Effectiveness of Sterilized Symbiotic Drink Containing Lactobacillus helveticus Comparable to Probiotic Alone in Patients with Constipation-Predominant Irritable Bowel Syndrome

This prospective, double-blind randomized controlled trial was conducted from January 2016 to July 2017. Patients fulfilling the Rome III IBS-C criteria were recruited from the Gastroenterology clinic and Endoscopy Unit at Universiti Kebangsaan Malaysia Medical Centre, Kuala Lumpur, Malaysia. Sample size calculation was performed according to a study by Min et al. on assessing IBS-C symptoms improvement [23]. Sixty-three pairs of subjects were required to be studied to be able to reject the null hypothesis that this response difference is zero with probability (power) of 0.8. The type I error probability associated with this test of this null hypothesis is 0.05. Therefore, a total of 160 subjects were recruited for this study with equal number on both arms (80 subjects) to be able to reject the null hypothesis for both designs of the study. The sample size was calculated based on Power and Sample Size (PS3) (version 3.1.2, Nashville, USA) [16, 24].

The exclusion criteria included age less than 18 years old, constipation due to other medical illnesses such as diabetes mellitus, inflammatory bowel disease, hypothyroidism, colorectal carcinoma, neurological diseases, major depression, pregnancy, breastfeeding women, and patients who were on chronic opioids or anti-depressants. Subjects must not have consumed any antibiotics, probiotics, prebiotics, symbiotics, and/or laxatives less than 2 weeks before recruitment. Lactose-intolerant patients were also excluded from this clinical trial through identification by clinical assessments during IBS screening. Our center did not perform hydrogen breath test to exclude lactose intolerance. Lactose intolerance is suggested when there is a resolution of symptoms with the elimination of lactose-containing food products and resumption of symptoms when lactose-containing food products reintroduction [25, 26].

Simple randomization was performed using a random number table. The subjects were randomized into two groups: Group A or symbiotic group, which received 350 mL of daily sterilized probiotic with L. helveticus and added 5.85 g polydextrose, and Group B or probiotic group, which received 350 mL of daily sterilized probiotic with L. helveticus without polydextrose for a total of 7 days (Fig. 1). During the period of study, subjects were advised not to take any laxatives, prebiotic supplement, yogurt, or other probiotic during the 1-week study period.

At the point of pre-consumption and upon completion of 7-day consumption, the symptoms and intestinal transit time were assessed with the collected stool samples. The subjects were reminded by phone to ensure compliance and empty bottles of probiotic were returned after completion.

Test products were packaged, labeled, and blinded to investigators and subjects. Simple randomization was performed to allocate subjects into two separate groups: Group A and Group B. For each bottle of 350 mL of sterilized probiotic, it contained L. helveticus with or without added 1.5 g/100 mL of polydextrose. A 0.5 g extra of polydextrose is added by taking into consideration the degradation compensation and preparation loss. The amount of L. helveticus is minimal as all probiotics were sterilized. Seven bottles were given to each subject. For both products, they were identical in weight, color, smell, taste, and package. These are the ingredient list: water, sugar, skimmed milk powder (cow), stabilizers (polydextrose (ins 1200 and ins 440), fermented milk [water, acidity regulator (ins 330), skimmed milk powder (cow), and lactobacillus], acidity regulators (ins 270 and ins 331 (iii)), soybean fiber, and flavoring. One serving contained 350 mL of cultured drink. For each 350 mL, there are 203 kcal/853 kJ energy, 45.5 g carbohydrate, 36.8 g total sugar, 36.8 g sucrose, 5.3 g lactose, 3.5 g protein, 0.0 g fat, 5.8 g dietary fiber, and 69 mg sodium. All doctors, research staff, and subjects involved were unaware of the treatment administered to the subject.

Red carmine capsule, a non-toxic food colorant, was used to determine the ITT [27]. Following consumption, there was a red discoloration of the stool. The ITT was determined from the time of ingestion to the first stool discoloration. Upon completion of 7 days of the test products, the subjects were required to repeat the ITT test.

Stool samples were collected twice; once prior to consumption and subsequent collection upon first defecation after consumption of the last treatment drink. Each subject was asked to fill 20 mL of fresh stool sample in a standardized stool container. The changes in fecal weight in both groups after the intervention were calculated, and the significance of the changes in fecal weight between the groups was analyzed. The volume of 20 mL was standardized and preferred due to the difficulty in collecting the total stool for the whole day and logistic reasons. Fecal weight was measured in gram (g), while pH was measured utilizing litmus pH paper.

To objectively assess constipation-related symptoms, we used Garrigues constipation questionnaires [28]. There are 21 components in the questionnaires, which include passing of hard stool, straining during defecation, incomplete emptying sensation feeling after bowel movement, blockage feeling in the anus, the need of pressing around perineum to assist defecation, spending more than 10 min in the toilet to defecate, and the number of bowel movements per week. All questions were graded as never, sometimes (less than 25% of time), often (more than 25% of time), or always.

Analysis was performed using IBM SPSS Statistics version 23 (IBM Corp., New York, USA), and p < 0.05 was considered to be statistically significant. For fecal pH and fecal weight, paired t test was used pre- and post-intervention separately based on intervention groups. General linear model with repeated measures was used for the statistical analyses pre- and post-intervention between the different groups for fecal pH and fecal weight. The fecal ITT was not normally distributed; therefore, Wilcoxon rank-sum test was used separately according to intervention group. Mann–Whitney test was also used for statistical analysis between different intervention groups for ITT assessment.

Initially, a total of 450 subjects were evaluated from clinic list. A total of 172 subjects fulfilled the diagnostic criteria for IBS-C. Out of 172 subjects, five declined participation while four subjects defaulted follow-up. A total of 163 subjects completed the study with 79 in Group A (symbiotic) and the remaining 84 subjects in Group B (probiotic). There were four subjects who withdrew from the study; one subject was due to intake of a different probiotic, and three of the subjects were lost to follow-up. These results are presented in Fig. 1.

The majority of IBS-C subjects (78.6%) were women then followed by men (21.4%). Three major ethnicities were recruited: Malays (73.1%), Chinese (24.5%), and Indians (1.2%). The median age for Group A was 34 years old, whereas for Group B was 27 years old. Median age was used as the subjects were not equally distributed. Based on the educational levels of the subjects, a higher percentage (74.9%) received tertiary level of education. However, 45.4% had low income of less than ringgit Malaysia (RM) 1500 per month and 54.6% had income more than RM 1500 per month with no correlation with their respective education level. In terms of social status, 96.3% were non-smokers and 99.4% never consumed alcohol. Dietary history among the recruited subjects showed 62% of them consumed low-fiber diet, 35% had moderate fiber intake, and only 3.1% had high-fiber intake. Pertaining to their lifestyles, 54% claimed that they sometimes had physical activity in their daily lives, 12.3% habitually performed exercise, and 33.7% did not actively exercise. Demographic and socioeconomic details are shown in Table 1. Mean bowel movement increased from 2.04 ± 2.25 to 4.90 ± 2.70 in Group A and from 2.19 ± 2.06 to 5.30 ± 2.30 in Group B.

The mean baseline fecal pH was nearly equal in both groups. Consumption of symbiotic and probiotic reduced fecal pH significantly. Fecal pH was normally distributed because of its skewness and kurtosis within ± 1. Median and mean value were similar. Most of the dots were on the line based on Q–Q Plot. Fecal pH for Groups A and B was reduced from 6.57 ± 0.96 to 6.13 ± 0.95 (p = 0.003) and 6.58 ± 1.0 to 5.87 ± 0.83 (p < 0.001), respectively. However, there was no statistical significant difference when compared between the two groups (p = 0.179) (Fig. 2).

Subjects in both groups at baseline had longer ITT with marked variations between individuals (33–117 h) [29]. Post-intervention showed a significantly shorter transit time for both groups. The ITT was significantly reduced with median of 58 h (IQR = 43–72) reduced to 45 h (IQR = 24–59) and median of 48 h (IQR = 31–72) reduced to 30 h (IQR = 24–49) for Groups A and B, respectively (p < 0.01). However, there was no statistical significant difference between the two groups (p = 0.262). The results are summarized in Table 2.

The baseline mean for fecal weight was 8.0 ± 6.4 g in Group A and 13.3 g ± 19.4 g in Group B. After a week consumption of symbiotic, the mean fecal weight increased to 9.8 g ± 7.6 g (p = 0.003) in Group A. However, mean fecal weight reduced to 11.2 g ± 6.6 g (p = 0.308) in Group B. Fecal weight was normally distributed because of its skewness and kurtosis within ± 1. Median and mean value were similar. Most of the dots were on the line based on Q–Q plot. The results were not statistically significant between the two groups (p = 0.076) (Fig. 3).

Clinical responses were assessed using the modified Garrigues constipation questionnaires [28]. IBS-C-related symptoms for Groups A and B improved at the end of the study, and no major adverse event was reported in both groups. Of particular interest were the changes in constipation, incomplete evacuation, passing on hard stool, and spending more than 10 min to evacuate, and straining symptoms were reported with the highest frequency at baseline. For each of these symptoms, the relative decrease in symptom frequency was approximately twofold greater in the post-intervention group compared to baseline. Pre-intervention, all subjects in Groups A and B have constipation. Post-intervention, 81% of subjects in Group A significantly improved constipation, while in Group B, 84% claimed that they have no constipation (p < 0.05). In terms of straining during defecation, there was reduction from 91 to 56% versus 77 to 48% in Groups A and B, respectively. For incomplete evacuation, there was reduction from 84 to 56% versus 93 to 47% in Groups A and B, respectively. Similar improvements were observed for passing on hard stool and the need of spending more than 10 min to defecate. There was a reduction from 97 to 66% versus 90 to 64% in Groups A and B, respectively, in terms of having hard stool. Post-intervention showed a reduction in number of subjects spending more than 10 min for complete defecation (85–69% in Group A and 52–43% in Group B). Similar improvements were observed for other constipation-related symptoms post-intervention including fewer subjects having anal blockage sensation or the need of pressing perineum to defecate at the end of the study. Results are shown in Figs. 4 and 5. However, there was no statistical significance between these two groups (p = 0.67).

All subjects were compliant to the study protocol. Adverse events were mild with the most frequent adverse event being loose stool, reported in 22 of the subjects (27.8%) in Group A and 18 subjects (21.4%) in Group B, which may be seen as a beneficial effect. Another reported adverse event was mild abdominal discomfort, reported in one subject (1.3%) in Group A and two subjects (2.8%) in Group B. The abdominal discomfort did not require any medical intervention or hospitalization.