The origins of the strains used in this study are shown in Table 1. For isolated of Bifidobacteria, fecal samples of healthy Koreans (20-30 years old) were collected by BBL's anaerobic sample collection and transport system to maintain anaerobic conditions, and were used within 24 h. Fecal samples were serially diluted 10-fold from 10^-1 to 10^-8, and 100 μl was spread onto selective BL (Blood Liver) (Nissui Pharm. Co. Ltd., Japan) agar containing 5% sheep blood. After 48 hr of incubation in anaerobic conditions (90% N₂, 5% H₂, 5% CO₂) (Bactron Anaerobic Chamber, Sheldon Manufacturing Inc., USA) at 37°C, brown or reddish-brown colonies 2-3 mm in diameter were selected for further identification 13 .

A fructose-6-phosphate phosphoketolase (F6PPK) test was performed 14 to ensure that the colonies selected were Bifidobacteria. To identify the isolated Bifidobacterium spp. at the species level, 16S rRNA sequencing was performed by Bioleaders (Daejeon, Korea).

All the participants were recruited from the Eden Adventist Hospital in Gyeonggi-do, Korea. Male and female nursing home residents with chronic constipation, and who were presently receiving intervention (eg. laxatives, enemas and other invasive procedures, such as manual removal of fecal impaction) were evaluated for enrolment in the study. Because a significant number of the study participants were nursing home residents, it was not always feasible to obtain a diagnosis according to "Rome II" criteria for constipation as a prerequisite for the inclusion criteria. Such a diagnosis would require a level of communication with the patient that was not always possible in this study. The exclusion criteria were participants with ileus, renal failure, dialysis, Crohn's disease, ulcerative colitis and chronic abdominal pain. Thus, we screened 25 participants and included 8 male and 11 female participants that chose to participate in the study.

This study was designed to assess the effects of LAB (3.0 × 10¹¹ CFU/g) in the regimen of nursing home residents suffering from chronic constipation, the dispensing of standard treatment by nursing home staffs and on costs of care and medications. Participants with chronic constipation were defined as residents who used laxatives at least once a week.

The study protocol and the informed consent forms were reviewed and approved by the Eden Adventist Hospital. Each participant, or his or her legal guardian, was informed both orally and in writing. Written informed consent was obtained from all participants, or from their legal guardians, before participation in the study.

Subjects were instructed on the aim and content of the trial, as well as test methods. All 19 subjects were assigned to receive LAB twice (to be taken 30 minutes after breakfast and dinner) a day for 2 weeks in November 2008. During the trial period, the participants were prohibited from administering their existing laxative medications and/or enemas in addition to the investigational product.

Before entering the trial, the subjects completed a questionnaire on name, age, sex and current defecation habits such as frequency of defecation in four steps from 'once more than three days' to 'more than twice a day', amount of stool in three steps from 'large' to 'small', state of stool in five steps from 'like stone' to 'like water' and yes or no about 'currently taking any medicine for defecation' (Appendix 1).

Subjects were observed for any untoward symptoms such as vomiting, diarrhea or abdominal pain. The nursing home staff recorded a defecation habit diary during the study and this included information such as frequency of defecation as well as amount and state of stool passed by the subjects.

After ending the trial, the subjects draw up a questionnaire on defecation habits.

We collected fecal samples from the subjects both before entering and after ending the trial, to investigate LAB levels and harmful enzyme activity.

Fecal samples (0.1 g) were suspended in 0.9 ml of 0.1 M phosphate buffer (pH 6.8 containing 0.5% cysteine) using a vortex, and 0.1 ml was then serially diluted 10-fold from 10^-1 to 10^-7. 1 ml was then poured into selective MRS broth (pH 7.0) (Difco, USA). After 48 h of incubation under anaerobic conditions (90% N₂, 5% H₂, 5% CO₂) (Bactron Anaerobic Chamber, Sheldon MFG. Inc., USA), colonies were counted as LAB 12 . The numbers of colony forming units (CFU) are expressed as log10 CFU per gram.

Harmful enzyme activities such as β-glucosidase, β-glucuronidase, tryptophanase, and urease of intestinal microflora related to colon cancer were tested in human fecal samples as previously described 15 16 17 .

β-glucosidase activity was assayed using 2 ml of a reaction mixture containing 0.8 ml of 2 mM p-nitrophenyl-β-D-glucopyranoside and 0.2 ml of the enzyme solution (suspended fecal sample), incubated for 30 min at 37°C, and then stopped by adding 1 ml of 0.5 N NaOH. The reaction mixture was then centrifuged at 3,000 rpm for 10 min. Enzyme activity was measured by monitoring absorbance at 405 nm.

β-glucuronidase activity was assayed using 2 ml of a reaction mixture consisting of 0.8 ml of 2 mM p-nitrophenyl-β-D-glucuronide and 0.2 ml of the enzyme solution, incubated for 30 min at 37°C, and then stopped by adding 1 ml of 0.5 N NaOH. The reaction mixture was centrifuged at 3,000 rpm for 10 min. Enzyme activity was measured by monitoring absorbance at 405 nm.

Tryptophanase activity was assayed using 2.5 ml of a reaction mixture consisting of 0.2 ml of complete reagent solution (2.75 mg of pyridoxal phosphate, 19.6 mg of disodium EDTA dihydrate, and 10 mg of bovine serum albumin in 100 ml of 0.05 M potassium phosphate buffer, pH 7.5), 0.2 ml of 20 mM tryptophan, and 0.1 ml of the enzyme solution, incubated for 1 h at 37°C, and then stopped by adding 2 ml of color reagent solution (14.7 g p-dimethylaminobenzaldehyde in 52 ml H₂SO₄ and 948 ml 95% ethanol). The reaction mixture was then centrifuged at 3,000 rpm for 10 min. Enzyme activity was measured by monitoring absorbance at 550 nm.

Urease activity was assayed using 0.5 ml of a reaction mixture consisting of 0.3 ml of urea substrate solution (4 mM urea in 20 mM sodium phosphate buffer, pH 7.0) and 0.1 ml of the enzyme solution, incubated for 30 min at 37°C and then stopped by adding 0.1 ml of 1 N (NH4)₂SO₄. Phenolnitroprusside reagent (1 ml) and alkaline hypochlorite reagent (NaClO, 1 ml) were added to the stopped reaction mixture and incubated for 20 min at 65°C. The reaction mixture was centrifuged at 3,000 rpm for 10 min. Enzyme activity was measured by monitoring absorbance at 603 nm.

The data were collected and analyzed independently of the investigators, who did not have access to it or to its analysis. Analysis data of questionnaire were processed using the SAS, and comparisons data of fecal LAB levels and harmful enzyme activity of intestinal microflora were analyzed using the unpaired Student's t-test. Differences were considered statistically significant at p < 0.05.

This study tested 19 subjects with a mean age of 77.1 and suffering from chronic constipation; 8 males of mean age 77.4 and 11 females of mean age 76.9. Of these subjects, only 15 subjects (79%) were currently receiving intervention such as laxatives and/or enemas: 8 males (100%) and 7 females (64%) (Table 2).

All 19 subjects were assigned to receive LAB (3.0 × 10¹¹ CFU/g) twice (to be taken 30 minutes after breakfast and dinner) a day for 2 weeks, and answered their defecation habits as follows; frequency of defecation: 11 individuals reported 'once more than three days' (57.9%), 3 'once two days' (15.8%), 4 'once a day' (21.1%) and 1 'more than twice a day' (5.3%) before LAB treatment; 6 'once more than three days' (31.6%), 7 'once two days' (36.8%), 6 'once a day' (31.6%) and 0 'more than twice a day' (0.0%) after LAB treatment; amount of stool: 4 individuals reported 'large' (21.1%), 11 'medium' (57.9%) and 4 'small' (21.1%) before LAB treatment; 8 'large' (42.1%), 9 'medium' (47.4%) and 2 'small' (10.5%) after LAB treatment; state of stool: 1 individual reported 'like stone' (5.3%), 5 'hard' (26.3%), 12 'soft' (63.2%), 1 'watery' (5.3%) and 0 'like water' (0.0%) before LAB treatment; 0 'like stone' (0.0%), 6 'hard' (31.6%), 11 'soft' (57.9%), 2 'watery' (10.5%) and 0 'like water' (0.0%) after LAB treatment (Table 3).

According to the defecation habit diary recorded by the nursing home staff, frequency of defecation (p = 0.676), amount of stool ('large'; p = 1, 'medium'; p = 0.664 and 'small'; p = 0.289) and state of stool ('like stone'; p = 0, 'hard'; p = 0.109, 'soft'; p = 0.365, 'watery'; p = 0.377 and 'like water'; p = 0) were not statistically significant (Table 4).

Fecal LAB levels were significantly increased from 4.4 log10 CFU/g to 7.3 log10 CFU/g after LAB treatment (p = 0.024) (Fig. 1.).

The harmful enzyme activities of intestinal microflora were shown in Table 5. After LAB treatment, tryptophanase and urease activities were decreased by 43% and 30%, respectively. Furthermore, there was a statistically significant decrease in both tryptophanase (p = 0.047) and urease (p = 0.005) activities. However, β-glucosidase and β-glucuronidase activities were increased, and there was not a statistically significant increase in both β-glucosidase (p = 0.074) and β-glucuronidase (p = 0.061) activities.

The following list shows the questions about currently your defecation habits.

1. Frequency of defecation?

①Once more than three days ②Once two days ③Once a day ④More than twice a day

2. Amount of stool?

①Large ②Medium ③Small

3. State of stool?

①Like stone ②Hard ③Soft ④Watery ⑤Like water

4. Are you currently take any medicine for defecate the stool?

①Yes (What is it?) ②No

The following list shows the questions about your defecation habits after LAB treatment.

1. Frequency of defecation?

①Once more than three days ②Once two days ③Once a day ④More than twice a day

2. Amount of stool?

①Large ②Medium ③Small

3. State of stool?

①Like stone ②Hard ③Soft ④Watery ⑤Like water