Correspondence to: Professor S. Chaussade, Department for Hepatogastroenterology, Hôpital Cochin, Pavillon Achard, 27 rue du faubourg Saint Jacques, 75679 Paris cedex 14, France. E-mail: firstname.lastname@example.org
Aim : To compare standard and maximum daily doses of polyethylene glycol 3350 plus electrolytes (Transipeg) and polyethylene glycol 4000 (Forlax) in a multicentre, double-blind, randomized, parallel-group study.
Methods : Ambulatory patients with idiopathic chronic constipation were randomized to receive Forlax (10 or 20 g) or Transipeg (5.9 or 11.8 g) for 1 month. The primary efficacy end-point was stool frequency. Secondary efficacy parameters included stool consistency, date of occurrence of first motion, straining on defecation, rectal evacuation, abdominal pain and distension. Adverse events were recorded.
Results : Stool frequency was significantly increased compared with baseline in all treatment groups (P = 0.0001). Most patients (≥ 67.3%) had their first stool within 1 day of starting treatment. Stool consistency significantly improved compared with baseline in all treatment groups (P = 0.0001). The percentage of patients with normal stool consistency was significantly higher for standard-dose Transipeg vs. both maximum-dose treatments (P < 0.01). Other secondary parameters were also significantly improved compared with baseline in all treatment groups (P = 0.0001). All medications were well tolerated.
Conclusions : Standard-dose Transipeg (5.9 g) normalized stool consistency with less semi-liquid or liquid stools than maximum-dose Transipeg and Forlax, with a non-significant trend towards less semi-liquid or liquid stools than standard-dose Forlax.
Constipation is a very common condition in the Western World, and its prevalence varies between 5 and 30% depending on the diagnostic criteria used.1 Constipation can be secondary to organic gastrointestinal or systemic disease, or may be due to non-organic gastrointestinal dysfunction. The Rome II diagnostic criteria define chronic idiopathic, or functional, constipation as the presence of two or more of the following for at least 12 weeks (which need not be consecutive) in the preceding 12 months: (i) straining in greater than one out of four defecations; (ii) lumpy or hard stools in greater than one out of four defecations; (iii) sensation of incomplete evacuation in greater than one out of four defecations; (iv) sensation of anorectal obstruction/blockade in greater than one out of four defecations; (v) manual manoeuvres to facilitate greater than one out of four defecations (e.g. digital evacuation, support of the pelvic floor); and/or (vi) less than three defecations per week.2
The pathophysiology of constipation involves the combination of slow colonic transit, hard stool consistency and/or problems with rectal evacuation. Constipation is often associated with a number of other symptoms, including abdominal pain, dyspepsia, flatulence, headache and asthenia.3 An increase in dietary fibre, increased hydration and physical exercise are recommended. However, these are often inadequate, and pharmacological intervention with a laxative is frequently required.3 In addition, the guidelines for the pharmacological treatment of chronic idiopathic constipation are not always well defined, and there is a need for effective, well-tolerated laxatives, as many existing agents cannot be used on a long-term basis due to unwanted, often serious side-effects.
Different types of laxatives are currently available, and these are associated with a range of side-effects. Bulking agents act by increasing stool weight and the production of volatile fatty acids. However, they can interfere with the absorption of certain drugs, and also have a low efficacy and a slow onset of action, which often result in poor compliance.1,4 Irritants, such as senna and bisacodyl, can be used to increase peristaltic activity.4 Side-effects of this approach include laxative dependency and electrolyte imbalances due to the loss of large amounts of water passed in the stools.4 In addition, increasing doses are required to maintain an effect with long-term use. Electrolyte imbalances are also associated with the frequent use of magnesium salts. Finally, synthetic disaccharides, such as lactulose and lactitol, are broken down by bacteria in the colon. A large amount of gas is produced as a result of the fermentation of these compounds, causing bloating and abdominal pain.1,4 Lactulose is also associated with changes in bacterial colonic flora and a subsequent lack of efficacy with long-term use.1
Polyethylene glycol (PEG), or macrogol, is an osmotic agent used in lavage solutions for gut cleansing before colonoscopy or bowel surgery.1 PEG is not absorbed from the gastrointestinal tract, and therefore is not metabolized, nor fermented by the bacterial flora. Commercially available PEG formulations for use as laxatives have been developed using PEG 3350 and PEG 4000, with the addition of electrolytes (iso-osmotic solutions) and without the addition of electrolytes (hypo-osmotic solutions). The major disadvantage of the use of PEG, without the addition of electrolytes, is the potentially poor safety due to plasma electrolyte loss. PEG 3350 is more hygroscopic than PEG 4000 due to its lower molecular weight,5 suggesting a higher potency on a per gram basis.
Transipeg is an iso-osmotic laxative containing PEG 3350 and electrolytes that has been developed for the safe, symptomatic treatment of constipation, regardless of its cause. The mechanism of action of Transipeg was recently investigated in healthy volunteers. Transipeg was found to increase the quantity of unbound free water (dilution or free water) without increasing ‘structural’ water, thereby modifying hard stools into soft moulded stools that were easy to evacuate.6,7 This effect on stool consistency was observed with a low PEG dose of < 10 g/day.7 Forlax is an osmotic laxative containing PEG 4000 without electrolytes; its mode of action involves the hydration and softening of stools. Both PEG 3350 with electrolytes3 (data on file, Roche Consumer Health) and PEG 40008 have been proven to be effective in the treatment of constipation. Therefore, a head-to-head comparison of these two laxatives in their standard and maximum recommended doses was considered to be appropriate.
The purpose of this study was to compare two different doses of two commercially available PEG formulations: standard and maximum recommended daily doses of PEG 3350 in an iso-osmotic solution (Transipeg) were compared with standard and maximum recommended daily doses of PEG 4000 in a hypo-osmotic solution (Forlax).
Materials and methods
The study had a prospective, multicentre, double-blind, randomized, parallel-group design and was performed in 69 centres in France. Patients fulfilling the entry requirements of the study (see below) were randomized to receive PEG 4000 alone (Forlax, Laboratoires Beaufour) or PEG 3350 combined with electrolytes (Transipeg F. Hoffmann–La Roche AG, Basle, Switzerland) for 1 month. In order to maintain double-blind conditions, a double-placebo technique was used. Therefore, patients received the following: (i) group 1 (standard Transipeg dose): PEG 3350 (5.9 g) plus electrolytes, one sachet per day; in addition, one sachet of Transipeg placebo and two sachets of Forlax placebo per day; (ii) group 2 (standard Forlax dose): PEG 4000 (10 g) without electrolytes, one sachet per day; in addition, one sachet of Forlax placebo and two sachets of Transipeg placebo per day; (iii) group 3 (maximum Transipeg dose): PEG 3350 (11.8 g) plus electrolytes, two sachets of 5.9 g per day; in addition, two sachets of Forlax placebo per day; (iv) group 4 (maximum Forlax dose): PEG 4000 (20 g) without electrolytes, two sachets of 10 g per day; in addition, two sachets of Transipeg placebo per day.
Each sachet of Forlax consisted of 10 g of PEG 4000, sodium saccharin and flavouring. Each sachet of Transipeg consisted of 5.9 g of PEG 3350, sodium chloride (0.146 g), potassium chloride (0.075 g), sodium bicarbonate (0.168 g), sodium sulphate (0.568 g), aspartame and flavouring. Each sachet had to be diluted in a glass of water before ingestion. All treatments were supplied in sachets of identical appearance.
All patients provided written informed consent and the study was conducted in compliance with the Helsinki Declaration (including the Tokyo, Venice and Hong Kong amendments) and local clinical study laws and regulations.
Male and female ambulatory patients aged 18 years or older, presenting with constipation in general practice, were recruited. To be included in the study, patients had to have a history of constipation for longer than 3 months and fulfil one of the following criteria: less than three motions per week, hard and/or lumpy stools requiring straining or a feeling of incomplete rectal evacuation. In addition, patients greater than 40 years of age had to have had a colonoscopy or barium enema in the previous 5 years.
Patients were excluded from the study if they suffered from severe systemic illness, severe psychiatric disease, advanced carcinoma or other colonic disease causing an excessively fragile mucosa, occlusion or subocclusion, abdominal pain of unidentified origin or an active gastrointestinal ulcer. Those patients on anti-ulcer treatment, those with known hypersensitivity to the study medication and those with poor compliance were also excluded from the study.
No physical or chemical laxatives other than the study treatments were permitted during the study, and opiates were also not permitted. During the treatment period, patients were allowed to use suppositories if they did not pass a stool for three consecutive days. Stools passed after the use of suppositories were not taken into account. Patients were assessed at baseline (week 0), and at the second (week 2) and fourth (week 4) week of treatment. At each visit, a clinical examination, including blood pressure, heart rate and body weight, was undertaken. A record was made of intercurrent/adverse events as volunteered by the patient or elicited at the interview. A review of bowel habits, including mean motions/week, stool consistency, straining at stool, feeling of rectal voiding, abdominal distension and pain, was also performed.
The primary efficacy end-point was the frequency of stools (mean motions per week), assessed at weeks 2 and 4. Stool frequency was assessed by investigators using information provided by the patient in the form of daily diary data, and from volunteered and elicited comments from the patients at each visit.
Secondary efficacy end-points were self-assessed by the patients using rating scales and recorded in daily diaries. Investigators assessed secondary end-points as for the primary end-point. The following parameters were assessed: stool consistency [1 (liquid) to 6 (very hard)]; date of occurrence of first motion; degree of straining on defecation [1 (none) to 4 (considerable)]; sensation of rectal evacuation [1 (complete) to 4 (none)]; abdominal distension [1 (none) to 4 (considerable)]; abdominal pain [1 (none) to 4 (considerable)]; quality of life (assessed using a visual analogue scale); and global impression of efficacy (assessed using a visual analogue scale).
In addition, the percentage (proportion) of patients with normal stool consistency, no or mild straining, complete or satisfactory evacuation, no or mild bloatedness and no or mild abdominal pain was also assessed for each group.
Clinical tolerance was assessed by: clinical examination by the investigator; adverse events as recorded in patient diaries and volunteered/elicited from patients at each visit; and patient and investigator global impression (visual analogue scale) at each visit.
The inter-group comparison at baseline was performed using the likelihood ratio χ2 test, without the restrictions that apply to the use of the χ2 test. Quantitative parameters were compared using analysis of variance (anova), with a Bonferroni multiple comparisons test in the event of a significant global effect, followed by a Kruskal–Wallis test if conditions for the use of anova were not met.
The primary end-point of mean motions per week was assessed using a two-way repeated measures anova for treatment and time (baseline, week 2 and week 4) and time–treatment interaction. Secondary (semi-quantitative) end-points and time to first motion (quantitative end-point) were assessed using a one-way anova. Comparisons of the primary end-point and all secondary parameters were performed between all medications at baseline, week 2 and week 4 using the Mann–Whitney U-test for the following comparisons: group 1 vs. group 3; group 2 vs. group 4; group 1 vs. group 2; group 3 vs. group 4. In addition, the results for each medication were compared between week 2 and baseline, week 4 and baseline, and week 4 and week 2, using the Wilcoxon signed rank test. Global impression of efficacy (visual analogue scale) and global satisfaction (visual analogue scale) were assessed using a two-way (treatment × time) anova. Differences were considered to be statistically significant at P < 0.05.
Population and demographics
Two hundred and seventy patients, reflecting a typical population presenting with constipation to general practice, were randomized. Four patients were subsequently excluded from the study, and 55 patients interrupted treatment prematurely due to adverse events (n = 28), treatment failure (n = 11), other reasons (n = 15) or were lost to follow-up (n = 1). Of a total of 266 patients, 263 (98.9%) fulfilled the inclusion criteria. The remaining three patients (one from group 1 and two from group 4) experienced abdominal pain and bloatedness at inclusion and were also included in the study.
The total population involved 266 patients, with similar numbers included in each treatment group: group 1, n = 67; group 2, n = 66; group 3, n = 70; group 4, n = 67. No significant difference was seen between patients in the four treatment groups in terms of patient demographics or baseline characteristics, including age, gender and constipation severity (Table 1). The mean age on inclusion was 51 years for women and 57.7 years for men (mean, 52.2 ± 18.5 years). The majority of patients (85%) included in the study were women. The mean duration of constipation was longer than 15 years.
Table 1. Demographic and baseline characteristics of the study population
Transipeg, 5.9 g
Forlax, 10 g
Transipeg, 11.8 g
Forlax, 20 g
No. of patients enrolled
Excluded from analysis
Included in efficacy and safety analysis
Dropouts, n (%)
Completed the study
Age, years (mean ± s.d.)
50.2 ± 17.4
50.6 ± 19.7
54.0 ± 17.4
52.2 ± 19.3
Women, n (%)
Period of constipation, months (mean ± s.d.)
194 ± 168
186 ± 176
190 ± 167
199 ± 182
Stool frequency (motions per week)
2.2 ± 1.3
1.8 ± 1.0
2.0 ± 0.9
2.4 ± 1.7
The primary efficacy parameter of stool frequency (mean motions per week) was significantly increased compared with baseline in all treatment groups at weeks 2 and 4 (P = 0.0001 for all groups) (Table 2), and there was no significant difference between the four groups. In addition, all medications had a fast onset of action. The time to first stool (days, mean ± s.d.) was somewhat shorter in both Transipeg groups compared with the Forlax groups, but this did not reach statistical significance: group 1, 0.85 ± 1.14; group 2, 1.38 ± 1.93; group 3, 0.89 ± 1.17; group 4, 1.15 ± 1.39 (P = 0.1385). There was also a trend towards an increased percentage of patients having their first motion within 1 day of starting treatment in both Transipeg groups compared with the Forlax groups, but again this did not reach statistical significance: group 1, 77.1%; group 2, 67.3%; group 3, 82.3%; group 4, 73.1%.
Table 2. Efficacy parameters at baseline (week 0), week 2 and week 4 — rating scale scores for the total study population
Transipeg, 5.9 g (n = 65)
Forlax, 10 g (n = 65)
Transipeg, 11.8 g (n = 69)
Forlax, 20 g (n = 67)
VAS, 100 mm visual analogue scale.
P = 0.0001 vs. week 0 (for all medications) using the Wilcoxon signed rank test.
P = 0.0001 vs. week 2 (for all medications) using the Wilcoxon signed rank test.
P < 0.01 vs. Transipeg 5.9 g using the Mann–Whitney U-test.
P < 0.001 vs. Transipeg 5.9 g using the Mann–Whitney U-test.
P = 0.001 vs. Forlax 10 g using the Mann–Whitney U-test.
Stool consistency was significantly improved compared with baseline in all treatment groups at weeks 2 and 4 (P = 0.0001 for all groups) (Table 2). Stool consistency in groups 3 and 4 was significantly more liquid compared with group 1 at week 2 (P < 0.001) and week 4 (P < 0.01) (Table 2). Stool consistency in group 4 was also significantly more liquid compared with group 2 at week 2 (P = 0.001) (Table 2). Straining at stool, rectal evacuation, abdominal bloatedness and abdominal pain were all significantly improved compared with baseline in all four treatment groups at weeks 2 and 4 (P = 0.0001 for all groups and parameters) (Table 2).
Similarly, the quality of life was significantly improved compared with baseline in all four treatment groups at weeks 2 and 4 (P = 0.0001 for all groups). An increase in quality of life was seen over time (from week 2 to week 4) in all four treatment groups (P = 0.0001 for all groups) (Table 2). Global patient satisfaction and investigator's global impression of efficacy were similar across the groups, with a significant increase observed for all four groups at week 4 compared with week 2 for investigator's global impression of efficacy (P = 0.0001).
The percentage of patients experiencing normal symptoms at each time point was also assessed for each treatment group. There was no significant difference between the four treatment groups for most parameters (degree of straining on defecation, sensation of rectal evacuation, abdominal distension and abdominal pain), with the exception of stool consistency (Figures 1 and 2). All four treatments improved stool consistency from baseline, with the percentage of patients with normal stool consistency increasing in each group at weeks 2 and 4. However, the percentage of patients with normal stool consistency was significantly higher in groups 1 and 2 at week 2 compared with groups 3 and 4 (Figure 1). At week 4, the percentage of patients with normal stool consistency was significantly higher in group 1 compared with groups 3 and 4 (P < 0.01) and in group 2 compared with group 3 (P < 0.05) (Figure 1). The percentage of patients with semi-liquid or liquid stool consistency increased from baseline to weeks 2 and 4 in all groups, except for group 1 (Figure 2). Fewer patients in group 1 had semi-liquid or liquid stools compared with groups 3 and 4 at weeks 2 and 4 (P < 0.01) (Figure 2). Also, fewer patients in group 1 had semi-liquid or liquid stools compared with group 2 at week 4, but this did not reach statistical significance (P = 0.06).
All medications were well tolerated with no differences observed between the four treatment groups. There were no deaths, and no serious or significant adverse events. Approximately 50–60% of patients reported at least one adverse event in each treatment group: group 1, 33/65; group 2, 33/65; group 3, 42/69; group 4, 36/67. The majority of adverse events were gastrointestinal: 136 patients reported at least one gastrointestinal event, and these were evenly distributed across the four treatment groups.
In terms of patients with events probably or possibly related to study treatment, 69 patients had one or more episodes of diarrhoea or liquid stools: group 1, 13.8%; group 2, 16.9%; group 3, 36.2%; and group 4, 35.8%. The incidence was significantly higher in the maximum-dose groups compared with the standard-dose groups (Figure 3). Other frequently reported gastrointestinal adverse events probably or possibly related to study treatment included distension, flatulence and abdominal pain; these were evenly distributed across all groups.
Vital signs and clinical examination were normal in over 95% of patients in the four treatment groups, and body weight did not change during the 4-week study period. The global impression of tolerability (investigators) was satisfactory and similar in all groups (70/100 mm visual analogue scale rating), with a significant increase in tolerability observed for all four groups at week 4 compared with week 2 (P = 0.0001).
This study compared the efficacy and tolerability of two different doses of PEG 3350 in an iso-osmotic solution (Transipeg) with two different doses of PEG 4000 in a hypo-osmotic solution (Forlax), in a typical general practice setting. The results showed that both doses of Transipeg and Forlax were effective in treating the symptoms of constipation and were well tolerated. However, the standard recommended dose of Transipeg normalized stool consistency with less semi-liquid or liquid stools than maximum-dose Transipeg or Forlax.
All medications (standard and maximum recommended doses) were effective in the treatment of idiopathic chronic constipation, as assessed by the frequency of stools, stool consistency, straining at stool, rectal evacuation, abdominal bloatedness and abdominal pain. All medications had a fast and very favourable onset of action, with the majority of patients having their first stool within 1 day of treatment initiation. Transipeg (PEG 3350 plus electrolytes) was more potent than Forlax (PEG 4000 without electrolytes) on a per gram basis. This finding is supported by previously reported data on the relative potencies of the two agents.8 This difference may be explained by the fact that PEG 3350 is more hygroscopic than PEG 4000 due to its lower molecular weight,5 although the role of electrolytes cannot be excluded.
All constipation symptoms were normalized in at least 60% of patients, with the exception of stool consistency. Stool consistency was normalized in a higher proportion of patients in the standard-dose groups, whilst the proportion of patients with liquid stools was higher in the maximum-dose groups. The lowest proportion of patients reporting diarrhoea and liquid stools was in the standard-dose Transipeg group; at week 4, standard-dose Transipeg resulted in fewer incidences than standard-dose Forlax, although this was not quite significant (P = 0.06). In addition, all medications resulted in an increase in patient-assessed quality of life and investigator-assessed global impression of efficacy, and the overall perception of the efficacy of standard-dose Transipeg (5.9 g) was very high.
Adverse events were reported by 50–60% of patients, and the majority were gastrointestinal. The incidence of diarrhoea and liquid stools was significantly higher in the high-dose groups compared with the standard-dose groups. All medications were well tolerated, with a significant increase in the global impression of tolerability observed for all four groups at week 4 compared with week 2. However, tolerability was higher with the standard doses of both preparations due to fewer episodes of diarrhoea and liquid stools.
These results confirm the balanced laxative effect of Transipeg 5.9 g (PEG 3350 and electrolytes) in the symptomatic treatment of constipation, with the combination of high efficacy and good tolerability. These results support those obtained from previous studies of Transipeg (data on file, Roche Consumer Health). In addition, Transipeg has also shown sustained efficacy in improving constipation symptoms over a 6-month period.3 Moreover, in this self-titration study, patients chose the 5.9 g dose as being the most effective and best tolerated for the relief of symptoms. In the current study, Transipeg was also shown to induce the first motion within the first day of treatment in the majority of patients.
Although this study was not placebo controlled, it was comparative and double-blind. Furthermore, the four treatment groups did not differ at baseline, and therefore the differences at weeks 2 and 4 between the four treatment groups are valid. Considering the complex pathophysiology of constipation, PEG may exert its effect by modifying the colonic transit time, stool weight or stool consistency. Whereas a dose of 6 g/day of PEG is effective at modifying stool consistency and therefore facilitating stool evacuation, 26 g/day decreases the colonic transit time, and doses > 26 g/day increase stool weight without modifying the oro-anal transit time.9 In addition, previous studies have demonstrated that PEG 3350 and 4000 are both effective in treating the symptoms of functional constipation3,8 (data on file, Roche Consumer Health). Transipeg is an iso-osmotic laxative containing PEG 3350 and electrolytes that has been developed for the symptomatic treatment of constipation, and therefore the balanced effect seen with the standard dose in this study was not unexpected. The equivalence of 5.9 g of Transipeg with 10 g of Forlax may be due to the higher hygroscopic nature of PEG 3350 compared with PEG 4000.5 This needs to be confirmed by the precise physical and chemical analysis of both PEGs.
Transipeg has a good long-term safety record, which supports the benefit of the iso-osmotic preparation (the addition of electrolytes); it is not absorbed and metabolized and therefore is not influenced by pharmacokinetic changes in different patient populations. In conclusion, standard-dose Transipeg (5.9 g) is effective, safe and well tolerated in the treatment of patients with chronic constipation symptoms.
The authors would like to thank Dr Meelis Reidla for critical review of the manuscript and productive comments. The efforts of the study group investigators (see below) and their study co-ordinators are also gratefully acknowledged. This study was supported by a grant from F. Hoffmann-La Roche AG, Basle, Switzerland.