Efficacy of probiotics on cognition, and biomarkers of inflammation and oxidative stress in adults with Alzheimer’s disease or mild cognitive impairment — a meta-analysis of randomized controlled trials

Probiotics are live microbes that confer health benefits to the host. Preliminary animal evidence supports the potential role of probiotics in ameliorating cognitive health, however, findings from clinical trials in Alzheimer’s disease (AD) or mild cognitive impairment (MCI) subjects are controversial. Thus, a meta-analysis is needed to clarify the efficacy of probiotics on cognition in AD or MCI patients. EMBASE, PubMed, Web of Science and Cochrane library were systematically searched and manually screened for relevant published randomized controlled trials (RCTs). Among the 890 citations identified, 5 studies involving 297 subjects met eligibility. There was a significant improvement in cognition (SMD = 0.37; 95% CI, 0.14, 0.61; P = 0.002; I2 = 24%), while a significant reduction in malondialdehyde (SMD = −0.60; 95% CI, −0.91, −0.28; P = 0.000; I2 = 0.0%) and high-sensitivity C-reactive protein (SMD = −0.57; 95% CI, −0.95, −0.20; P = 0.003; I2 = 0.0%) post-intervention levels between the probiotics and control group. This meta-analysis indicated that probiotics improved cognitive performance in AD or MCI patients, possibly through decreasing levels of inflammatory and oxidative biomarkers. However, current evidence is insufficient, and more reliable evidence from large-scale, long-period, RCT is needed.

Alzheimer* or AD or Dement* or "cognitive" or "mild cognitive impairment" or MCI or "cognitive fail*" or "cognitive performance" or "cognitive impair*" or "cognitive decline*" or "memory function" or "Mental capacity" or "cognitive defect" RCT or "randomized controlled trials" or "Controlled Clinical Trials" or "random*" or "control*" or "trial*" The incidence of Alzheimer's disease (AD) is increasing globally and has reached the point of being a costly public health issue [1]. Mild cognitive impairment (MCI) is a syndrome defined as cognitive deterioration that does not compromise daily functioning; however, amnestic MCI has a high risk of progression to AD and has been considered as a typical prodromal stage of AD [2]. Unfortunately, there is currently no curative treatment for cognitive impairment and dementia [3].
The gut microbiota (GM) consists of a vast bacterial community that resides primarily in the lower gut and lives in a symbiotic relationship with the host [4]. Accumulating evidence has demonstrated the close interconnection between the gastrointestinal tract and the brain, known as the gut-brain axis [5]. Recently, the GM has been found to regulate brain development and behavior via the gut-brain axis, and this has been called the microbiota-gut-brain axis (MGB) [6]. It has been found that dysfunction in behavior and cognition are associated with GM dysbiosis [7], and further activation of gut inflammation has been regarded as a possible pathogenic cofactor in cognitive deterioration and dementia [8]. Moreover, decreased abundance of antiinflammatory bacterial species such as Bifidobacterium breve strain A1 and increased abundance of proinflammatory flora phyla such as Firmicutes and Bacteroidetes are the most distinctive alterations in the GM observed in AD patients [9]. Therefore, the GM has been proposed as a key player in the pathogenesis of AD and might be a new potential therapeutic target for the prevention and treatment of AD [10].
Probiotics are live microbes that confer health benefits to the host when administered in adequate amounts [11], possibly through their anti-inflammatory or antioxidative effects [12,13]. Recently, some probiotics have been shown to influence the central nervous system (CNS) and behavior via modulation of the MGB [14]. Moreover, eleven preclinical studies have shown that neither single strains nor multi-strain probiotics were beneficial for improving cognitive function in animal models [15]. These preclinical results have indicated that probiotics supplementation might be an effective dietary intervention to ameliorate ageassociated cognitive deficits. Nevertheless, findings from available clinical trials focusing on the effects of probiotics in patients with AD or MCI are inconsistent [16][17][18][19][20]. Additionally, previous relevant reviews mainly focused on the effect of probiotics in neurodegenerative and neurodevelopmental disorders with both animal models and human trials [15,21]. The evidence for the effects of psychobiotics on mental and neurological conditions/disorders remains limited. Thus, a metaanalysis is needed to clarify the efficacy of probiotics on cognition in adults suffering from AD or MCI, as well as the possible underlying mechanisms. Given the potential importance of such a meta-analysis, this protocol provides an in-depth description of the research objectives as well as the methodological and analytical approaches that will be used to identify, appraise, and synthesize the relevant studies.

Objectives
(1) Conduct a meta-analysis to clarify the efficacy of probiotics on cognition in adults suffering from AD or MCI.
(2) To examine the effect of probiotics on inflammatory and oxidative biomarkers in adults with AD or MCI.

Study selection
Eligibility criteria (1) Adult human participants who had a diagnosis of AD or MCI (aged over 18 y); (2) The study was an RCT and published in peerreviewed journals in English; (3) No constraint on dosage, strain, or form of probiotics (either stand alone or in combination with www.aging-us.com 9 AGING other compounds) was applied; (4) Probiotic interventions had no significant difference in form, appearance, taste, and smell compared to the control interventions; (5) No restrictions were applied on validated measures of cognitive assessment; (6) Continuous data at baseline and post-intervention, or the change from baseline, were reported or could be calculated from the data reported in the article. (1) The publications were abstracts, reviews, conference papers, study protocols, cross-sectional studies, nonhuman (in vitro and animal) studies or papers that did not report on any outcome of interest;

Exclusion criteria
(2) No post-intervention or change from baseline on scale scores of cognition was reported, and these data could not be calculated based on the information in the article; (3) The study reported on a sample that overlapped the sample in another study. In this case, only the study with the larger sample size was included.

Information sources
The preliminary search will be performed through electronic databases including the EMBASE, PubMed, Web of Science and the Cochrane Central Register of Controlled Trials (CENTRAL). And a snowball search will be manually carried out by searching reference lists from relevant published reviews and the retrieved papers.

Search strategy
The search strategy was developed, finalized and adapted for each database with a combination of free text and controlled vocabulary keywords, including the following terms: (Alzheimer's disease OR dementia OR mild cognitive impairment OR cognitive dysfunction OR cognitive defect OR cognition OR memory OR mental capacity) AND (adult OR human) AND (probiotic OR yeast OR yoghurt OR fermented product OR lactobacillus OR bifidobacterium OR fermented dairy product OR synbiotics OR cultured milk products). Predefined search strategy for each electronic databases are as follows:   Alzheimer* or AD or Dement* or "cognitive" or "mild cognitive impairment" or MCI or "cognitive fail*" or "cognitive performance" or "cognitive impair*" or "cognitive decline*" or "memory function" or "Mental capacity" or "cognitive defect" #20

Data management
All electronic databases citations retrieved using the above search strategy will be imported into Endnote (Endnote X6, Thomson Reuters, San Francisco, CA) to manage and delete duplicate records. Studies retrieved from reference lists of published reviews and retrieved articles will be entered into Microsoft Excel spread sheet for de-duplication and screening.

Selection process
Screening will be performed by screening the titles and abstracts followed by the retrieval and screening of full text articles using the inclusion and exclusion criteria. We will contact the study authors for full text articles with the maximum of three attempts for any articles that we are unable to retrieve. Articles were initially and independently screened for eligibility by two investigators. Any disagreements that arise between the reviewers will be resolved through discussion, or with a third reviewer. Moreover, duplicate and non-eligible articles will be excluded and reasons for exclusion will be recorded. A PRISMA flow chart will be used to trace the overall process.

Data collection process
Two reviewers will extract data from the selected studies independently and crosscheck all articles for accuracy. Disagreements will be resolved by discussion between the reviewers or with a third reviewer.

Data items
We will extract data according to our predetermined form(as shown in Supplementary Table 1), containing five main categories: 1) Basic information of the included studies (e.g., study title, name of first author, year of publication, whether single-center or multicenter, country of study, total sample size); 2) Characteristics of the participants (e.g., gender, age, disease type, diagnostic criteria); 3) Intervention-related variables (e.g., type of strains, dosage, duration); 4) Outcome measurements (e.g., MMSE, TAC, GSH, MDA, hs-CRP, NO); 5) Main findings of each included studies. To perform the metaanalysis, the following data form of outcome assessments are to be extracted: the mean change score along with the associated variance (standard deviation [SD] or standard error of the mean [SEM]). When change scores are not available, the scores (mean ± SD or mean ± SEM) and the number of participants at baseline and post-intervention will be extracted.

Primary outcomes
The primary outcomes of this study are the standardized mean differences (SMDs) of change from baseline between probiotics and placebo group, including pooled estimate of cognition, malondialdehyde (MDA), highsensitivity C-reactive protein (hs-CRP), total glutathione (GSH) and nitric oxide (NO).

Secondary outcomes
Where available, secondary outcomes for this review will include subgroups such as cognitive rating scales (MMSE versus non-MMSE), disease type (AD versus MCI), strains of flora (multiple versus sole).

Risk of bias in individual studies
An assessment of risk of bias will be incorporated into our analysis. By assessing the quality of the studies that will be included in the meta-analysis, we can assess the strength of the body of evidence. In particular, information relating to bias will be extracted from each study during the data extraction process. This assessment will follow the same procedure with the data collection process where disagreements will be resolved by discussion between the reviewers or with a third reviewer. To facilitate the appraisal of possible risk of bias, paired reviewers will evaluate independently the risk of bias of included RCTs using the tool of the Cochrane Handbook for Systematic Reviews of Interventions, containing the following six criteria: random sequence generation, allocation concealment, blindness of participants and personnel, blindness of outcome assessment, incomplete outcome data, selective reporting, and other sources of bias. Furthermore, to evaluate the risk of bias, reviewers will assess and rate each of the above items into three ratings: "low risk of bias" or "high risk of bias" or "uncertain-risk". In particular, "low risk" indicates that further research is very unlikely to change the confidence of the estimates whereas "high risk" indicates that further research is very likely to change the estimate. The results of the risk of bias assessment will be pooled into Revman 5.3, and generate a "summary of risk of bias assessment" table.

Data synthesis
We will use STATA software (version 12; StataCorp) to perform the statistical analyses. The primary outcomes of this study are the standardized mean differences (SMDs) of change from baseline between probiotics and placebo group. The SMDs will be tested by a Z statistic, and a two-tailed P < 0.05 will be regarded as statistically significant. To determine the extent of variation between the selected studies, tests of heterogeneity will be performed. The inter-study heterogeneity was examined by chi-square (χ 2 ) statistics and I 2 statistics. Higgins and Colleagues provided tentative benchmarks for I 2 where values below 25% might be considered as low, 50-75% as moderate and above 75% as high. In our study, the heterogeneity among the different studies was considered high if P < 0.1 for the χ 2 statistic or I 2 > 50% [22]. A random-effects model was used if significant heterogeneity was shown among trials. Otherwise, the results were obtained from a fixed-effects model. Subgroup analyses will be performed to examine the possible source of heterogeneity within these studies.

Meta-bias(es)
We will use STATA software (version 12; StataCorp) to perform sensitivity analysis and publication bias. A sensitivity analysis will be conducted to test the reliability of the findings using the leave-one-out method, while publication bias will be assessed by Egger's test and Begg's test.