Use of tegaserod along with polyethylene glycol electrolyte solution for colonoscopy bowel preparation: a prospective, randomized, double-blind, placebo-controlled study


Dr M. R. Sanaka, Division of Gastroenterology, MetroHealth Medical Center/CaseWestern Reserve University, 2500 MetroHealth Drive, Cleveland, OH 44109, USA.


Background  Polyethylene glycol electrolyte solution (PEG-EL) used for colonoscopy preparation is not well tolerated by several patients. A significant number of patients have inadequate bowel preparation despite taking PEG-EL.

Aims  To determine the effect of prokinetic agent, tegaserod when given in addition to PEG-EL on patient tolerance, quality of colonic preparation and adverse side effects experienced.

Methods  In this prospective, randomized, placebo-controlled, double-blind study, a total of 130 patients scheduled for colonoscopy were enrolled. They were instructed to take three pills of either tegaserod 6 mg each or placebo (one pill twice on the day prior to and third pill in the morning on the day of colonoscopy) in addition to standard 4L of PEG-EL in the evening prior to the day of colonoscopy. Patient tolerance of preparation, quality of bowel preparation, overall satisfaction and adverse side effects were compared between the two groups.

Results  Fifty-five patients in placebo group and 58 patients in tegaserod group completed the study. There was no difference between the two groups in the tolerance of preparation, quality of bowel preparation, overall satisfaction and the side effects.

Conclusion  Addition of tegaserod to polyethylene glycol electrolyte solution during colonoscopy preparation does not improve patient tolerance, quality of colonic preparation or the adverse side effects.


The polyethylene glycol-electrolyte lavage (PEG-EL) solution has been the standard for colonoscopy bowel preparation since 1980.1 Although this preparation is effective, the need to drink large volumes and the poor taste of PEG-EL are limiting factors that affect patient compliance.2–13 Hence alternative low volume bowel preparations like sodium phosphate solution (NAP) have been developed. NAP has been shown to be better tolerated than PEG-EL.2–15 However, comparative studies of NAP and PEG-EL have yielded conflicting results regarding the quality of bowel preparation2–14 with some studies showing unsatisfactory preparation with NAP.6–14

The prokinetic agent, cisapride has been co-administered with PEG-EL to improve the tolerance and quality of preparation. But the combination of cisapride with PEG-EL has shown conflicting results on efficacy and tolerance.16–20 The use of cisapride is currently restricted worldwide because of the concern of induction of possible cardiac dysrhythmias. Tegaserod is a partial 5-hydroxy tryptamine-4 (5HT4) agonist and stimulates the peristaltic reflex and intestinal secretion, as well as inhibiting visceral sensitivity. Both pre-clinical and clinical investigations have shown that tegaserod can stimulate motility throughout the gastrointestinal tract.21–23 Long-term safety of tegaserod has been established and it is devoid of cardiac dysrhythmias.24–26

We hypothesized that the combined actions of tegaserod on the gastrointestinal tract would make it a favourable adjunct to PEG-EL for colonoscopy bowel preparation. The aims of our study are to evaluate the efficacy of tegaserod when given in conjunction with PEG-EL in improving patient tolerance of the preparation and quality of bowel preparation.


This was a prospective, randomized, double-blind, placebo-controlled trial. The study was conducted at MetroHealth Medical Center (MHMC), a tertiary care, academic medical center, Cleveland, OH, USA. MHMC has a gastroenterology fellowship-training program with seven fellows in training and 10 attending physicians. The protocol was approved by the Institutional Review Board at MHMC.

Inclusion criteria were that the patient must be an out-patient scheduled for routine colonoscopy, over 18 years of age and willing to sign the informed consent. Exclusion criteria included mainly contraindications to take tegaserod, i.e. severe renal impairment (serum creatinine over 2 mg/dL), moderate to severe hepatic impairment (cirrhosis of liver with Child–Pugh scores of B and C), history of bowel obstruction in the past, symptomatic gall bladder disease, suspected sphincter of Oddi dysfunction or abdominal adhesions and known hypersensitivity to the drug. Patients who were taking prokinetic agents and stimulant laxatives were also excluded.

Precolonoscopy preparation

Each patient received oral and written instructions from a trained Registered Nurse (RN) in the clinic regarding the diet, taking the study medication and drinking of PEG-EL. Opaque numbered envelopes containing three pills of identical appearance, each of either placebo or tegaserod 6 mg, were distributed to the patients while instructing them. Both the RN and the patients were blinded to the kind of study medication contained in the envelope. A questionnaire for assessing the baseline gastrointestinal symptoms of patients was completed by all patients at the time of enrollment. They were asked if they had constipation, diarrhoea, abdominal pain, nausea or vomiting.

No solid foods were allowed during the day of preparation i.e. the day prior to scheduled colonoscopy. Clear liquids were allowed ad libitum throughout the day until midnight. On the day of preparation, patients were instructed to take one pill in the morning and the second pill at 3.30 pm. They were instructed to begin consuming the PEG-EL at 4.00 pm and complete ingestion of 4 L before 8 pm. Patients were allowed to take any flavoured PEG-EL of their choice and the type of flavour was not recorded. Patients were instructed to take the third pill in the morning on the day of colonoscopy.

Patient tolerance and symptoms during preparation

A questionnaire for assessing the patient symptoms experienced during the preparation was completed by all patients on their arrival to endoscopy unit before the colonoscopy. Using a 4-point scale (1 = absent, 2 = mild, 3 = moderate and 4 = severe), patients were asked to grade the severity of five symptoms during preparation (nausea and vomiting, abdominal pain or cramps, bloating and anal irritation). Patients were asked to evaluate the overall tolerance of bowel preparation on a 5-point scale (1 = not at all distressing, 2 = mildly distressing, 3 = moderately distressing, 4 = severely distressing and 5 = unable to finish). A score of 3 and less was considered as acceptable tolerance and a score 4 or 5 were considered as unacceptable tolerance. Finally, patients were asked to state their willingness to repeat the assigned preparation and whether they would prefer another preparation.

Colonoscopy and quality of bowel preparation

All colonoscopies were performed by either a senior fellow (individual experience of performing at least 300 colonoscopies) under the direct supervision of an attending physician or by the attenders themselves. All the endoscopists were blinded to the kind of study medication received by the patient. The endoscopist scored the quality of preparation on both the right and left colon (right-proximal to splenic flexure; left-distal to splenic flexure) with a 5-point scale as follows: 1 = no retained fluid or stool; 2 = clear liquid; 3 = liquid stool; 4 = semi-solid stool; and 5 = solid stool. Scores of 1and 2 were considered excellent preparation, score 3 indicated satisfactory preparation, and scores 4 and 5 as an unsatisfactory preparation. The quality of preparation is defined as adequate if the score is 3 or less, and inadequate if the score was 4 and above. All colonoscopies are performed under conscious sedation by giving a combination of narcotic, meperidine along with a benzodiazepine, midazolam.

Statistical analysis

Intent to treat per protocol analyses was performed. The differences between the two groups were analysed by t-test family for parametric interval data (i.e. patient age, amount of medication used for sedation); Mann–Whitney U-test for ordinal or non-parametric interval data (i.e. patient symptom scores) and Chi-Square test for categorical variables (i.e. successful cleanout). A P-value of <0.05 (two tail) was considered as statistically significant. Assuming that 65% in placebo group and 90% in tegaserod group will have acceptable tolerance to bowel preparation, a sample size of 56 per group will be required for detecting the difference with a power of 90% and a Type I error of 0.05. This improved difference in patient acceptance (25%) was regarded as clinically relevant as was proposed in the previous studies.20


Seventeen of 130 patients randomized were excluded from the final analysis (12 patients failed to keep their appointment for colonoscopy and five patients opted out of the study). Hence a total of 113 patients (55 in placebo group and 58 in tegaserod group) were analysed. Both groups were comparable with regards to patient demographics and baseline gastrointestinal symptoms (Table 1). No serious adverse events occurred in either of the groups. There were two patients in the Placebo group and one in the tegaserod group that took only two of the three pills given.

Table 1.  Demographics of patients
 Placebo group (n = 55)Tegaserod group (n = 58)P-value
Mean age in years (s.d.)51.9 (11.0)52.5 (13.9)0.82
 Female34 (61.8%)37 (63.8%)0.83
 White31 (56.4%)35 (60.3%)0.55
Baseline symptoms
 Abdominal pain17 (30.9%)24 (41.4%)0.29
 Constipation13 (23.6%)15 (25.9%)0.78
 Diarrhoea13 (23.6%)21 (36.2%)0.15
 Nausea11 (20%)10 (17.2%)0.71
 Vomiting4 (7.3%)1 (1.7%)0.15
Patients who had colonoscopy in past17 (30.9%)20 (34.5%)0.69
Patients who had abdominal/pelvic surgery in past19 (34.5%)26 (44.8%)0.26
Patients with diabetes mellitus9 (16.4%)10 (17.3%)0.90

Side effects and patient tolerance

There was no difference in adverse side effects experienced by patients during bowel preparation between the two groups (Table 2). Tolerance of preparation was similar in both groups. Also abdominal pain and bloating associated with colonoscopy procedure were similar in the two groups.

Table 2.  Side effects reported by patients during colonoscopy preparation, colonoscopy procedure and post-procedure (Scale 1–5; lower scores indicate lesser symptoms)
 Placebo group (n = 55)Tegaserod group (n = 58)P-value
  1. * Data expressed as mean (s.d.) and percentages.

Nausea/vomiting1.40 (0.87)1.48 (0.90)0.54
Abdominal discomfort/cramps1.40 (0.74)1.45 (0.65)0.41
Bloating1.60 (0.74)1.74 (0.83)0.39
Anal irritation1.38 (0.68)1.41 (0.70)0.69
Tolerance of preparation
 Mean score2.04 (1.20)2.14 (0.96)0.27
 No. of patients with acceptable tolerance* (score of 3 and less)47 (85.5%)53 (91.4%)0.32
 Abdominal discomfort during colonoscopy1.76 (0.84)1.81 (0.85)0.79
Symptoms 1 h after colonoscopy
 Abdominal discomfort1.44 (0.74)1.40 (0.70)0.86
 Bloating1.62 (0.87)1.53 (0.63)0.96
Symptoms one day after colonoscopy
 Abdominal discomfort1.24 (0.43)1.23 (0.42)0.92
 Bloating1.18 (0.43)1.25 (0.54)0.64

Efficacy of colon cleansing and colonoscopic Characteristics

There was no difference found between the Placebo and tegaserod groups, findings are summarized in Table 3. Amount of sedation required for colonoscopy was also similar in both the groups.

Table 3.  Quality of bowel preparation and colonoscopic characterisitics
  Placebo group (n = 55)Tegaserod group (n = 58)P-value
  1. * Data expressed as mean (s.d.) and percentages.

Quality of bowel preparation (Scale 1–5, lower scores indicate better preparation)
 Left side of colon2.28 (1.10)2.19 (1.03)0.64
 Right side of colon2.65 (0.96)2.45 (1.13)0.34
 No. of patients with adequate preparation on left side49/54 (90.7%)51/57 (89.5%)0.82
 No. of patients with adequate preparation on right side47/54 (87%)46/56 (82.1%)0.48
 No. of patients with excellent preparation on left side27/54 (50%)33/57 (57.9%)0.40
 No. of patients with excellent preparation on right side18/54 (33.3%)26/56 (46.4%)0.16
 Insertion time to cecum in minutes12.9 (7.23)11.8 (8.01)0.47
 Colonoscopy completion rate54/55 (98.2%)56/58 (95.6%)0.50
 Technical difficulty (Scale 1–10, lower scores indicate less difficulty)4.67 (2.16)4.58 (2.29)0.81
Amount of sedation required
 Meperidine (mg)77.4 (38.4)73.6 (40.9)0.33
 Midazolam4.76 (1.4)4.9 (1.7)0.52
 Diverticulosis seen*15 (27.3%)19 (32.3%)0.53

Overall satisfaction and willingness to repeat the preparation

There was no difference in overall satisfaction with preparation between the placebo and tegaserod groups [mean score of 7.24 (S.E.M., 2.88) vs. 7.24 (S.E.M., 2.82), P-value 0.98]. Also the willingness to repeat the assigned preparation in future was similar between the placebo and tegaserod groups, 76.4% vs. 77.6% (P-value 0.88).


The key finding of this prospective randomized-controlled study was that the addition of tegaserod to standard PEG-EL for colonoscopy bowel preparation does not help in improving patient tolerance of taking preparation or the quality of bowel preparation.

This is the first study, to our knowledge, that has evaluated the effect of tegaserod when used in conjunction with standard PEG-EL for precolonoscopy bowel preparation. The combined effects of tegaserod on the gastrointestinal tract, i.e. increasing motility throughout the gut and decreasing visceral sensitivity and flatulence, prompted us to conduct this study. After extreme restrictions on the use of cisapride worldwide because of the risk of cardiac dysrhythmias, tegaserod is the only available drug that promotes pangastrointestinal motility. Adult dosage for tegaserod is 6 mg twice a day for constipation and constipation predominant irritable bowel syndrome,22, 27 hence we chose to give 6 mg twice a day dosage on the colonoscopy preparation day. A third dose in the morning of the day of procedure is given to help decrease the abdominal discomfort and flatulence during and in the post-procedure period. The peak plasma concentration of tegaserod is reached approximately 1 h after oral dosing and has an estimated half-life of 11 ± 5 h.

In our study, there was no difference in any of the parameters compared between the tegaserod and placebo groups. Based on the acceptable tolerance rates in previous studies,14, 18 we assumed that 65% of the patients in placebo group would have acceptable tolerance of bowel preparation. The acceptable tolerance of standard PEG-EL without the addition of tegaserod was high (85%) in our study. This may have been related to unrestricted use of flavoured PEG-EL, which are known to be better tolerated than plain PEG-EL.28 Even though the acceptable tolerance in tegaserod group was 91.4%, the difference between the groups was not statistically significant. Negative result of this study may be due to inadequate power, as our study was designed to be able to detect a major effect (25%) of tegaserod in improving patient tolerance of preparation. Because of the high tolerance rate of PEG-EL alone, further improvement with the addition of tegaserod might be difficult to demonstrate.

Tegaserod has been shown to improve motility throughout the gastrointestinal tract.21–23 This property might make it a favourable adjunct to standard PEG-EL especially in certain patient groups such as those with gastroparesis, constipation and slow transit. Drinking a large volume i.e. 4L of standard PEG-EL might be a difficult task because of adverse side effects in these patients. Hence, baseline tolerance of standard PEG-EL is expected to be low in these patient groups and addition of tegaserod may show a significant improvement in patient tolerance, adverse side effects and the quality of bowel preparation.

There are some limitations in our study. The symptoms reported by patients in our study were subjective and there was no objective scoring used to confirm these. Multiple endoscopists performed the colonoscopies and rated the quality of preparation, hence there may be no concordance between the different endoscopists. The strengths of our study include the study design: a prospective, randomized, placebo-controlled, double-blinded study and the intent to treat analysis used for analyzing the data.

In conclusion, tegaserod when given in conjunction with PEG-EL does not improve the patient tolerance or the quality of bowel preparation. Further studies are required to evaluate the efficacy of tegaserod in conjunction with PEG-EL in subgroups of patients with gastroparesis, constipation and slow transit.


This study was presented as an abstract in American College of Gastroenterology 70th Annual Scientific Meeting, 28 October to 2 November 2005, Honolulu, HI, USA.

This study was supported in part through the General Clinical Reseach Center at Case Western Reserve University, Cleveland, OH, USA (NIH no. MO1 RR00080).