A total of 345 consecutive adult out patients referred to the endoscopy unit of the John Radcliffe hospital for colonoscopy on routine clinical indications were prospectively invited to participate. Exclusion criteria were ileus or suspected bowel obstruction, significant gastroparesis or gastric outlet obstruction, toxic colitis or megacolon, pregnancy, or lactation. Patients included were referred mainly due to alarm symptoms such as altered bowel habit in patients > 45 years of age, blood per rectum, iron-deficiency anaemia, unintentional weight loss, and/or other symptoms suggestive of malignant disease. No patient refused to participate. Patients who could not be sedated for colonoscopy were excluded from further evaluation. All patients gave informed consent before participation.

From July to November 2006, we prospectively audited the performance of the large bowel cleansing with magnesium citrate (citramag-only group). The colon preparation protocol with magnesium citrate and senna (citramag/senna group) was audited from August 2007 to January 2008.

All patients were asked to refrain from taking iron tablets 7 d prior to colonoscopy and any medications reducing gastrointestinal motility, e.g. loperamide, 4 d before colonoscopy, but continued all other medications. Two days before colonoscopy, all patients were instructed to eat a low-fiber diet such as white fish, chicken, white bread, eggs, cheese, or potato without skins. High-fiber kinds of food such as red meat, fruit, or vegetables, were to be avoided and patients were advised to consume plenty of fluids. On the day before colonoscopy patients, were instructed to have a low-fiber breakfast. Following that, patients were instructed not to eat any solid food until after colonoscopy and to consume plenty of clear fluids. On the day before colonoscopy at 5:00 pm, patients were instructed to dissolve one sachet of citramag in 200 mL of hot water, which was to be consumed half an hour later when cool. One sachet contains 11.6 g magnesium carbonate and 17.8 g anhydrous citric acid. At 7:00 pm, the patients were instructed to dissolve half of the second sachet of citramag in 100 mL of water and consume it once cooled. Patients were instructed to drink clear fluids (at least a cupful every 30 min) throughout the day and evening before colonoscopy. On the day of colonoscopy, at 6:00-7:00 am in the case of a morning appointment or at 9:00-10:00 am in the case of an afternoon appointment, patients were instructed to consume the other half of a sachet of citramag as described above.

Patients in the citramag-only group underwent colonoscopy after bowel cleansing as described above (two sachets of citramag). Patients in the citramag/senna group were instructed to follow the above instructions but they were also asked to consume one sachet of senna in a cup of warm water and consume it at 2:00 pm on the day before colonoscopy (two sachets of citramag and one sachet of senna).

In order to eliminate interobserver variability, all colonoscopies were performed by the same experienced endoscopist who was, however, not blinded as to the cleansing regimen used. The colonoscopist rated the overall cleansing of the bowel on a 4-point Likert scale as in previous studies[11–13]: 1 = “unacceptable” (large amounts of solid and semisolid fecal residue requiring additional cleansing resulting in rebooking); 2 = “poor” (enough feces or fluid to prevent a completely reliable examination); 3 = “satisfactory” (small amounts of feces or fluid not interfering with the exam); 4 = “good” (no more than small bits of adherent feces/fluid). For the primary efficacy variable, scores of 3 and 4 were considered “adequate” and scores of 1 or 2 were considered “inadequate”.

Prior to colonoscopy, patients were asked to describe their tolerance to the cleansing protocol. Symptoms of nausea, vomiting, bloating, abdominal pain/discomfort, and headache were documented. Patients were asked to rate their symptom severity on a 5-point Likert scale as in previous bowel cleansing studies[12–14]: 1 = “none”; 2 = “mild”; 3 = “moderate”; 4 = “severe”; and 5 = “extreme”. Any adverse event reported during bowel cleansing and/or during the procedure and/or during stay in the recovery room after the procedure was also recorded.

Data are shown as mean and standard deviation or as median and ranges, as appropriate. The χ² test was used for comparisons between categorical variables and the Student’s t test was used for comparisons between quantitative variables. All tests were two-tailed and conducted at a 5% significance level.

In six of 160 patients (3.8%) using the citramag only cleansing protocol, bowel preparation was so poor with solid fecal residues that the procedure was abandoned in the rectosigmoid, and we did not attempt further insertion. In five other patients the cecum was reached but the view was unacceptable due to fecal residues, especially on the right side. Therefore, a repeat colonoscopy was indicated in these patients as well. Totally, 11/160 (6.9%) of the patients had to be rebooked for colonoscopy or virtual colonoscopy because of insufficient bowel cleansing after the citramag regimen.

Complete colonoscopy with visualization of the cecum was reached in 137/160 colonoscopies (86%). Adjustment for tumor strictures (4), cancellation of the procedure due to poor bowel preparation (6) or severity of colitis (1) raised the completion rate to 92% (137/149).

Five out of 182 patients (2.7%) were not examined further than the rectosigmoid because of insufficient bowel cleansing after taking the senna protocol, and three had to be rebooked because of insufficient views despite intubation of the cecum. Thus, eight of 182 patients (4.4%) had unacceptable bowel cleansing under the senna/citramag regimen.

The unadjusted completion rate for cecal intubation was 92%. Adjustment for tumor strictures[4] and exclusion of procedures abandoned in the rectosigmoid due to poor bowel preparation[5] resulted in a completion rate of 97%.

Table 2 shows the quality of colon cleansing results for the two regimens used in the current study as assessed by the endoscopist. The combined citramag/senna regimen proved superior in bowel cleansing as it achieved “adequate” colon visualization (quality of colon cleansing rated as “good” or “satisfactory” in 148/182 (81.3%) compared to 108/160 (67.5%) colonoscopies using the citramag protocol (P = 0.004; Table 3). The colonic polyp detection rate was higher in the citramag/senna group compared to the citramag-only group (P < 0.03; Table 1).

Both protocols were well tolerated. None of the side effects observed were categorized as extreme, and five patients reported severe side effects during bowel preparation. The two bowel cleansing regimens did not significantly differ in the occurrence and intensity of nausea, vomiting, bloating or headache (Table 3). Abdominal cramps occurred more often in the citramag/senna group (P < 0.003; Table 3). Two patients reported “severe” abdominal pain/cramps both of whom had a stricturing tumor in the rectosigmoid.

Adequate colon preparation is essential for the quality of colonoscopy. The completion and polyp detection rate as well as the complication risk and the duration of the colonoscopy are strongly affected by poor bowel cleansing.

As far as we know, this is the first study to evaluate the efficacy of senna as an adjunct to magnesium citrate for large bowel preparation prior to colonoscopy. Our findings demonstrate that the addition of senna to the bowel cleansing protocol with magnesium citrate significantly improves the cleansing outcome, but also increases the occurrence of side effects such as (mild) abdominal pains and discomfort.

Our study results suggest that senna can be used as a helpful adjunct to a cleansing protocol based on magnesium citrate and will increase the efficacy of the bowel preparation before colonoscopy.

Magnesium citrate acts as an osmotic laxative which is often used for bowel preparation; senna is a stimulative agent increasing peristaltic movement.

This interesting study evaluates how the addition of senna alters the cleansing outcome and the occurrence of side effects under a bowel preparation protocol based on magnesium citrate. The results suggest that senna might be a helpful adjunct to standard bowel preparation protocols.