Benefits
Lumbar spine bone density after menopause
In a 12-month trial in 249 healthy early postmenopausal women in Sweden, one capsule a day kept lumbar spine bone density steady (-0.01 percent) while it fell 0.72 percent on placebo, the trial's main outcome. Probi funded it and employed two of the authors. In a second 12-month trial in 114 women, spine density fell slightly more on the blend than on placebo, though not after correction for multiple tests.
Shin and forearm bone density in a second 12-month trial
In 114 early postmenopausal women in Australia, bone density at the shin (distal tibia), the main outcome, fell slightly more on the blend than on placebo over 12 months (difference -3.7 mg HA/cm3). Spine density also fell more on the blend, though not after correction for multiple tests, and forearm loss was greater at borderline significance. Probi funded the trial, which a university ran.
Hip and femoral neck bone density
Neither trial found a benefit at the hip. The Australian trial found no difference from placebo at the total hip or femoral neck, and its authors describe the Swedish trial as showing no benefit at sites other than the spine. A news report of the Swedish trial gave a femoral neck loss of 1.34 percent on the blend against 0.88 percent on placebo.
Bone turnover and inflammatory blood markers
In the Australian trial, 12 months on the blend did not change bone turnover markers such as PINP and CTX, calcium-regulating hormones or inflammatory markers compared with placebo. Bone turnover markers are lab measures of bone remodeling and are not by themselves evidence of stronger bones.
Bone and lean mass in mouse studies (lab research)
In mice whose ovaries or testes were removed to mimic hormone-related bone loss, the same three strains given in drinking water limited cortical or trabecular bone loss, and in female mice they raised whole-body bone density and lean mass on a low-fat but not a high-fat diet. These are animal results, and the two human trials did not agree with each other.
Mechanism of action
Immune signaling in bone (animal research)
In ovariectomized mice, the blend lowered expression of the inflammatory signals TNF-alpha and IL-1beta in cortical bone, raised osteoprotegerin, which restrains bone-resorbing cells, and kept regulatory T cells in bone marrow from falling. A later mouse study again linked its bone effects to regulatory T cells. This has not been shown in people.
Short-chain fatty acids from the gut (animal research)
In male mice, the blend raised cecal levels of the short-chain fatty acids propionate and acetate, and of succinate and lactate, alongside its bone effects. In female mice the strains were recovered from cecal content, showing that they reached the gut, and the bone and lean mass gains depended on a low-fat diet.
Inflammation and vitamin K2 markers in people
In the 12-month Australian trial, inflammatory markers such as TNF-alpha did not differ from placebo, and neither did undercarboxylated osteocalcin, a marker of vitamin K activity. The authors concluded that a vitamin K2 route was unlikely for these strains.
Clinical trials
Randomized, double-blind, placebo-controlled trial at four centers in Sweden; one capsule a day of L. paracasei DSM 13434, L. plantarum DSM 15312 and L. plantarum DSM 15313 (10 billion CFU per capsule) or placebo for 12 months; registered NCT02722980 with lumbar spine bone density by DXA as the primary outcome; funded and sponsored by Probi AB, two authors were Probi employees, and two others, Ohlsson and Sjögren, are named inventors on a Probi patent application covering these strains (per the 2014 mouse study disclosure), and a news report says three authors were coinventors on a related patent (Jansson et al. 2019, Lancet Rheumatol)
249 healthy early postmenopausal women were randomized and 234 (94 percent) had the spine scans needed for the primary analysis.
Lumbar spine bone density fell 0.72 percent on placebo and was unchanged on the blend (-0.01 percent), a mean difference of 0.71 percent (95 percent CI 0.06 to 1.35). Adverse events were similar between groups. A news report of the trial gave femoral neck loss of 1.34 percent on the blend and 0.88 percent on placebo, and the later Australian trial's authors describe no benefit at sites other than the spine. Only one dose was tested.
Randomized, double-blind, placebo-controlled trial (Probone21); one capsule a day of L. plantarum DSM 15312 and DSM 15313 with L. paracasei DSM 13434 (10 billion CFU per capsule) or maize starch placebo for 12 months; registered ACTRN12621000810819 with distal tibia volumetric bone density by HR-pQCT as the primary outcome; funded by Probi AB and run by La Trobe University, with no Probi employees among the authors, who declared no conflicts and state the funder had no role in design, analysis or writing (Resciniti et al. 2026, Am J Clin Nutr)
114 healthy women 1 to 8 years after menopause (58 probiotic, 56 placebo), mean age about 56; women with osteoporosis-range T-scores were excluded.
Distal tibia volumetric bone density, the primary outcome, fell more on the blend than on placebo (difference -3.7 mg HA/cm3, 95 percent CI -6.9 to -0.4, p = 0.027). Lumbar spine areal density also fell more on the blend (p = 0.042), not significant after correction for multiple tests, and forearm loss was greater at borderline significance; total hip and femoral neck did not differ. Bone turnover, calcium-regulating and inflammatory markers did not differ. The authors conclude the blend does not attenuate bone loss.