We searched MEDLINE, Embase and CENTRAL from inception to 25 July 2018. We also searched references, trials registers, and conference proceedings.
SELECTION CRITERIA: Fifteen studies (547 participants) were included. Four studies assessed treatment of acute
pouchitis. Five studies assessed treatment of chronic
pouchitis. Six studies assessed prevention of
pouchitis. Three studies were low risk of bias. Three studies were high risk of bias and the other studies were unclear.Acute
pouchitis: All
ciprofloxacin participants (7/7) achieved remission at two weeks compared to 33% (3/9) of
metronidazole participants (RR 2.68, 95% CI 1.13 to 6.35, very low certainty evidence). No
ciprofloxacin participants (0/7) had an AE compared to 33% (3/9) of
metronidazole participants (RR 0.18, 95% CI 0.01 to 2.98; very low certainty evidence). AEs included
vomiting,
dysgeusia or transient
peripheral neuropathy. Forty-three per cent (6/14) of
metronidazole participants achieved remission at 6 weeks compared to 50% (6/12) of
budesonide enema participants (RR 0.86, 95% CI 0.37 to 1.96, very low certainty evidence). Fifty per cent (7/14) of
metronidazole participants improved clinically at 6 weeks compared to 58% (7/12) of
budesonide enema participants (RR 0.86, 95% CI 0.42 to 1.74, very low certainty evidence). Fifty-seven per cent (8/14) of
metronidazole participants had an AE compared to 25% (3/12) of
budesonide enema participants (RR 2.29, 95% CI 0.78 to 6.73, very low certainty evidence). AEs included
anorexia,
nausea,
headache,
asthenia,
metallic taste,
vomiting, paraesthesia, and depression. Twenty-five per cent (2/8) of
rifaximin participants achieved remission at 4 weeks compared to 0% (0/10) of placebo participants (RR 6.11, 95% CI 0.33 to 111.71, very low certainty evidence). Thirty-eight per cent (3/8) of
rifaximin participants improved clinically at 4 weeks compared to 30% (3/10) of placebo participants (RR 1.25, 95% CI 0.34 to 4.60, very low certainty evidence). Seventy-five per cent (6/8) of
rifaximin participants had an AE compared to 50% (5/10) of placebo participants (RR 1.50, 95% CI 0.72 to 3.14, very low certainty evidence). AEs included
diarrhea,
flatulence,
nausea, proctalgia,
vomiting, thirst, candida,
upper respiratory tract infection, increased hepatic
enzyme, and
cluster headache. Ten per cent (1/10) of Lactobacillus GG participants improved clinically at 12 weeks compared to 0% (0/10) of placebo participants (RR 3.00, 95% CI 0.14 to 65.90, very low certainty evidence).Chronic
pouchitis: Eighty-five per cent (34/40) of De Simone Formulation participants maintained remission at 9 to 12 months compared to 3% (1/36) of placebo participants (RR 20.24, 95% CI 4.28 to 95.81, 2 studies; low certainty evidence). Two per cent (1/40) of De Simone Formulation participants had an AE compared to 0% (0/36) of placebo participants (RR 2.43, 95% CI 0.11 to 55.89; low certainty evidence). AEs included
abdominal cramps,
vomiting and
diarrhea. Fifty per cent (3/6) of
adalimumab patients achieved clinical improvement at 4 weeks compared to 43% (3/7) of placebo participants (RR, 1.17, 95% CI 0.36 to 3.76, low certainty evidence). Sixty per cent (6/10) of
glutamine participants maintained remission at 3 weeks compared to 33% (3/9) of
butyrate participants (RR 1.80, 95% CI 0.63 to 5.16, very low certainty evidence). Forty-five per cent (9/20) of patients treated with
bismuth carbomer foam
enema improved clinically at 3 weeks compared to 45% (9/20) of placebo participants (RR 1.00, 95% CI 0.50 to 1.98, very low certainty evidence). Twenty-five per cent (5/20) of participants in the
bismuth carbomer foam
enema group had an AE compared to 35% (7/20) of placebo participants (RR 0.71, 95% CI 0.27 to 1.88, very low certainty evidence). Adverse events included
diarrhea, worsening symptoms, cramping,
sinusitis, and
abdominal pain.
PREVENTION: At 12 months, 90% (18/20) of De Simone Formulation participants had no episodes of acute
pouchitis compared to 60% (12/20) of placebo participants (RR 1.50, 95% CI 1.02 to 2.21, low certainty evidence). Another study found 100% (16/16) of De Simone Formulation participants had no episodes of acute
pouchitis at 12 months compared to 92% (11/12) of the no treatment control group (RR 1.10, 95% 0.89 to 1.36, very low certainty evidence). Eighty-six per cent (6/7) of Bifidobacterium longum participants had no episodes of acute
pouchitis at 6 months compared to 60% (3/5) of placebo participants (RR 1.43, 95% CI 0.66 to 3.11, very low certainty evidence). Eleven per cent (1/9) of Clostridium butyricum MIYAIRI participants had no episodes of acute
pouchitis at 24 months compared to 50% (4/8) of placebo participants (RR 0.22, 95% CI 0.03 to 1.60, very low certainty evidence). Forty-six per cent (43/94) of
allopurinol participants had no episodes of
pouchitis at 24 months compared to 43% (39/90) of placebo participants (RR 1.06, 95% CI 0.76 to 1.46; low certainty evidence). Eighty-one per cent (21/26) of
tinidazole participants had no episodes of
pouchitis over 12 months compared to 58% (7/12) of placebo participants (RR 1.38, 95% CI 0.83 to 2.31, very low certainty evidence).
AUTHORS' CONCLUSIONS: