 2 pm on the day before the procedure. Diet effect on colonoscopy was not the main focus of this study; however, it was shown that his formulated regimen had similar efficacy when compared with other colon-cleansing agents [19]. In another study by El-Sayed et al., colonic preparation with split-dose PEG-ES (2 L of PEG-ES, plus one tablet of bisacodyl the day before colonoscopy, and then 1 L on the same day) with minimal dietary restriction the day before, provided better-quality colon cleansing and higher compliance than the usual whole 3-L dose of the same solution along with the liquid diet [20]. Moreover, a split dose of PEG-ES (2 L the night before colonoscopy, followed by 2 L the morning of the procedure) with no dietary restrictions was shown by Aoun et al. to provide a better quality of bowel preparation than the whole 4-L dose regimen. This dietary modification had no significant impact on patient tolerability and side effects, except for bloating. Individuals in the split-dose regimen group in this study were more willing to repeat the preparation if offered the option [21]. In a prospective study conducted by Rapier and Houston [22], the use of pre-packaged low-residue diet kits given in combination with PEG or magnesium citrate and bisacodyl on the day before colonoscopy was associated with adequate colon cleansing and patient tolerability. Similarly, Delegge and Kaplan concluded from their trial that the use of a pre-packaged, low-fiber diet with a magnesium citrate/bisacodyl cathartic resulted in significantly better colon cleansing and patient tolerance as well as a willingness to repeat the cathartic preparation when compared with the standard sodium phosphate-CLD regimen [23]. Scott et al. studied the efficacy and tolerance of sodium phosphate solution after diet liberalization. He randomized 200 patients into two groups. The first received a light breakfast followed by clear fluids on the day before colonoscopy. The second group of patients had a normal breakfast and a low-residue lunch followed by clear fluids on the day before the procedure. No significant difference in clinical efficacy was noted between the two groups. Fewer cases of hunger were reported, and more patients were able to resume their usual daily activity in the latter group [24].

To our knowledge, the current study is the first randomized, controlled prospective clinical study that evaluates the effect of diet liberalization during bowel preparation using 4 L PEG-ES alone on the day before colonoscopy. Our patients were representative of the population that undergoes colonoscopy in terms of age, sex, and indications for colonoscopy.

We compared the effect of FFD to that of CLD when consumed in combination with 4 L of PEG-ES solution the day prior to colonoscopy. The quality of bowel cleansing appeared to be significantly better when patients were instructed to drink the solution along with the FFD. The quality of bowel preparation is directly related to the amount of purgative intake regardless of the type of diet. Although both groups were instructed to drink a maximum of 4 L of PEG-ES, more patients in the FFD were able to drink most of the recommended PEG-ES, which in turn improved the quality of the colon preparation. Patients in the FFD group showed a higher propensity to tolerate and adhere to the purgative as a single dose. They reported fewer side effects (nausea, vomiting, bloating, and hunger), and more energy, even after adjusting for sex and age. They also showed willingness to take the PEG-ES with the FFD. As previously demonstrated by Hale et al., the present study showed that women had a greater tendency to report adverse events than men. However, in contrast to the Hale study [25], which showed that increasing age was positively correlated with preparation-related symptoms, our study did not reveal this effect.

Our study suffers from the following limitations. The presence of other co-morbidities that can negatively impact on the colon-cleansing quality, such as diabetes, was not accounted for in our cohort of patients [26]. A detailed history of medication intake and disease conditions was not included in the questionnaire, both of which have been shown to correlate with symptom prevalence [25]. A larger sample size is required to clarify the effect of colonoscopy indications on the quality of bowel preparation. However, due to the randomization process, we feel these limitations should not affect the validity of the results. The exact time between completion of the preparation and the performance of the procedure, which is an important factor that can affect the quality of the bowel cleansing, was not considered in the present study, with patients undergoing colonoscopy the following day from any time between 7 : 30 and 13 : 00 [27] [28].

Few published studies have addressed the effect of liberalizing the diet in the process of bowel cleansing. The whole 4-L dose of PEG-ES was much better tolerated and provided better bowel preparation quality, along with improved patient acceptance, compliance, and adherence when consumed with the FFD during the day before colonoscopy. This would reduce the rate of repeat procedures, and would probably have a positive impact on overall cost reduction as well as patient tolerance and satisfaction. Therefore, we highly recommend this FFD regimen to be included in preparing patients for colonoscopy.

Competing interests: None

 

References

A. SoweidMD 

American University of Beirut Medical Center

P.O. Box 11-0236
Riad El Solh 1107 2020
Beirut
Lebanon

Fax: +961-1-366098

Email: as25@aub.edu.lb

 

References

A. SoweidMD 

American University of Beirut Medical Center

P.O. Box 11-0236
Riad El Solh 1107 2020
Beirut
Lebanon

Fax: +961-1-366098

Email: as25@aub.edu.lb

   Permissions and Reprints