Efficacy of tegaserod for functional constipation in Chinese subjects: a randomized double-blind controlled trial in a single centre



This article is corrected by:

  1. Errata: Errata / Corrigenda Volume 25, Issue 8, 1002, Article first published online: 30 March 2007

Dr A. O. On Chan, Department of Medicine, University of Hong Kong, Queen Mary Hospital, Hong Kong.
E-mail: aoochan@hku.hk




Tegaserod has been shown to be effective in chronic constipation in Western population.


We investigated if tegaserod is equally effective in Chinese population.

Materials and methods

Two hundred and fifty patients were randomized to a double-blinded 8-week treatment of tegaserod 6 mg b.d. or placebo. Response during weeks 1–4 was defined as an increase in complete spontaneous bowel motion ≥1/week. Secondary efficacy included response during weeks 1–8, individual symptoms and scores, quality of life and global assessment of bowel habits and constipation.


One hundred and nine patients from the treatment group and 107 from the placebo group completed the 8-week treatment. Responder rates was 47.7% vs. 29% for the treatment and placebo groups (P = 0.005). The sustained complete spontaneous bowel motion rate was 29.4% vs. 15.7% in the two groups (P = 0.016). The response rates were higher than that reported previously in the Caucasian studies. There was improvement in the scores for stool form scale, bothersomeness of constipation, abdominal distension/bloating and satisfaction of bowel habit (P < 0.05). The mental score was higher in the treatment group (46.8 ± 9 vs. 43.6 ± 10, P = 0.01).


Tegaserod is effective in relieving chronic constipation in Chinese population. The efficacy observed may be higher than that in Western population.


Functional constipation is a common problem in clinical practice. In the Western population, the prevalence was reported to be as high as 24% in elderly subjects and more commonly among women.1 We observed a prevalence of 14% of constipation in the Hong Kong population.2

Functional constipation (idiopathic constipation) usually can be subdivided into slow transit and anorectal dyssynergia.3 tegaserod is a new class of 5-HT4 agonists. Activation of 5-HT4 receptors triggers the release of neurotransmitters from the enteric nerves resulting in increased contractility and stimulation of the peristaltic reflex.4–8 In animal models, tegaserod acts as a motility-enhancing agent, exerting activity throughout the gastrointestinal tract.9 tegaserod has also been shown to significantly accelerate bowel transit in healthy volunteers and in patients with constipation-predominant irritable bowel syndrome. 10, 11 In addition, Johanson et al.12 and Kamm et al.13 have also reported the efficacy of tegaserod (6 mg b.d.) in chronic constipation to be around 40–43%, with the responder defined by ≥1 increase of complete spontaneous bowel motion (CSBM) per week compared with baseline. Subjects from the two studies were recruited from the Western population.

The Chinese population has a completely different diet and cultural behaviour from that of the Western countries. Despite that a minority of the recruited subjects from the two studies by Johanson et al. and Kamm et al. was Asian, the efficacy of tegaserod in the Western population may not be directly applicable to the Chinese population. In addition, there is scarce data on constipation in the Chinese population. Despite the prevalence of functional constipation in Hong Kong is approaching that of the Western population, most of the population in Hong Kong is Chinese. Thus the aim of the current study was to investigate the efficacy of tegaserod 6 mg b.d. in the Chinese patients with functional constipation in a single centre. We adopted the same protocol as reported by Kamm et al.13 in order to have better comparison with the Western reports.


Protocol, assignment and randomization

We adopted the protocol as reported by Kamm et al.13 Except that the current study was a single centre double-blinded placebo-controlled trial from April 2005 to April 2006, consisting of a 2-week baseline assessment and 8-week treatment period. Briefly, patients were given same size tablets containing tegaserod 6 mg b.d. or placebo b.d., each taken within 30 min of morning and evening meals throughout the 8-week period. Patients were seen by the investigator on day 1, week 4 and 8. Bisacodyl was given as rescue medicine and patients were instructed to take if no bowel motion for ≥4 days. Randomization was performed using a validated computer system. All involved personnel were blinded to medications given until study conclusion. Concealed allocation was employed in this study. The study was performed in accordance with the Declaration of Helsinki regarding informed patient consent and institutional review board approval.

Patients and controls

Constipated subjects were referred from general practitioners to the Constipation Clinic in Queen Mary Hospital, Hong Kong, which is a tertiary referral centre. The constipated subjects were recruited after assessment by a gastroenterologist and satisfied the criteria by Kamm et al.:13 age ≥18 years, constipation for 6 months or more, with an average of less than three CSBM per week; and at least one of the following occurring for more than 25% of the time: straining, passage of lumpy or hard stools and sensation of incomplete evacuation. CSBM referred to the feeling that defecation led to complete passage of stool rather than partial or incomplete evacuation of stool without relief of symptoms. Patients with a previous history of constipation predominant irritable bowel syndrome using the Rome II criteria were excluded from the present study. Additional exclusion criteria were inability to understand Chinese, constipation because of secondary causes, history of malignancy and significant systemic disease. Pregnant or breast feeding women were also excluded. Patients who failed to complete the diary or if the constipation was not confirmed by diary were excluded.


Assessment of constipation symptoms was done in a diary format. Patients recorded their constipation symptoms in a diary throughout the 2-week baseline and 8-week treatment period. On a daily basis, they recorded the symptoms of CSBM, straining score (using a 3-point score: no straining/acceptable straining/too much straining), a feeling of incomplete evacuation (yes/no), the 7-point Bristol stool scale form (range from 1, separate hard lumps, to 7, watery with no solid pieces),14 and the intake of Bisacodyl as rescue medicine. Weekly assessment include the patients’ satisfaction with bowel habits over the past week (using a 5-point ordinal scale, 0 = very satisfied, 4 = not at all satisfied), and the bothersomeness of constipation, the bothersomeness of abdominal discomfort/pain and the bothersomeness of distension/bloating (each using a 5-point ordinal scale, 0 = not at all, 4 = a very great deal).

Patients also completed the shorter form of SF-36 questionnaire (SF-12) questionnaire15 at baseline and at week 8 in order to evaluate the effect of treatment on quality of life and general health status. The scale of SF-12 is scored using 0–100, with higher scores indicating better performance.

Assessment of bowel transit time

Normal and slow transit constipation were confirmed by X-ray and colonic motility studies performed in all patients during the assessment period. Colonic transit time was assessed through the use of radiopague markers. In brief, four sets of distinctive radiopaque markers of different shapes and size (circle on day 1, semi-cylinder on day 2, dot on day 3 and cylinder on day 4) were ingested by the volunteers on four consecutive days. X-ray of the abdomen was taken on day 5 and day 7 to assess the mouth to anal transit and segmental colon transit. Transit in the right, left and rectosigmoid colon was calculated by adding all markers seen in these regions on days 5 and 7. Slow total colonic transit was defined as >67 h, the mean transit plus 2 standard deviations averaged from published studies.16, 17


Sample size

Assuming the responder rate in the treatment group was 43.2% and placebo group was 25.1%,12 107 patients per arm would be sufficient for achieving 80% of power to detect treatment difference, at a significance level of 0.05%. The calculation was done by nQuery Advisor® software (Statistical Solutions, Saugus, MA, USA).

Primary efficacy variable

The primary efficacy variable was the responder rate for CSBM during the first 4 weeks of treatment. Patients with a mean increase of CSBM ≥1/week compared with the last 14 days of baseline were defined as responders, provided that they had completed at least 7 days of treatment.

Secondary efficacy variables

These included the change from baseline in scores for individual constipated symptoms (stool form, straining scores, bothersomeness of constipation, abdominal distension/bloating, and abdominal pain/discomfort, and satisfaction of bowel habit). Days of laxatives used and percentage of patients needed laxatives were assessed. Quality of life before and after treatment by SF-12 was assessed.

Statistical analysis was performed using SPSS (SPSS, Chicago, IL, USA) and JMP (SAS, Cary, NC, USA) statistical software. Demographics for patients and controls were summarized by calculating means (s.d.) [or median (range)] for continuous variables (e.g. age and severity score) and proportions for categorical variables (e.g. sex). Comparisons were performed with the Student's t-test for continuous variables and with chi-squared test for categorical data. All statistical tests were two-sided and P-value of <0.05 was considered to indicate a statistically significant difference.


Demographic data

There were 293 patients screened (43 were excluded because 38 withdrew consent, three were to follow-up and two no longer required treatment) (Figure 1). Two hundred and fifty patients were randomized and 216 (86.4%) completed the 8-week treatment (109 from the treatment group and 107 from the placebo group). In the treatment group, 10 patients dropped out from because of unsatisfactory therapeutic effect, six were lost to follow-up. In the placebo group, 15 patients dropped out and three were lost to follow-up (Figure 1). The demographic data, baseline constipation symptoms and colonic transit time are summarized in Table 1. Although there was statistical difference in age between the treatment and placebo groups, the difference was only 4 years. There was no difference in the constipation symptoms or the mean colonic transit time in the two groups at baseline (Table 1).

Figure 1.

 Summary of patient disposition by treatment group.

Table 1.   Demographic data and constipation symptoms of the treatment group and placebo group
 Treatment group n = 125Placebo group n = 125 P-value
Sex (female)92%89%0.39
Female to male ratio11.5:18:1 
Ethnicity (Chinese)100%100% 
Age (years ± s.d.)43.9 ± 1247.1 ± 120.04
Duration of constipation (years ± s.d.)15.7 ± 1416.7 ± 130.6
Complete spontaneous bowel motion per week (mean ± s.d.)1.9 ± 21.9 ± 20.9
Passage of hard stool80%83.9%0.6
Incomplete evacuation84%86%0.6
Slow colonic transit35.8%40.2%0.5
Mean total colonic transit time (h)51.7 ± 2956.2 ± 280.22

Primary efficacy

The percentage of patients with mean increase of CSBM ≥1/week was 47.7% and 29%, respectively, in the treatment and placebo groups (P = 0.005), while the intention to treat analysis was 41.6% and 24.8%, respectively (P = 0.005). The treatment group also has a higher rate of sustained CSBM ≥3/week over the 8-week treatment period (29.4% vs. 15.7%, P = 0.016).

Secondary efficacy

The scores for stool form, straining, bothersomeness of constipation, abdominal distension/bloating, and abdominal pain/discomfort, and satisfaction of bowel habit are summarized in Table 2. The scores for Bristol stool scale, bothersomeness of constipation and abdominal distension/bloating and satisfaction of bowel habit were significantly lower in the treatment group. Despite the number of days of laxatives used did not show any difference in the two groups, the percentage of patients needed laxatives as rescue therapy was lower in the treatment group (Table 2). In addition, the mental component summary score of the SF 12, but not the physical component summary score, was higher in the treatment group which indicated a better response (Table 2).

Table 2.   Secondary efficacy variables for the treatment and placebo groups over the 8-week treatment
 Treatment group, n = 109Placebo group, n = 107 P-value
Bristol stool scale (mean ± s.d.)
 Baseline3.3 ± 1.92.9 ± 1.90.13
 Treatment period4.1 ± 2.73.2 ± 1.70.003
Straining (mean ± s.d.)
 Baseline0.5 ± 0.50.5 ± 0.50.9
 Treatment period0.6 ± 0.50.7 ± 0.50.2
Bothersomeness of constipation (mean ± s.d.)
 Baseline2.4 ± 1.22.5 ± 1.40.8
 Treatment period2.1 ± 1.12.5 ± 0.90.02
Bothersomeness of abdominal distension/bloating (mean ± s.d.)
 Baseline2.4 ± 0.92.3 ± 1.10.8
 Treatment period2.1 ± 1.22.4 ± 1.10.04
Bothersomeness of abdominal pain/discomfort (mean ± s.d.)
 Baseline2.2 ± 1.52.1 ± 1.30.7
 Treatment period1.8 ± 1.22.1 ± 1.20.07
Satisfaction of bowel habit (mean ± s.d.)
 Baseline2.6 ± 0.92.5 ± 0.70.8
 Treatment period2.2 ± 12.6 ± 0.90.001
Number of days taking laxatives(rescue therapy)/week (mean ± s.d.)
 Baseline2.0 ± 2.61.7 ± 2.40.25
 Treatment period0.8 ± 1.61.1 ± 1.70.17
Percentage of patients taking laxatives
 Treatment period43%58.1%0.02
Physical component score (mean ± s.d.)
 Baseline46.7 ± 846.2 ± 90.6
 Treatment period46.4 ± 847.3 ± 80.4
Mental component score (mean ± s.d.)
 Baseline44.1 ± 842.2 ± 110.1
 Treatment period46.8 ± 943.6 ± 100.01

Safety and side effects

There was no serious side effect reported. In there treatment group, there were two patients (1.8%) reporting ankle oedema, two (1.8%) had vomiting, seven (6.4%) had belching and bloating, and one (0.9%) had haemorrhoids, while in the placebo group, there were two patients (1.9%) with haemorrhoids, one (1.9%) with palpitation, three (2.8%) with belching and bloating, and three (2.8%) with diarrhoea. The overall side effect prevalence was 11% in the treatment group vs. 8.4% in the placebo group (P = 0.5).


The current study demonstrated efficacy of tegaserod in Chinese constipated patients, with significant improvement in CSBM, decrease laxatives used, improved stool form, decrease bothersomeness of constipation and abdominal distension/bloating, and improved satisfaction of bowel habit. More importantly the mental score of SF-12, which reflects the quality of life, was shown to improve significantly after treatment.

On the other hand, it is interesting to note that the responding rate in the treatment group, 47.7%, in our study, is higher when compared with 43.2% by Johanson et al.12 and 40.2% by Kamm et al.13 In addition, the sustained responder rate (CSBM ≥3/week over the 8-week treatment period) is also higher in our patients than that by Kamm et al. and Johanson et al. (29.4% vs. 25.2% vs. 22.0%, respectively). Similar is also true for the stool form scale: 4.1 vs. 3.5 vs. 3.1. Although the differences between the Asian and Western results do not reach statistical significance, this may be due to the relatively small sample in our study. We estimate that if the sample size in our study is similar to that by Kamm et al. and Johanson et al., a significant statistical difference will be observed. More importantly, similar trend have also been observed in the tegaserod study in constipation predominant irritable bowel syndrome in the Asian pacific region by Kellow et al.18 They observed a responder rate of 56% vs. that of 46.3% in the European study19 and 40.5% in a US study,20 showing that the respond rate in the Asian Pacific region is also considerably higher in the Western population.

The above observations suggested that tegaserod is effective in Chinese population with chronic idiopathic constipation. In addition, the efficacy may be better than that observed in Western population. The age and sex distribution of the patients, and the duration of constipation was similar in our study to that by Kamm et al., but a better efficacy was observed in ours. We therefore postulated that this may be due to the fact that the Chinese subjects had a milder degree of constipation; the efficacy may also be partially related to the type of diet intake, or that the underlying aetiology for chronic constipation may be different in different populations. The percentage of patients with slow constipation (35–40%) in our study is comparable to those reported previously.21 However, there was no direct comparison of the mean colonic transit time between the Chinese and Western constipated subjects. The transit time in the Chinese constipated subjects may, in fact, be less prolonged than the Western. In addition other genetic factors such as serotonin transporter gene polymorphism may also be involved. The presence of 5-HTTLPR allele S has been reported to be associated with slow colonic transit in the Chinese population.22 While tegaserod is a serotonin receptor agonist, this might explain the enhanced response in the Chinese. Unfortunately, no similar studies have been performed in the Western population.

Our study showed that tegaserod therapy is effective in Chinese patients with functional constipation. Further studies may be needed to compare the response between Chinese and Caucasian patients.


Declaration of funding interests: the study was partially funded by Novartis Pharmaceuticals (HK) Ltd (CHTF919EHK01).