The strain with the most new research, and the trial that found nothing

Akkermansia muciniphila now draws more research attention than any other organism in this catalogue. Among 825 newly indexed papers covering 84 strains, it accounts for 128 of them. The largest human trial in that set tested it directly, and found no effect on the outcome it was designed to measure.

Two studies, one strain, opposite answers

Two papers published this year on the same organism reached different conclusions, and both deserve to be read. A meta analysis of seven trials in children with functional abdominal pain found no consistent benefit from Limosilactobacillus reuteri DSM 17938 over placebo at follow up, with a modest reduction at end of treatment and weak underlying evidence. A separate randomized trial in children aged four to twelve reported that the same strain cut pain frequency and intensity substantially more than placebo at four and eight weeks.

Both can be read honestly at once. A single trial can be positive while the pooled picture stays unconvincing, and that is usually what an immature evidence base looks like from the inside. The useful posture is not to pick the result you prefer but to note that the question is open, and that anyone claiming it is settled is reading selectively.

What a probiotic adds, when something else is already working

One study design recurs often enough to be worth naming, because it makes attribution difficult by construction: a strain is tested on top of an intervention that already produces an effect. The cleanest example: twelve weeks of continuous energy restriction in men living with obesity reduced fat mass and leptin, raised ghrelin and improved eating behaviour measures. Those changes appeared regardless of whether participants also took a four strain probiotic, which added no measurable benefit beyond the calorie restriction itself.

The authors are careful about their own null, noting the analysis was underpowered for several of the hormonal measures, and that caution belongs here too. What the design does show clearly is the trap. When a supplement is tested alongside something that already works, a good outcome belongs to the combination until a trial separates them.

Where the newer human work is pointing

Two trials measured outcomes this field has historically studied mostly in animals. In older adults, six weeks of Lacticaseibacillus rhamnosus HN001 was associated with changes in brain connectivity and anxiety scores, with some effects still detectable four to six weeks after supplementation stopped. In surgical patients, an eighty four person trial run alongside standard care measured greater bone mineralization on CT at twelve weeks in the group receiving a multistrain preparation.

Both are single trials with modest samples, and neither establishes that these strains do these things. What they do establish is that the questions are now being asked in people rather than only in cell culture, which is the direction this literature has needed to move for a long time.

The shape of the rest

Most of the 825 papers are laboratory and animal work: 139 animal studies and 75 in vitro studies among those scored primary or secondary, against fifteen randomized trials and four meta analyses. A further 253 of the new strain links are passing references, where an organism appears as a comparator, a quality control target in an antimicrobial assay, or one taxon named in a list of many. Those are indexed because a complete index is more useful than a flattering one. They are not evidence about the strain, and they are not counted as such here.

On the question this catalogue exists to answer

One new paper bears directly on plastics. In zebrafish, combined exposure to a cyanobacterial toxin and polystyrene nanoplastics produced worse gut and neurological damage than either compound alone, and supplementing with Lacticaseibacillus rhamnosus GG reduced the resulting neuroinflammation by shifting gut bacterial composition. It is one study, in fish, of a strain given alongside an exposure rather than of a supplement taken by people. It does not show that any product does anything for anyone. It is also the only paper of its kind in 825, which is the more informative fact.

Four other papers placed strains alongside environmental contaminants. The strongest is a mouse study in which Akkermansia muciniphila was given during cadmium exposure, with the authors reporting reduced cognitive impairment. The rest cover an aflatoxin binding assay run in buffer and cheese rather than in any living system, silver nanoparticle effects on bacterial viability, and a flame retardant compound. Read together they describe a field that has barely begun asking this question. That is the honest state of the record, and reporting it accurately is worth more than a number that flatters the shelf.