2. Probiotics vs placebo for prevention of diarrhoea induced by chemotherapy.

Study ID and participants	Intervention(s)	Results
Chen 2014 Participants with colorectal cancer undergoing intravenous chemotherapy	Intervention: combined Clostridium butyricum and Bifidobacterium capsule (n = 35) Control: placebo (n = 35) Both interventions (3 capsules, 3 times a day) were administered from 5 days before surgery for colorectal cancer to 7 days after surgery in each group. All participants received intravenous chemotherapy (calcium folinate 300 mg, fluorouracil 500 mg) during surgery	Proportion of participants with diarrhea: 6/30 versus 12/30 (RR 0.50, 95% CI 0.22 to 1.16) Quality of life: not assessed Severity of diarrhoea: not assessed Time to rescue medication for diarrhoea: not assessed Proportion of participants requiring rescue medication for diarrhoea: not assessed Adverse events: abdominal distension: 18/30 versus 11/30 (RR 1.64, 95% CI 0.94 to 2.85) Systemic inflammatory response syndrome (SIRS): 7/30 versus 9/30 (RR 0.78, 95% CI 0.33 to 1.82) Infection of incisional wound: 3/30 versus 3/30 (RR 1.0, 95% CI 0.22 to 4.56) Pulmonary infection: 2/30 versus 2/30 (RR 1.0, 95% CI 0.15 to 6.64) Urinary tract infection: 2/30 versus 2/30 (RR 1.0, 95% CI 0.15 to 6.64) Duration of fever (days): 4.5 ± 1.0 versus 4.6 ± 1.2 (MD ‐0.10 days, 95% CI ‐0.66 to 0.46) Hypoproteinaemia: 5/30 versus 4/30 (RR 1.25, 95% CI 0.37 to 4.21) Side effects relevant to drug: 0/30 versus 0/30 Mortality: study authors reported no deaths
Liu 2000 Participants with cancer of the lung, stomach, colon, rectum, or breast or with metastatic neck carcinoma who were to receive chemotherapy	Cross‐over study with 22 participants Intervention: Bifidobacterium combined with chemotherapy Control: chemotherapy alone Bifidobacterium capsule (2 capsules per time, 2 times a day) was taken from 1 day before chemotherapy to the sixth day of chemotherapy in each phase. Length of the washout period in this cross‐over study was about 21 days	Proportion of participants with diarrhoea: 6/22 versus 10/22 (no paired analysis presented) Quality of life: not assessed Severity of diarrhoea: Grade 2+: 3/22 versus 7/22 (no paired analysis presented) Grade 3+: 1/22 versus 4/22 (no paired analysis presented) Time to rescue medication: not assessed Proportion of participants requiring rescue medication for diarrhoea: not assessed Adverse events: not assessed Mortality: study authors reported no deaths
Mego 2015 Participants with colorectal cancer starting a new line of chemotherapy	Intervention: probiotic formula Colon Dophilus™ 3*1 capsule per day orally for 12 weeks "Each capsule contained 10*109 CFU of bacteria. Each capsule contained 10 lyophilized probiotic strains including Bifidobacterium breve HA‐129 (25%), Bifidobacterium bifidum HA‐132 HA (20%), Bifidobacterium longum HA‐135 (14.5%), Lactobacillus rhamnosus HA‐111 (8%), Lactobacillus acidophilus HA‐122 (8%), Lactobacillus casei HA‐108 (8%), Lactobacillus plantarum HA‐119 (8%), Streptococcus thermopilus HA‐110 (6%), Lactobacillus brevis HA‐112 (2%), Bifidobacterium infantis HA‐116 (0.5%) (n = 23)" Control: placebo (n = 23). "Each capsule with placebo contained only inactive ingredients without probiotic bacteria, and placebo capsules were prepared by the central pharmacy. The placebo was indistinguishable from the capsule with probiotics in terms of color, appearance, taste, smell, size, shape, and other properties and contained the same additives as probiotic capsule"	Proportion of participants with diarrhoea: 9/23 versus 14/23 (RR 0.64, 95% CI 0.35 to 1.18) Quality of life: not assessed Severity of diarrhoea: Grade 2+: 4/23 versus 6/23 (RR 0.67, 95% CI 0.22 to 2.05) Grade 3+: 0/23 versus 4/23 (RR 0.11, 95% CI 0.01 to 1.95) Grade 4: 0/23 versus 1/23 (RR 0.33, 95% CI 0.01 to 7.78) Time to rescue medication for diarrhoea: not assessed Proportion of participants requiring rescue medication for diarrhoea: "participants on probiotic arm used less loperamide and diphenoxylate/atropine compared to participants on placebo arm" Adverse events: "We received filled study diaries from 38 (82.6%) of patients. ….. We did not observe any infection caused by probiotic strains used in this study" Mortality: study authors reported no deaths

CI: confidence interval.

RR: risk ratio.

SIRS: systemic inflammatory response syndrome.