Gastroenterology 2020;-:1–9

1
2
AGA SECTION 61
62
3 63
4
5
AGA Clinical Practice Guidelines on the Role of Probiotics in the 64
65
6 Management of Gastrointestinal Disorders 66
7 67
8 Q9 Grace L. Su,1,2 Cynthia W. Ko,3 Premysl Bercik,4 Yngve Falck-Ytter,5,6 Shahnaz Sultan,7 68
9 Adam V. Weizman,8 and Rebecca L. Morgan9 69
10 70
11 1 2
Division of Gastroenterology and Hepatology, University of Michigan, Ann Arbor, Michigan; Gastroenterology Section, 71
12 3
Veterans Administration Ann Arbor Healthcare System, Ann Arbor, Michigan; Division of Gastroenterology, University of 72
13 Washington Medical School, Seattle, Washington; 4Division of Gastroenterology, McMaster University, Hamilton, Ontario, 73
14 Canada; 5Division of Gastroenterology, Case Western Reserve University, Cleveland, Ohio; 74
6
15 Louis Stokes Veterans Affairs Medical Center, Cleveland, Ohio; 7Division of Gastroenterology, University of Minnesota, 75
16 Minneapolis, Minnesota; 8Division of Gastroenterology, Mount Sinai Hospital, Department of Medicine, University of Toronto, 76
9
Toronto, Ontario, Canada; and Department of Health Research Methods, Evidence and Impact, McMaster University,
17 77
Hamilton, Ontario, Canada
18 78
19 79
20 80
21
22
T his document presents the official recommendations are not considered drugs in the United States or Europe, the
of the American Gastroenterological Association regulatory status is not the same as would normally
(AGA) on the role of probiotics in the management of accompany a pharmaceutical product. The industry is
81
82
23 gastrointestinal disorders. The guideline was developed by largely unregulated and marketing of product is often 83
24 the AGA Institute’s Clinical Guidelines Committee and geared directly at consumers without providing direct and 84
25 approved by the AGA Governing Board. It is accompanied by consistent proof of effectiveness.5,6 This has led to wide- 85
26 a technical review that provides a detailed synthesis of the spread use of probiotics with confusing evidence for clinical 86
27 evidence from which these recommendations were formu- efficacy.7 It is estimated that 3.9 million American adults 87
28 lated.1 To get a better understanding of these guidelines, we used some form of probiotics or prebiotics (nutrients that 88
29 recommend reading the accompanying technical review. promote growth or beneficial functions of beneficial mi- 89
30 Development of this guideline and the accompanying tech- crobes)8 in 2015, an amount that is 4 times that in 2007.9,10 90
31 nical review was fully funded by the AGA Institute without Given widespread use and often biased sources of infor- 91
32 additional outside funding. Members of the Guideline Panel mation, it is essential that clinicians have objective guidance 92
33 and Technical Review Panel were selected by the AGA for their patients about the appropriate use of and in- 93
34 Governing Board in consultation with the Clinical Guidelines dications for probiotics. 94
35 Committee with careful consideration of all Institute of Although there has been a substantial number of studies 95
36 Medicine recommendations for clinical guideline develop- examining probiotics in various gastrointestinal diseases, 96
37 ment. A patient representative was also included in the the studies have been extremely varied, including differ- 97
38 development and review process and had no recommended ences in the strain of microbes used, dose, and route of 98
39 changes. The guideline and accompanying technical review administration, as well as the research methodology, 99
40 underwent independent peer review, and a 30-day open including differences in the reporting of end points and 100
41 public comment period; all comments were collated by the outcomes.5 Furthermore, most of the studies with probiotics 101
42 AGA staff and were reviewed and carefully considered by involved a relatively small number of patients compared to 102
43 the Guideline Panel and Technical Review teams, respec- trials investigating the effects of pharmacological in- 103
44 tively. Changes were incorporated in revised documents and terventions. Conclusions drawn from meta-analyses or 104
45 where changes were not accepted, a thoughtful response systematic reviews can be misleading if different studies 105
46 document was created. In accordance with the Clinical with different patient populations, different reported end 106
47 Guidelines Committee policies, all clinical guidelines are points and outcomes, or different strains or combinations of 107
48 reviewed annually at the AGA Clinical Guideline Committee probiotics are grouped together inappropriately.7 Within 108
AGA SECTION

49 meeting for new information. The next update for these species, different strains can have widely different activities 109
50 guidelines is anticipated in 3 years from publication (2023). and biologic effects. Many immunologic, neurologic, and 110
51 Within the last 20 years, there has been increasing biochemical effects of gut microbiota are likely to not only 111
52 recognition and interest in the role of the gut microbiome in 112
53 gastrointestinal health.2 Defined by the Food and Agricul- 113
54 ture Organization of the United Nations and the World Abbreviations used in this paper: AGA, American Gastroenterological 114
55 Health Organization as “live microorganisms which when Association; CI, confidence interval; GRADE, Grading of Recommenda- 115
tions Assessment, Development and Evaluation; IBS, irritable bowel syn-
56 administered in adequate amounts confer a health benefit drome; MD, mean difference; NEC, necrotizing enterocolitis; OR, odds 116
57 on the host,”3 probiotics hold the promise of an effective ratio; RCT, randomized controlled trial; RR, relative risk. Q1 Q2 117
58 way to alter the microbiome for our benefit. Enthusiasm and 118
59 popularity within the community for probiotics has led to a © 2020 by the AGA Institute
0016-5085/$36.00
119
60 multibillion-dollar industry worldwide.4 Because probiotics https://doi.org/10.1053/j.gastro.2020.05.059 120

PGL 5.6.0 DTD
YGAST63501_proof
6 July 2020
2:53 pm
ce

 2 Su et al Gastroenterology Vol. -, No. -

121 Table 1.Quality of Evidence treatment of gastrointestinal diseases (not prebiotic use). 181
122 Each research question identified the population, interven- 182
Quality Description tion, comparison, and patient-important outcomes. The
123 183
124 High We are very confident that the true effect lies close
Technical Review Panel initially reviewed and assessed 184
125 to that of the estimate of the effect. relevant systematic reviews that addressed the clinical 185
126 Moderate We are moderately confident in the effect estimate. questions, updating high-quality systematic reviews through 186
127
The true effect is likely to be close to the December 2018 to inform the recommendations when 187
128
estimate of the effect, but there is a possibility possible.1 For situations in which there was either no recent 188
that it is substantially different. systematic review available or the recent systematic review
129 Low Our confidence in the effect estimate is limited. The 189
130 was not deemed high quality, the Technical Review Panel 190
true effect may be substantially different from
131 the estimate of the effect. conducted the systematic review de novo. The findings from 191
132 Very Low We have very little confidence in the effect each systematic review were assessed using the GRADE 192
133 estimate. The true effect is likely to be approach and presented in an evidence profile. The Guide- 193
substantially different from the estimate of line Panel and the authors of the technical review met face
134 194
135
effect to face on May 9, 2019 to discuss the findings from the 195
136 technical review. After this meeting, the Guideline Panel 196
137 independently formulated guideline recommendations; the 197
138 Technical Review Panel was not involved in the formulating 198
139 be strain-specific, but also dose-specific.6 Furthermore, or finalizing of recommendations. Although the quality of 199
140 combinations of different microbial strains may also have evidence (Table 1) was a key factor in determining the 200
141 widely different activities, as some microbial activities are strength of the recommendations (Table 2), the Panel also 201
142 dependent on interactions between different strains. In considered the balance between the benefits and harms of 202
143 developing this guideline, we have examined the evidence the interventions, as well as patients’ values and prefer- 203
144 presented in the accompanying technical review with these ences, resource use (ie, cost), health equity, acceptability, 204
145 constraints in mind. We prioritized Clostridioides difficile– and feasibility (the Evidence to Decision Framework). The 205
146 associated diseases, inflammatory bowel disease, irritable recommendations, certainty of evidence, and strength of 206
147 bowel syndrome, infectious gastroenteritis, and necrotizing recommendations are summarized in Table 3. The guideline 207
148 enterocolitis (NEC) because these were conditions for which and technical review went through a 30-day public 208
149 probiotics were commonly considered. We focused on comment period between February 16, 2020 and March 17, 209
150 patient-important outcomes, such as induction and main- 2020. AGA staff collated the comments, the Guideline Panel 210
151 tenance of disease, treatment of disease, prevention of deliberated in their response and, when appropriate, 211
152 sepsis, and all-cause mortality. modified the document text. We hope to provide clinicians 212
153 This guideline was developed utilizing a process out- with clear guidance regarding the appropriate use of spe- 213
154 lined previously.11 Briefly, the AGA process for developing cific probiotics in the context of specific gastrointestinal 214
155 clinical practice guidelines follows the Grading of Recom- diseases. The target audience for this guideline includes 215
156 mendations Assessment, Development and Evaluation health care providers, dietitians, and patients. The guide- 216
157 (GRADE) approach11 and best practices as outlined by the lines include recommendations for specific populations, 217
158 National Academy of Science (formerly Institute of Medi- including adults, children, and neonates. 218
159 cine).12 A priori, the Guideline Panel and methodologist In addition, we were not able to assess the viability of 219
160 identified and formulated clinically relevant questions about each formulation reported in the studies, as this information 220
161 the use of probiotic formulations for the prevention and was not routinely available. We recognize that different 221
162 222
163 223
164 Table 2.Strength of Recommendation 224
165 225
166 Strength of 226
recommendation For the patient For the clinician
167 227
168 Strong Most individuals in this situation would want the Most individuals should receive the recommended 228
AGA SECTION

169 recommended course of action and only a small course of action. Formal decision aids are not 229
170 proportion would not. likely to be needed to help individuals make 230
171 decisions consistent with their values and 231
172 preferences. 232
Conditional The majority of individuals in this situation would Different choices will be appropriate for different
173 233
want the suggested course of action, but many patients. Decision aids may be useful in helping
174 would not. individuals in making decisions consistent with 234
175 their values and preferences. Clinicians should 235
176 expect to spend more time with patients when 236
177 working towards a decision. 237
178 238
179 239
180 240

PGL 5.6.0 DTD
YGAST63501_proof
6 July 2020
2:53 pm
ce

 - 2020 AGA Clinical Practice Guidelines on Probiotics and Gastrointestinal Disorders 3

241 Table 3.Summary of recommendationsa 301

242 Recommendations Strength of recommendation Quality of evidence 302
243 303
244 1. In patients with C difficile infection, we recommend the use of No recommendation Knowledge gap 304
245 probiotics only in the context of a clinical trial. 305
246 2. In adults and children on antibiotic treatment, we suggest the use Conditional Low 306
of S boulardii; or the 2-strain combination of L acidophilus
247 307
CL1285 and L casei LBC80R; or the 3-strain combination of L
248 acidophilus, L delbrueckii subsp bulgaricus, and B bifidum; or the 308
249 4-strain combination of L acidophilus, L delbrueckii subsp 309
250 bulgaricus, B bifidum, and S salivarius subsp thermophilus over 310
251 no or other probiotics for prevention of C difficile infection. 311
252 Comment: Patients who place a high value on the potential 312
253 harms (particularly those with severe illnesses) or a high value 313
on avoiding the associated cost and a low value on the small
254 risk of C difficile development (particularly in the outpatient
314
255 setting), would reasonably select no probiotics. 315
256 3. In adults and children with Crohn’s disease, we recommend the No recommendation Knowledge gap 316
257 use of probiotics only in the context of a clinical trial. 317
258 4. In adults and children with ulcerative colitis, we recommend the No recommendation Knowledge gap 318
259 use of probiotics only in the context of a clinical trial. 319
5. In adults and children with pouchitis, we suggest the 8-strain Conditional Very low
260 320
combination of L paracasei subsp paracasei DSM 24733, L
261 plantarum DSM 24730, L acidophilus DSM 24735, L delbrueckii
321
262 subsp bulgaricus DSM 24734, B longum subsp longum DSM 322
263 24736, B breve DSM 24732, B longum subsp infantis DSM 323
264 24737, and S salivarius subsp thermophilus DSM 24731 over no 324
265 or other probiotics. 325
266 Comment: Patients for whom the feasibility and cost of using 326
this combination of bacterial strain is problematic may
267 327
reasonably select no probiotics.
268 6. In symptomatic children and adults with irritable bowel syndrome, No recommendations Knowledge gap 328
269 we recommend the use of probiotics only in the context of a 329
270 clinical trial. 330
271 7. In children with acute infectious gastroenteritis, we suggest Conditional Moderate 331
272 against the use of probiotics. 332
273 8. In preterm (less than 37 weeks gestational age), low-birth-weight Conditional Moderate/high 333
infants, we suggest using a combination of Lactobacillus spp and
274 Bifidobacterium spp (L rhamnosus ATCC 53103 and B longum
334
275 subsp infantis; or L casei and B breve; or L rhamnosus, L 335
276 acidophilus, L casei, B longum subsp infantis, B bifidum, and B 336
277 longum subsp longum; or L acidophilus and B longum subsp 337
278 infantis; or L acidophilus and B bifidum; or L rhamnosus ATCC 338
279 53103 and B longum Reuter ATCC BAA-999; or L acidophilus, B 339
bifidum, B animalis subsp lactis, and B longum subsp longum), or
280 340
B animalis subsp lactis (including DSM 15954), or L reuteri (DSM
281 17938 or ATCC 55730), or L rhamnosus (ATCC 53103 or ATC
341
282 A07FA or LCR 35) for prevention of NEC over no and other 342
283 probiotics. 343
284 344
285 a
345
286 Please see the accompanying technical review for the supporting evidence.1 346
287 347
288 348
AGA SECTION

289 manufacturers use different processes, which may affect the The AGA makes no recommendations for the use of 349
290 actual content of the probiotic utilized, but this is not within probiotics in the treatment of C difficile infection. Inci- 350
291 the scope of this guideline and therefore we provided the dence of C difficile infection is rising and is responsible 351
292 granular data regarding each strain as specified in the for almost half a million infections in the United States 352
293 published reports. When the information was available, we in 2011,13 with recurrence rates of up to 19.9%, and 353
294 considered the strain specificity when evaluating outcomes. leading to 29,000 deaths. Fecal microbiota trans- 354
295 plantation is highly effective in treating recurrent C 355
296 In patients with Clostridioides difficile infection, we difficile infection,14 but the data supporting the use of 356
297 recommend the use of probiotics only in the context probiotics in initial or recurrent C difficile infection are 357
298 of a clinical trial. No recommendation, knowledge gap, 358
less convincing.
299 359
300 360

PGL 5.6.0 DTD
YGAST63501_proof
6 July 2020
2:53 pm
ce

 4 Su et al Gastroenterology Vol. -, No. -

361 The technical review identified 5 placebo-controlled The AGA suggests the use of certain strains and strain 421
362 randomized controlled trials (RCTs) evaluating probiotics combination of probiotics in the prevention of C difficile 422
363 as adjunct treatment with antibiotics, testing 4 different infection. Although there is a large body of literature 423
364 probiotic formulations. The patient populations across studying the role of probiotics in preventing antibiotic- 424
365 studies differed, including patients with an initial C difficile associated C difficile infection, the studies are very hetero- 425
366 infection, recurrent infection, or both. Probiotics or placebo geneous. The technical review identified 39 studies that 426
367 was administered together with metronidazole or vanco- were previously evaluated by Cochrane review published in 427
368 mycin at low dose or high doses. Due to these variations in 2017.15 A total of 9955 patients were included but the 428
369 the study design, as well as in clinical outcomes, data were populations studied were extremely varied, including pedi- 429
370 deemed too heterogeneous to be pooled in the analysis. All 5 atric, adult, and elderly patients—utilizing a variety of 430
371 published studies contained uncertain or high risk of bias antibiotic regimens in both inpatient and outpatient set- 431
372 regarding blinding of outcome assessment and selective tings—who have very different risks for the development of 432
373 reporting. C difficile infection. The Cochrane review found that pro- 433
374 The probiotic formulations studied included Saccharo- biotics reduced the overall risk of C difficile infection vs 434
375 myces boulardii, Lactobacillus plantarum 299v, Lactobacillus placebo (RR, 0.40; 95% CI, 0.30–0.52); however, the bene- 435
376 rhamnosus ATCC 53103, and the 4-strain combination of ficial effect was driven by the population of patients with 436
377 Lactobacillus acidophilus ATCC 700396, Lactobacillus para- high risk of developing C difficile infection, with no signifi- 437
378 casei subsp paracasei ATCC 335, Bifidobacterium animalis cant effects observed in patients with low or baseline risk. 438
379 subsp lactis ATCC SD5220 and B animalis subsp lactis ATCC The technical review did not identify any new RCTs 439
380 SD5219. The largest study, involving 134 patients, reported between the 2017 Cochrane review and November 2018 440
381 that S boulardii may have a beneficial effect on cessation and thus assessed the certainty of evidence from these 39 441
382
Q6 (relative risk [RR], 1.33; 95% confidence interval [CI], 1.02– trials.1 The overall certainty of the evidence was down- 442
383 1.74) and recurrence of diarrhea (RR, 0.59; 95% CI, 0.35– graded from Moderate to Low due to unclear or high risk of 443
384 0.98), but the quality of evidence was low. The smaller trials bias in most of the trials across all domains for all outcomes 444
385 with L plantarum 299v or the 4-strain combination sug- assessed. Several studies were published as abstracts only 445
386 gested that these probiotics also may have beneficial effects or referenced unpublished data. Publication bias was 446
387 on diarrhea but the evidence was very uncertain, while the considered, as a large number of registered trial protocols 447
388 administration of L rhamnosus ATCC 53103 resulted in on this topic were not associated with subsequent peer- 448
389 increased recurrence of C difficile infection compared to reviewed publications. 449
390 placebo (RR, 2.63; 95% CI, 0.35–19.85). The overall cer- Subgroup analyses of individual probiotic strains or 450
391 tainty of evidence across all critical outcomes for probiotics strain combinations that may have an effect compared with 451
392 used as adjunctive treatment for C difficile infection was low. placebo found that the risk of C difficile infection develop- 452
393 Furthermore, the technical review identified a potential risk ment was reduced by S boulardii (RR, 0.41; 95% CI, 0.22– 453
394 of publication bias due to multiple registered trials that 0.79); the 2-species combination of L acidophilus CL1285 454
395 were not linked to a published report. While currently and L casei LBC80R (RR, 0.22; 95% CI, 0.11–0.42); the 3- 455
396 available data suggest that some probiotics might be bene- strain combination of L acidophilus, L delbrueckii subsp 456
397 ficial in treatment of C difficile, further studies with stan- bulgaricus, and B bifidum (RR, 0.35; 95% CI, 0.15–0.85); as 457
398 dardized study design and larger number of patients are well as the 4-strain combination of L acidophilus, L del- 458
399 needed to define those probiotics, as well as to identify brueckii subsp bulgaricus, B bifidum, and S salivarius subsp 459
400 which patient populations may benefit from this thermophilus (RR, 0.28; 95% CI, 0.11–0.67), with the overall 460
401 intervention. quality of the evidence rated as Low. It should be pointed 461
402 out that beneficial effect of probiotics was seen mainly in 462
In adults and children on antibiotic treatment, we patients with very high risk of developing C difficile infection
403 suggest the use of S boulardii; or the 2-strain 463
404 (>15% baseline risk) and that the analysis of most studies 464
combination of L acidophilus CL1285 and had a wide CI that includes the potential for some benefit, as
405 Lactobacillus casei LBC80R; or the 3-strain 465
406 combination of L acidophilus, Lactobacillus well as for some harm. Thus, patients who place a high value 466
407 delbrueckii subsp bulgaricus, and Bifidobacterium on avoiding associated financial cost or potential harms 467
408 bifidum; or the 4-strain combination of L acidophilus, (especially those immunocompromised patients) and who 468
AGA SECTION

409 L delbrueckii subsp bulgaricus, B bifidum, and have low risk of developing C difficile infection (mainly 469
Streptococcus salivarius subsp thermophilus over no outpatients in the community) may choose not to use any
410 or other probiotics for prevention of C difficile 470
411 probiotics. 471
infection.
412 Conditional recommendation, low quality of evidence. 472
In adults and children with Crohn’s disease, the AGA
413 Comment: Patients who place a high value on the
recommends use of probiotics only in the context of 473
414 potential harms (particularly those with severe illnesses)
a clinical trial. No recommendation, knowledge gap. 474
415 or a high value on avoiding the associated cost and a 475
low value on the small risk of C difficile development
416 476
(particularly in the outpatient setting), would reasonably
417 select no probiotics. The AGA recommends use of probiotics on only in the 477
418 context of a clinical trial for adults and children with Crohn’s 478
419 479
420 480

PGL 5.6.0 DTD
YGAST63501_proof
6 July 2020
2:53 pm
ce

 - 2020 AGA Clinical Practice Guidelines on Probiotics and Gastrointestinal Disorders 5

481 disease. Alterations in the gut microbiome in patients with study design, patient populations, and the specific probiotics 541
482 Crohn’s disease are being explored increasingly, and interest that have been studied. 542
483 in microbiota-based therapies, such as probiotics and fecal The technical review identified 11 studies on the use of 543
484 microbiota transplantation, is growing. However, studies of probiotics for induction of remission in adults and children 544
485 probiotics for induction or maintenance of remission in with ulcerative colitis. The probiotic formulations under 545
486 Crohn’s disease have been limited by small sample sizes, evaluation included the 3-strain combination of B breve 546
487 heterogeneity in patient populations, heterogeneity in study Yakult, B bifidum Yakult, and L acidophilus; B longum Reuter 547
488 design, and differences in the probiotic formulations tested. ATCC BAA-999; E coli Nissle 1917; Lactobacillus reuteri 548
489 The technical review searched for studies of both in- ATCC 55730; and the 8-strain combination of L paracasei 549
490 duction and maintenance of remission in Crohn’s disease in subsp paracasei DSM 24733, L plantarum DSM 24730, L 550
491 adults and children. Only 1 study of 11 subjects was iden- acidophilus DSM 24735, L delbrueckii subsp bulgaricus DSM 551
492 tified for induction of remission in either adults or children. 24734, B longum subsp longum DSM 24736, B breve DSM 552
493 This study found no evidence of benefit for L rhamnosus 24732, B longum subsp infantis DSM 24737, and S salivarius 553
494 ATCC 53103 compared to placebo for induction of remission subsp thermophilus DSM 24731. The comparators varied 554
495 (odds ratio [OR], 0.80; 95% CI, 0.04–17.20), with CI that among studies, and in some studies included mesalamine. 555
496 were wide and thus did not exclude the potential for benefit Four studies compared the 8-strain probiotic combina- 556
497 or harm. tion to mesalamine or balsalazide for induction of remission, 557
498 Eleven studies of probiotics for maintenance of remis- suggesting potential for benefit but with very low certainty 558
499 sion in adults and children with Crohn’s disease were of the evidence (RR, 1.72; 95% CI, 0.78–3.32). Two studies 559
500 identified. The identified studies were heterogeneous in examined the effectiveness of oral E coli Nissle 1917 560
501 inclusion criteria, the probiotic studied, and study design. compared to mesalamine for this indication, again with 561
502 The probiotic formulations studied included Escherichia coli uncertain benefit (RR, 0.86; 95% CI, 0.49–1.49). One of 562
503 Nissle 1917; S boulardii; L rhamnosus ATCC 53103; Lacto- these studies also allowed adjunctive treatment with ste- 563
504 bacillus johnsonii NCC 533; and an 8-strain probiotic com- roids and gentamicin, while the other allowed co-therapy 564
505 bination of L paracasei subsp paracasei DSM 24733, L with topical prednisolone. One study of rectally adminis- 565
506 plantarum DSM 24730, L acidophilus DSM 24735, L del- tered E coli Nissle 1917 did not show any clear evidence of 566
507 brueckii subsp bulgaricus DSM 24734, Bifidobacterium lon- benefit compared to placebo. Rectally administered L reuteri 567
508 gum subsp longum DSM 24736, Bifidobacterium breve DSM ATCC 55730 was tested in children, with suggestion of an 568
509 24732, Bifidobacterium longum subsp infantis DSM 24737, increased clinical response rate compared to placebo (RR, 569
510 and S salivarius subsp thermophilus DSM 24731. In addition, 1.83; 95% CI, 1.14–2.92). Other probiotics were tested in 570
511 the studies differed on whether remission was induced by single studies only, with no demonstrated benefit for in- 571
512 medical or surgical therapy, and only 1 study enrolled duction of remission. 572
513 children. Lastly, some of the studies used mesalamine as the The technical review identified 6 studies of probiotics 573
514 comparator, or allowed co-therapy with mesalamine in the for maintenance of remission in ulcerative colitis. Two 574
515 probiotic arm. There was no overall evidence of benefit from studies of E coli Nissle 1917 and 1 study of L rhamnosus 575
516 any of the probiotic therapies studied for maintenance of ATCC 53103 (RR, 0.82; 95% CI, 0.60–1.11) did not show 576
517 remission. clear benefit of the probiotic compared to mesalamine for 577
518 The overall quality of the evidence was rated as low for maintenance of remission. In addition, compared to placebo, 578
519 induction of remission and maintenance of remission. Given the 2-strain combination of L acidophilus LA-5 and B ani- 579
520 the overall small study samples, as well as heterogeneity in malis subsp lactis Bb12; the 2-strain combination of B breve 580
521 patient populations, probiotic strains studied, and study Yakult and L acidophilus; and the 3-strain combination of 581
522 design, it is unclear whether there is potential for specific Enterococcus faecalis T-111, Clostridium butyricum TO-A, 582
523 probiotic strains to be beneficial for either induction or and Bacillus mesentericus TO-A did not show any evidence 583
524 maintenance of remission in Crohn’s disease. Further of benefit for this indication, although these formulations 584
525 studies are needed to define specific populations of patients were tested in single studies only. 585
526 with Crohn’s disease who might benefit from probiotics, as The overall quality of evidence for probiotics for in- 586
527 well as the most effective probiotic strains. duction or maintenance of remission in ulcerative colitis 587
528 was rated as low. Available evidence is limited by small 588
AGA SECTION

529 sample sizes, differences in patient populations, variability 589

In adults and children with ulcerative colitis, the AGA
530 in study design, and heterogeneity in the probiotic formu- 590
recommends the use of probiotics only in the
531 context of a clinical trial. No recommendation, lations used. The most extensively tested formulation was 591
532 knowledge gap. the 8-strain combination of L paracasei subsp paracasei 592
533 DSM 24733, L plantarum DSM 24730, L acidophilus DSM 593
534 24735, L delbrueckii subsp bulgaricus DSM 24734, B longum 594
The AGA recommends the use of probiotics in adults and subsp longum DSM 24736, B breve DSM 24732, B longum
535 595
children with ulcerative colitis only in the context of a subsp infantis DSM 24737, and S salivarius subsp thermo-
536 596
clinical trial. As with Crohn’s disease, interest in microbiota- philus DSM 24731 for induction of remission, although even
537 597
based therapies for ulcerative colitis is growing. However, here the available studies were limited by potential for bias
538 598
available evidence is limited because of heterogeneity in and pooled results did not show evidence of benefit. Further
539 599
540 600

PGL 5.6.0 DTD
YGAST63501_proof
6 July 2020
2:53 pm
ce

 6 Su et al Gastroenterology Vol. -, No. -

601 research is needed to identify specific patient populations underwent an ileal pouch–anal anastomosis for conditions 661
602 that might benefit most from treatment with probiotics and other than chronic ulcerative colitis, such as familial 662
603 to define the most effective probiotic formulations. adenomatous polyposis. 663
604 664
In adults and children with pouchitis, the AGA In symptomatic children and adults with irritable
605 suggests the use of the 8-strain combination of L bowel syndrome, we recommend the use of
665
606 paracasei subsp paracasei DSM 24733, L plantarum probiotics only in the context of a clinical trial. No 666
607 DSM 24730, L acidophilus DSM 24735, L delbrueckii recommendations, knowledge gap. 667
608 subsp bulgaricus DSM 24734, B longum subsp 668
609 longum DSM 24736, B breve DSM 24732, B longum 669
610 subsp infantis DSM 24737, and S salivarius subsp The AGA makes no recommendations for the use of 670
611 thermophilus DSM 24731 over no or other probiotics. probiotics in children and adults with irritable bowel syn- 671
Conditional recommendation, very low quality of drome (IBS). While there are many studies examining this
612 evidence. 672
613 question, they are marked by significant heterogeneity in 673
Comment: Patients for whom the feasibility and cost of
614 using this combination of bacterial strain is problematic study design, outcome, and probiotics used. 674
615 may reasonably select no probiotics. The technical review found a total of 76 RCTs that used 44 675
616 different probiotic strains or combinations of strains.1 For the 676
617 majority of studies that reported a benefit, the data were 677
618 The AGA suggests the use of the 8-strain combination of derived from a single RCT. Only 2 formulations (S boulardii 678
619 L paracasei subsp paracasei DSM 24733, L plantarum DSM and the 8-strain combination) had more than 1 RCT that 679
620 24730, L acidophilus DSM 24735, L delbrueckii subsp bul- measured the same outcome, allowing for combined analysis. 680
621 garicus DSM 24734, B longum subsp longum DSM 24736, B Three studies tested S boulardii in 232 adults with IBS and 681
622 breve DSM 24732, B longum subsp infantis DSM 24737, and while the studies used different outcome measures, all re- 682
623 S salivarius subsp thermophilus DSM 24731 over no or other ported an abdominal pain score that was not different be- 683
624 probiotics in patients with pouchitis. Pouchitis is a frequent tween those treated with S boulardii and those treated with 684
625 post-surgical complication after total proctocolectomy and placebo. Two RCTs tested the 8-strain combination (L para- 685
626 ileal pouch–anal anastomosis for ulcerative colitis, and a casei subsp paracasei DSM 24733, L plantarum DSM 24730, L 686
627 role for the gut microbiota has been suggested in its path- acidophilus DSM 24735, L delbrueckii subsp bulgaricus DSM 687
628 ogenesis. The possibility of microbiota-directed therapy for 24734, B longum subsp longum DSM 24736, B breve DSM 688
629 this condition has been suggested. 24732, B longum subsp infantis DSM 24737, and S salivarius 689
630 The technical review identified 7 studies of probiotics subsp thermophilus DSM 24731) in 73 adults with IBS and 690
631 for treatment or prevention of pouchitis in adult patients abdominal pain and although this demonstrated a decrease in 691
632 with an ileal pouch–anastomosis for management of ulcer- the abdominal pain score using the visual analog scale (mean 692
633 ative colitis. The 8-strain probiotic formulation for mainte- decrease, 3.78; 95% CI, 4.93–2.62), the overall sample size 693
634 nance of remission in chronic pouchitis was tested in 2 was small and there was unclear risk of selection, reporting, 694
635 studies including a total of 76 patients, with a potential and detection bias. In addition, the patients enrolled were of 695
636 benefit in the proportion of patients who maintained variable IBS subtypes. 696
637 remission at 12 months compared to placebo (RR, 20.24; In the remainder of the studies, the majority of the single 697
638 95% CI, 4.28–95.81, low certainty of evidence). Two addi- RCTs using different probiotic and probiotic combinations 698
639 tional studies suggested a benefit of the same 8-strain of variable duration reported some benefit, but the sample 699
640 combination for prevention of an initial episode of acute sizes were all relatively small and had significant differences 700
641 pouchitis, but with very low certainty of evidence (RR for no in study subjects and designs. The overall quality of evi- 701
642 episodes of pouchitis, 1.29; 95% CI, 1.03–1.61). Single trials dence was very low. There was also significant concern for 702
643 of L rhamnosus ATCC 53103, B longum Reuter ATCC BAA- publication bias, as the Technical Review team found 703
644 999, and C butyricum CBM 588 did not show clear evi- numerous registered protocols that yielded no peer- 704
645 dence of benefit for treatment or prevention of pouchitis, reviewed publications or results that were publicly avail- 705
646 although samples sizes were extremely small in all available able. Although there has been significant interest and po- 706
647 studies. tential for the use of probiotics in IBS, further studies are 707
648 The overall quality of evidence was rated as very low needed to clarify this important question. 708
AGA SECTION

649 due to risk of bias, small sample sizes, and heterogeneity in 709
650 the patient populations and interventions tested. The ma- 710
jority of evidence came from studies of the 8-strain pro- In children with acute infectious gastroenteritis, we
651 suggest against the use of probiotics. 711
652 biotic combination of L paracasei subsp paracasei DSM 712
Conditional recommendation, moderate quality of
653 24733, L plantarum DSM 24730, L acidophilus DSM 24735, L evidence. 713
654 delbrueckii subsp bulgaricus DSM 24734, B longum subsp 714
655 longum DSM 24736, B breve DSM 24732, B longum subsp 715
infantis DSM 24737, and S salivarius subsp thermophilus The AGA suggests against the use of probiotics in children
656 716
DSM 24731. Other probiotic formulations need further with acute infectious gastroenteritis in the United States and
657 717
testing for this indication. It also unclear whether these Canada. The majority of the data supporting the use of pro-
658 718
results would apply to children or to patients who biotics in children with acute infectious gastroenteritis were
659 719
660 720

PGL 5.6.0 DTD
YGAST63501_proof
6 July 2020
2:53 pm
ce

 - 2020 AGA Clinical Practice Guidelines on Probiotics and Gastrointestinal Disorders 7

721 from studies performed outside of United States and Canada, The AGA suggests the use of certain probiotic strain or 781
722 while 2 high-quality studies performed in the United States strain combination for the prevention of NEC in preterm 782
723 and Canada did not show any benefit. infants less than 37 weeks gestational age and low birth 783
724 The technical review identified 89 studies, 58 were included weight. Preterm birth is common, affecting 10% of new- 784
725 in a Cochrane Review published in 2010 and 31 additional borns in the United States and 15 million pregnancies 785
726 studies were published after 2010. Most of the studies that worldwide each year. Premature infants have increased risk 786
727 showed a benefit of probiotics were published in India, Italy, of mortality and multiple morbidities, including NEC. NEC is 787
728 Poland, Turkey, and Pakistan and had 1 or more concerns the most important gastrointestinal emergency among 788
729 regarding risk of bias.1 Of the 89 studies, 58 reported duration of preterm neonates, characterized by mucosal or even deeper 789
730 diarrhea as an outcome. Combining these 58 studies that utilized intestinal necrosis of the bowel with common long-term 790
731 various strains of probiotics, the mean duration of diarrhea was sequelae, including short bowel syndrome and impaired 791
732 reduced by 21.91 hours (95% CI, 16.17–27.64 hours), but the neurodevelopment. Microbiota differs in infants with NEC 792
733 level of evidence was low. The most commonly studied probiotic compared to healthy infants providing a rationale for 793
734 was S boulardii, which was utilized in 21 RCTs. Only 9 studies microbiota-oriented treatments. 794
735 reported on the mean duration of diarrhea, which was reduced The technical review presented results from a recent 795
736 by 28.9 hours (95% CI, 16.78–41.03 hours), but the level of systematic review and network meta-analysis that assessed 796
737 evidence was very low. The second most commonly used strain the role of probiotics in the prevention of mortality and 797
738 was L rhamnosus ATCC 53103, which was evaluated in 19 RCTs. morbidity in preterm infants.16 In total, 63 studies 798
739 Of these 19 RCTs, 14 studies reported mean duration of diarrhea comparing single- and multiple-strain probiotics to placebo 799
740 as an outcome, which was reduced by 23.13 hours (95% CI, in patients with severe NEC were included and multiple 800
741 12.33–33.94 hours). outcomes, such as all-cause mortality, severe NEC (stage II 801
742 While some strains of bacteria improved diarrhea duration or greater), culture-proven sepsis, and duration of hospi- 802
743 in children, few of the studies were performed in North America talization were assessed. 803
744 until 2 recent multicenter, randomized, double-blind, placebo- Combinations of Lactobacillus spp and Bifidobacterium 804
745 controlled trials conducted by the Pediatric Emergency Care spp (L rhamnosus ATCC 53103 and B longum subsp infantis; 805
746 Applied Research Network and the Pediatric Emergency or L casei and B breve; or L rhamnosus, L acidophilus, L casei, 806
747 Research Canada. These studies enrolled 943 and 827 children B longum subsp infantis, B bifidum, and B longum subsp 807
748 from 10 and 6 emergency departments in the United States and longum; or L acidophilus and B longum subsp infantis; or L 808
749 Canada, respectively. The US study used L rhamnosus ATCC acidophilus and B bifidum; or L rhamnosus ATCC 53103 and 809
750 53103 and the Canadian study used a combination of L rham- B longum Reuter ATCC BAA-999; or L acidophilus, B bifidum, 810
751 nosus R0011 and Lactobacillus helveticus R0052 for 5 days. B animalis subsp lactis, and B longum subsp longum; or B 811
752 Neither showed any benefit in the occurrence of moderate-to- animalis subsp lactis [including DSM 15954], or L reuteri 812
753 severe gastroenteritis between placebo and probiotics groups. [DSM 17938 or ATCC 55730], or L rhamnosus [ATCC 53103 813
754 Two additional studies in the United States and Canada using or ATC A07FA or LCR 35]) reduced all-cause mortality 814
755 the same strains of bacteria confirmed the lack of benefit. Given compared to placebo (OR, 0.56, 95% CI, 0.39–0.80), while 815
756 likely differences in host genetics, diet, sanitation, and endemic severe NEC was reduced by combinations of Lactobacillus 816
757 enteropathogens between North America and the other global spp and Bifidobacterium spp (L rhamnosus ATCC 53103 and 817
758 regions, as well as different causes of acute infectious gastro- B longum subsp infantis; or L casei and B breve; or L 818
759 enteritis in children, we do not feel that the studies conducted in rhamnosus, L acidophilus, L casei, B longum subsp infantis, B 819
760 other regions can be generalized to the population served by the bifidum, and B longum subsp longum; or L acidophilus and B 820
761 AGA and thus suggest against the use of probiotics for acute longum subsp infantis; or L acidophilus and B bifidum; or L 821
762 infectious gastroenteritis in children. rhamnosus ATCC 53103 and B longum Reuter ATCC BAA- 822
763 999; or L acidophilus, B bifidum, B animalis subsp lactis, 823
In preterm (less than 37 weeks gestational age), low- and B longum subsp longum; OR, 0.35; 95% CI, 0.20–0.59), B
764 birth-weight infants, we suggest using a combination 824
765 animalis subsp lactis (including strain DSM 15954; OR, 0.31; 825
of Lactobacillus spp and Bifidobacterium spp (L 95% CI, 0.13–0.74), L reuteri (DSM 17938 or ATCC 55730;
766 rhamnosus ATCC 53103 and B longum subsp 826
767 infantis; or L casei and B breve; or L rhamnosus, L OR, 0.55; 95% CI, 0.34–0.91), or L rhamnosus (ATCC 53103 827
768 acidophilus, L casei, B longum subsp infantis, B or ATC A07FA or LCR 35; OR, 0.44; 95% CI, 0.21–0.90), all 828
AGA SECTION

769 bifidum, and B longum subsp longum; or L supported by moderate- or high-quality evidence. 829
770 acidophilus and B longum subsp infantis; or L There was low to very low quality of evidence to support 830
acidophilus and B bifidum; or L rhamnosus ATCC beneficial effects of combinations of Lactobacillus spp, Bifi-
771 53103 and B longum Reuter ATCC BAA-999; or L 831
772 dobacterium spp, and Enterococcus spp (L acidophilus, B 832
acidophilus, B bifidum, B animalis subsp lactis, and
773 longum subsp longum, and E faecalis; or L gasseri PTA-5845, 833
B longum subsp longum), or B animalis subsp lactis
774 (including DSM 15954), or L reuteri (DSM 17938 or B longum subsp infantis PTA-5843, and E faecium PTA- 834
775 ATCC 55730), or L rhamnosus (ATCC 53103 or ATC 5844; or L acidophilus, B longum subsp longum, and E fae- 835
776 A07FA or LCR 35) over no and other probiotics. cium; or L acidophilus, B longum subsp infantis, and E fae- 836
Conditional recommendation, moderate/high quality calis (OR, 0.28; 95% CI, 0.16–0.49), combinations of
777 837
of evidence. Bifidobacterium spp and S salivarius subsp thermophilus (B
778 838
779 longum subsp infantis, B bifidum, and S salivarius subsp 839
780 840

PGL 5.6.0 DTD
YGAST63501_proof
6 July 2020
2:53 pm
ce

 8 Su et al Gastroenterology Vol. -, No. -

841 thermophilus; or B longum subsp infantis DSM 33361, B longum subsp infantis; or L casei and B breve; or L rham- 901
842 animalis subsp lactis DSM 15954, and S salivarius subsp nosus, L acidophilus, L casei, B longum subsp infantis, B 902
843 thermophilus TH-4) (OR, 0.38; 95% CI, 0.19–0.75) or a B bifidum, and B longum subsp longum; or L acidophilus and 903
844 subtilis and E faecium (OR, 0.23; 95% CI, 0.08–0.63) in se- B longum subsp infantis; or L acidophilus and B bifidum; or 904
845 vere NEC reduction compared to placebo. L rhamnosus ATCC 53103 and B longum Reuter ATCC BAA- 905
846 Combinations of Lactobacillus spp, Bifidobacterium spp, 999; or L acidophilus, B bifidum, B animalis subsp lactis, 906
847 and S boulardii (L rhamnosus, L acidophilus, B longum subsp and B longum subsp longum) may prevent the develop- 907
848 longum, and S boulardii; or L acidophilus, B bifidum, and S ment of NEC. For children with acute gastroenteritis in 908
849 boulardii) reduced days to reach full enteral feeds (mean North America, however, the current evidence does not 909
850
Q7 difference [MD], –3.30; 95% CI, –5.91 to –0.69), supported support the use of probiotics. While other society guide- 910
851 by moderate- or high-quality evidence. Similar effects, lines17–19 have previously recommended the use of pro- 911
852 although based on low or very low quality of evidence, were biotics in this population, these guidelines were developed 912
853 shown with combinations of Lactobacillus spp and Bifido- without utilizing GRADE methodology and also relied on 913
854 bacterium spp (L casei and B breve; or L rhamnosus, L aci- data outside of North America, which became available 914
855 dophilus, L casei, B longum subsp infantis, B bifidum, and B after the recommendations were made. This is an area that 915
856 longum subsp longum; or L acidophilus and B bifidum; or L will require further study and recommendations evolve as 916
857 acidophilus, B bifidum, B animalis subsp lactis, and B longum more direct high-quality data become available. 917
858 subsp longum; MD, –2.15; 95% CI, –3.78 to –0.51), or with L We identified that significant knowledge gaps exist in Q8 918
859 reuteri (DSM 17938 or ATCC 55730 (MD, –2.62; 95% CI, this very promising and important area of research due to 919
860 –4.53 to –0.71). Finally, B animalis subsp lactis and L reuteri the significant heterogeneity between studies and vari- 920
861 (DSM 17938 or ATCC 55730) significantly shortened hos- ability in the probiotic strains studied. The lack of 921
862 pitalization based on moderate- or high-quality evidence consistent harms reporting makes it difficult to assess true 922
863 (MD, –13.00; 95% CI, –22.71 to –3.29 and MD, –7.89; 95% harms. The lack of product manufacturing details pro- 923
864 CI, –11.60 to –4.17, respectively). hibits true comparisons and decreases the feasibility of 924
865 obtaining certain products by patients. Future high-quality 925
866 studies are urgently needed that address these pitfalls. 926
867
Discussion These guidelines will undergo a review and consideration 927
868 The gut microbiome plays an important role in for an update within 3–5 years or earlier if practice- 928
869 gastrointestinal health and disease and probiotics repre- changing evidence becomes available. 929
870 sent a promising modality for therapeutic intervention. As 930
871 outlined in the accompanying technical review,1 the cur- 931
872 rent evidence suggests that the use of certain probiotic References 932
873 strains or probiotic strain combinations may prevent C 933
1. Preidis GA, Weizman AV, Kashyap PC, et al. American
874 difficile infections for adults and children on antibiotic 934
Gastroenterological Association Institute Technical Review
875 treatment. However, the quality of evidence was low and 935
on the Role of Probiotics in the Management of Gastro-
876 the reporting of potential harms was not always consis- intestinal Disorders. Gastroenterology 2020;XX. XX–XX. 936
877 tent. Thus, for patients who place a high value on avoid- 937
2. Sanders ME, Merenstein DJ, Reid G, et al. Probiotics and
878 ance of potential harms, particularly those with severe 938
prebiotics in intestinal health and disease: from biology
879 illnesses or immunosuppression, it would be reasonable to to the clinic. Nat Rev Gastroenterol Hepatol 2019; 939
880 select not to use probiotics. While there was evidence for 16:605–616. 940
881 probiotics in the prevention of C difficile, the technical 3. Food and Agricultural Organization of the United Nations 941
882 review found significant knowledge gap in the use of (FAO)/World Health Organization (WHO). Health and 942
883 probiotics in treatment of C difficile and recommend this Nutritional Properties of Probiotics in Food including 943
884 as an area for further study. Similar knowledge gaps exist Powder Milk with Live Lactic Acid Bacteria. Report of a 944
885 in the use of probiotics in irritable bowel syndrome and Joint FAO/WHO Expert Consultation on Evaluation of 945
886 inflammatory bowel disease (Crohn’s disease and ulcera- Health and Nutritional Properties of Probiotics in Food 946
887 tive colitis). In the subset of patients with pouchitis, cur- Including Powder Milk with Live Lactic Acid Bacteria; 947
888 rent evidence supports the use of the 8-strain combination Cordoba, Argentina; October 1–5, 2001. 948
AGA SECTION

889 (L paracasei subsp paracasei DSM 24733, L plantarum 4. Sniffen JC, McFarland LV, Evans CT, et al. Choosing an 949
890 DSM 24730, L acidophilus DSM 24735, L delbrueckii subsp appropriate probiotic product for your patient: An 950
891 bulgaricus DSM 24734, B longum subsp longum DSM evidence-based practical guide. PLoS One 2018;13: 951
24736, B breve DSM 24732, B longum subsp infantis DSM e0209205.
892 952
893 24737, and S salivarius subsp thermophilus DSM 24731) if 5. Caselli M, Cassol F, Calo G, et al. Actual concept of 953
feasibility of obtaining the combination is not a barrier. In "probiotics": is it more functional to science or business?
894 954
preterm infants less than 37 weeks gestational age, the World J Gastroenterol 2013;19:1527–1540.
895 955
probiotic strains B animalis subsp lactis or L reuteri (DSM 6. de Simone C. The unregulated probiotic market. Clin
896 956
17938 or ATCC 55730) or L rhamnosus (ATCC 53103 or Gastroenterol Hepatol 2019;17:809–817.
897 957
ATC A07FA or LCR 35) or combination of Lactobacillus spp 7. Suez J, Zmora N, Segal E, et al. The pros, cons, and many
898 958
and Bifidobacterium spp (L rhamnosus ATCC 53103 and B unknowns of probiotics. Nat Med 2019;25:716–729.
899 959
900 960

PGL 5.6.0 DTD
YGAST63501_proof
6 July 2020
2:53 pm
ce

 - 2020 AGA Clinical Practice Guidelines on Probiotics and Gastrointestinal Disorders 9

8. Bindels LB, Delzenne NM, Cani PD, et al. Towards a paper by the ESPGHAN Working Group for Probiotics
961 1021
more comprehensive concept for prebiotics. Nat Rev and Prebiotics. J Pediatr Gastroenterol Nutr 2014;
962 1022
Gastroenterol Hepatol 2015;12:303–310. 58:531–539.
963 1023
9. Parker EA, Roy T, D’Adamo CR, et al. Probiotics and 19. Shane AL, Mody RK, Crump JA, et al. 2017 Infectious
964 1024
gastrointestinal conditions: an overview of evidence from the Diseases Society of America Clinical Practice Guidelines
965 1025
Cochrane Collaboration. Nutrition 2018;45:125–134 e11. for the Diagnosis and Management of Infectious Diar-
966 1026
10. Clarke TC, Black LI, Stussman BJ, et al. Trends in the rhea. Clin Infect Dis 2017;65:1963–1973.
967 1027
use of complementary health approaches among adults:
968 1028
United States, 2002–2012. Natl Health Stat Report 2015
969 Feb 10;(79):1–16. 1029
970 Correspondence 1030
11. Sultan S, Falck-Ytter Y, Inadomi JM. The AGA Institute Address correspondence to: Chair, Clinical Guidelines Committee, American
971 process for developing clinical practice guidelines part Gastroenterological Association, National Office, 4930 Del Ray Avenue, 1031
972 one: grading the evidence. Clin Gastroenterol Hepatol Bethesda, Maryland 20814. e-mail: clinicalpractice@gastro.org Q3 1032
973 2013;11:329–332. Acknowledgments
1033
974 12. Graham R, Mancher M, Dianne W, et al. Clinical Practice The Clinical Guidelines Committee included: Shahnaz Sultan, MD, MHSc, 1034
975 Guidelines We Can Trust. Washington, DC: National
AGAF, Chair; Karen Chachu, MD, PhD; Lukejohn Day, MD, AGAF; Benjamin 1035
Lebwohl, MD, AGAF; Thiruvengadam Muniraj, MD, PhD, MRCP; Amit Patel,
976 Academies Press, 2011. MD; Anne F. Peery, MD; Raj Shah, MD; Harminder Singh, MD; Siddharth 1036
977 13. Lessa FC, Mu Y, Bamberg WM, et al. Burden of Clos-
Singh, MD; Stuart J. Spechler, MD, AGAF; Grace L. Su, MD, AGAF; Aaron P. 1037
Thrift, BSc, PhD; Jennifer M. Weiss, MD, MS, AGAF; and Adam V. Weizman,
978 tridium difficile infection in the United States. N Engl J MD. Q4 1038
979 Med 2015;372:825–834. 1039
The Guideline Panel included: Grace L. Su (Chair): Division of Gastroenterology
980 14. Vaughn BP, Rank KM, Khoruts A. Fecal microbiota and Hepatology, University of Michigan, Ann Arbor, Michigan and
1040
981 transplantation: current status in treatment of GI and liver Gastroenterology Section, Veterans Administration Ann Arbor Healthcare 1041
982 disease. Clin Gastroenterol Hepatol 2019;17:353–361.
System, Ann Arbor, Michigan; Cynthia W. Ko (Member): Division of 1042
Gastroenterology, University of Washington Medical School, Seattle, WA;
983 15. Goldenberg JZ, Yap C, Lytvyn L, et al. Probiotics for the Premysl Bercik (Member): Division of Gastroenterology, McMaster University, 1043
984 prevention of Clostridium difficile-associated diarrhea in Hamilton, Ontario, Canada. 1044
985 adults and children. Cochrane Database Syst Rev 2017; The Technical Review Panel included: Rebecca L. Morgan (GRADE 1045
986 12:CD006095. Methodologist): Department of Health Research Methods, Evidence and 1046
Impact, McMaster University, Hamilton, Ontario, Canada; Adam V. Weizman
987 16. Morgan RL, Preidis GA, Kashyap PC, et al. Probiotics for (Methodologist Trainee): Division of Gastroenterology, Mount Sinai Hospital, 1047
988 prevention of mortality and morbidity in preterm infants: Department of Medicine, University of Toronto, Toronto, Ontario, Canada; 1048
989 a systematic review and network meta-analysis of ran- Purna Kashyap (Content Expert): Department of Internal Medicine, Mayo
Clinic, Rochester, Minnesota; Geoffrey A. Preidis (Content Expert):
1049
990 domized trials. Gastroenterology 2020;XX:xx–xx. Department of Pediatrics, Baylor College of Medicine, Houston, Texas. 1050
991 17. Guarino A, Ashkenazi S, Gendrel D, et al. European So- 1051
Conflicts of interest
992 ciety for Pediatric Gastroenterology, Hepatology, and These authors disclose the following: Premysl Bercik received financial 1052
993 Nutrition/European Society for Pediatric Infectious Dis- support/remuneration from Nestex, Allergan Canada, IM HealthScience, and 1053
LUPIN Pharma Canada. Adam V. Weizman received financial support/
994 eases evidence-based guidelines for the management of renumeration from AbbVie, Janssen Pharmaceutica, Takeda Pharmaceutical 1054
995 acute gastroenteritis in children in Europe: update 2014. Company, and Ferring Pharmaceuticals. The remaining authors disclose no 1055
conflicts. All members were required to complete disclosure statement.
996 J Pediatr Gastroenterol Nutr 2014;59:132–152.
These statements are maintained at the American Gastroenterological
1056
997 18. Szajewska H, Guarino A, Hojsak I, et al. Use of probiotics Association Institute headquarters in Bethesda, Maryland and pertinent 1057
Q5
998 for management of acute gastroenteritis: a position disclosures are published with the report. 1058
999 1059
1000 1060
1001 1061
1002 1062
1003 1063
1004 1064
1005 1065
1006 1066
1007 1067
1008 1068
AGA SECTION

1009 1069
1010 1070
1011 1071
1012 1072
1013 1073
1014 1074
1015 1075
1016 1076
1017 1077
1018 1078
1019 1079
1020 1080

PGL 5.6.0 DTD
YGAST63501_proof
6 July 2020
2:53 pm
ce