Buttermilk: the fraction has the trials, the glass does not

Buttermilk carries the milk fat globule membrane, and that fraction has been pooled across 13 randomised trials in disease-free adults. At 19.8 to 5000 mg a day for 2 to 16 weeks it lowered fasting total cholesterol by 0.18 mmol/L (95% CI -0.30 to -0.07), LDL cholesterol by 0.12 mmol/L and apolipoprotein B by 0.05 g/L, at moderate certainty by GRADE. Those trials fed the fraction as an isolate or an enriched food against dairy matched for energy, so the number belongs to the fraction rather than to a glass of buttermilk. The same synthesis found no clear effect on HDL cholesterol, triacylglycerol, glucose, insulin or blood pressure, and grades that null as low certainty. Buttermilk's own trials disagree with each other. Forty-five grams a day for four weeks in 34 adults with LDL cholesterol below 5.0 mmol/L lowered serum cholesterol by 3.1 percent (p = 0.019) and triacylglycerol by 10.7 percent (p = 0.007), with LDL cholesterol at p = 0.057. Those 45 g were a supplement-style dose rather than a glass of the drink. Twelve weeks of 80 mL of traditionally prepared buttermilk at lunch in 92 mildly hypercholesterolemic adults lowered neither total nor LDL cholesterol.

The numbers, per 100 g

How much is actually in it

• The same buttermilk, from the same dairy, carried 81.30 mg of total polar lipids per 250 mL portion in winter and 69.73 mg in spring, a seasonal gap of about 14 percent. — Buttermilk samples used as the trial product, measured by LC-ESI-MRM-MS/MS, winter and spring sampling (n = 2 seasonal samples) Evidence D source
  Total polar lipids 81.30 plus or minus 1.95 mg per daily portion in winter against 69.73 plus or minus 2.23 mg in spring

What your body absorbs

• One egg a day raised total and LDL cholesterol in women by about 0.6 mmol/L over 12 weeks, and the same egg yolk stirred into a buttermilk drink raised them by about 0.3 mmol/L, which was no longer significant. — 97 volunteers over 12 weeks, 57 taking one extra egg a day, 20 taking one egg yolk in a buttermilk drink, 20 continuing one to two eggs a week (n = 97) Evidence B source
  Extra egg in women: total cholesterol +0.63 mmol/L and LDL +0.59 mmol/L (both P < 0.05); same yolk in buttermilk: +0.33 mmol/L (P = 0.66) and +0.31 mmol/L (P = 0.55)
• In laboratory digestion models, buttermilk matrices raised the measured bioaccessibility of fat-soluble compounds such as curcuminoids, beta-carotene and omega-3 fatty acids compared with non-dairy systems. — 17 in vitro gastrointestinal digestion reports, 13 of them with data usable for pooling, no human participants (n = 17 reports) Evidence D source
  Direction is consistent and compound-dependent, and the authors decline to give a single pooled effect size

What it does to your health

• Pooled across 13 randomised trials, 2 to 16 weeks of milk fat globule membrane intake lowered fasting total cholesterol by 0.18 mmol/L (95% CI -0.30 to -0.07) in disease-free adults. — Disease-free adults in 13 randomised controlled trials (14 articles), doses 19.8 to 5000 mg/day for 2 to 16 weeks, compared with isoenergetic dairy low in these components (n = 13 trials) Evidence A source
  Fasting total cholesterol -0.18 mmol/L (95% CI -0.30, -0.07); LDL cholesterol -0.12 mmol/L (95% CI -0.22, -0.03); apolipoprotein B -0.05 g/L (95% CI -0.09, -0.01)
• The same synthesis of 13 randomised trials found no clear effect of the milk fat globule membrane fraction on HDL cholesterol, triacylglycerol, glucose, insulin or blood pressure. — Disease-free adults in 13 randomised controlled trials of MFGM or milk polar lipids against isoenergetic dairy comparators (n = 13 trials) Evidence A source
  No clear effect on fasting HDL cholesterol, triacylglycerol, glucose, insulin or blood pressure
• In 34 adults, 45 g of buttermilk a day for 4 weeks lowered serum cholesterol by 3.1 percent and triacylglycerol by 10.7 percent against a nutrient-matched placebo. — 34 men and women with serum LDL cholesterol below 5.0 mmol/L at screening, mean LDL 3.8 mmol/L, 45 g/day of a supplement-style buttermilk dose rather than a glass of the drink, 4 weeks in each arm (n = 34) Evidence B source
  Serum cholesterol -3.1% (P = 0.019), LDL cholesterol -3.1% (P = 0.057), triacylglycerol -10.7% (P = 0.007)
• In the same 34 adults, 45 g of buttermilk a day for 4 weeks lowered systolic blood pressure by 2.6 mm Hg and plasma angiotensin converting enzyme by 10.9 percent. — 34 men and women with LDL cholesterol below 5 mmol/L and normal blood pressure below 140 mm Hg, 45 g/day of a supplement-style buttermilk dose rather than a glass of the drink, 4 weeks in each arm (n = 34) Evidence B source
  Systolic blood pressure -2.6 mm Hg (P = 0.009), mean arterial pressure -1.7 mm Hg (P = 0.015), angiotensin I-converting enzyme -10.9% (P = 0.003)
• In 92 mildly hypercholesterolemic adults, 12 weeks of daily traditionally prepared buttermilk did not lower serum total or LDL cholesterol. — 92 mildly hypercholesterolemic adults in four arms, 80 mL of traditionally prepared buttermilk or skimmed milk at lunch with or without lutein-enriched egg yolk, 12 weeks (n = 92) Evidence B source
  No reduction in total cholesterol (trend P = 0.077) or LDL cholesterol (P = 0.204), and no blunting of the egg yolk effect

Safety, limits and interactions

• In an Austrian outbreak investigation where 29 percent of 362 residents of a rehabilitation clinic fell ill, a toxin-producing Bacillus cereus strain was isolated from the creme spinach and the strawberry buttermilk alike. The one culture-positive patient had eaten both dishes and carried the same strain. — Three 2013 food poisoning outbreaks in Austria, the buttermilk finding in outbreak C among 362 residents of a rehabilitation clinic, 29 percent of whom fell sick (n = 362 exposed in outbreak C) Evidence C source
  Enterotoxigenic B. cereus of the highest-risk phylogenetic group recovered from strawberry buttermilk and from the stool of the one culture-positive case-patient
• Among 1043 food samples tested in Babol, northern Iran, in 2019, the buttermilk and yogurt samples grew no Escherichia coli, no Staphylococcus aureus and no other target organisms. — 1043 food samples from preparation and distribution centres in Babol, northern Iran, across five food groups, 2019 (n = 1043 samples) Evidence C source
  All target organisms negative in the buttermilk and yogurt samples

What people believe that the data does not support

• In United States law the carton labelled cultured buttermilk is a cultured milk under the standard of identity 21 CFR 131.112, which is why it is a fermented drink rather than the liquid left from churning butter. — Federal standard of identity for cultured milk, United States Evidence E source
  The nomenclature paragraph permits the name cultured buttermilk for a cultured milk made with lactic acid-producing organisms

Who this works differently for

• In 61 perimenopausal women in Latvia, 28 days of 250 mL fermented buttermilk a day left total cholesterol 0.485 mmol/L higher (95% CI 0.023 to 0.948) and LDL cholesterol 0.330 mmol/L higher (95% CI 0.024 to 0.636) than in the control group. — 61 perimenopausal women, 31 drinking 250 mL of fermented buttermilk daily for 28 days and 30 controls, winter and spring cohorts (n = 61) Evidence B source
  Total cholesterol beta 0.485 (95% CI 0.023, 0.948) and LDL cholesterol beta 0.330 (95% CI 0.024, 0.636) higher in the intervention group after adjustment

The bottom line

The blood pressure result has the same problem. In those same 34 adults, all of them normotensive, 45 g a day for four weeks lowered systolic blood pressure by 2.6 mm Hg (p = 0.009) and plasma angiotensin converting enzyme by 10.9 percent. The pooled synthesis of 13 trials of the same fraction found no clear blood pressure effect, so that single number should not stand alone. A newer trial runs the other way on lipids. In 61 perimenopausal women in Latvia, 28 days of 250 mL of fermented buttermilk a day left total cholesterol 0.485 mmol/L higher (95% CI 0.023 to 0.948) and LDL cholesterol 0.330 mmol/L higher (95% CI 0.024 to 0.636) than the control group after adjustment. That was a secondary, exploratory analysis rather than the trial's primary endpoint, and the trial was registered only retrospectively. Put the set together and a daily glass has not been shown to move cholesterol in either direction. Two findings are about what buttermilk does to other food. In 97 volunteers over 12 weeks, one extra egg a day raised total cholesterol by 0.63 mmol/L and LDL cholesterol by 0.59 mmol/L in women. The same yolk stirred into a buttermilk drink raised them by 0.33 and 0.31 mmol/L, which was no longer significant in an arm of 20. A lost significance in 20 people is a hint that the food around the cholesterol matters, not evidence that the drink blocked anything. Seventeen laboratory digestion reports, with no human participants at all, point the same way for curcuminoids, beta-carotene and omega-3 fatty acids, and their authors decline to give a single pooled number. The product itself is not one constant thing. The same buttermilk from the same dairy carried 81.30 mg of total polar lipids per 250 mL portion in winter and 69.73 mg in spring, from two seasonal samples of one product, in work the authors call exploratory. The name is looser still. In United States law the carton labelled cultured buttermilk is a cultured milk under 21 CFR 131.112, a fermented drink made with lactic acid-producing organisms, while the trials above used traditionally prepared buttermilk, the liquid left from churning butter. On safety this page holds two records pointing different ways. In an Austrian investigation of a 2013 outbreak at a rehabilitation clinic, where 29 percent of 362 residents fell ill, an enterotoxigenic Bacillus cereus of the highest-risk phylogenetic group was recovered from the creme spinach and the strawberry buttermilk alike, and from the stool of the one culture-positive patient, who had eaten both. Isolation is not attribution, and the dishes carrying the statistical risk in that outbreak were others. Among 1043 food samples tested in Babol, northern Iran, in 2019, the buttermilk and yogurt samples grew no Escherichia coli, no Staphylococcus aureus and nothing else on the target list. The report does not say how many of those samples were buttermilk. Neither figure is a rate of risk, and both say only that a cultured dairy drink is not a sterile food. We found no trial in pregnancy, in children or in kidney disease, so anything beyond that is not something this page can tell you.

Sources

1. btml-r2-09 · food composition analysis
   The winter sample provided an estimated 81.30 ± 1.95 mg of total PL per daily portion, whereas the spring sample provided 69.73 ± 2.23 mg.
   Nutrients, 2026 · checked 2026-09-24
2. btml-r2-06 · randomised parallel trial
   The increase in serum TC and LDL-C concentration was no longer significant in women consuming the same egg yolk incorporated in a buttermilk drink (0.33 mmol/L [P = 0.66] and 0.31 mmol/L [P = 0.55], respectively).
   Nutrition, 2013 · checked 2026-09-24
3. btml-r2-07 · in vitro digestion studies
   We identified a limited but growing body of in vitro evidence indicating that buttermilk-based matrices generally increase the intestinal bioaccessibility of curcuminoids, β-carotene, omega-3 fatty acids, vitamin and MFGM phospholipids...Methodological heterogeneity and the exclusive reliance on in vitro models are important
   Molecules, 2026 · checked 2026-09-24
4. btml-r2-01 · systematic review and meta-analysis of RCTs
   Pooled analyses demonstrated that 2 to 16 wk of MFGM/MPL intake (19.8-5000 mg/d) reduced fasting total cholesterol [weighted mean difference (WMD): -0.18 mmol/L; 95% confidence interval (CI): -0.30, -0.07]
   Adv Nutr, 2026 · checked 2026-09-24
5. btml-r2-02 · systematic review and meta-analysis of RCTs
   No clear effects were observed for fasting circulating HDL cholesterol, triacylglycerol, glucose, insulin, or blood pressure (low certainty).
   Adv Nutr, 2026 · checked 2026-09-24
6. btml-r2-03 · randomised crossover trial
   Consumption of buttermilk led to reduction in serum cholesterol (-3.1%, P = 0.019), LDL-C (-3.1%, P = 0.057) and triacylglycerol (-10.7%, P = 0.007).
   Nutr Metab Cardiovasc Dis, 2013 · checked 2026-09-24
7. btml-r2-04 · randomised crossover trial
   Buttermilk consumption significantly reduced systolic blood pressure (-2.6 mm Hg; P = 0.009), mean arterial blood pressure (-1.7 mm Hg; P = 0.015), and plasma levels of the angiotensin I-converting enzyme (-10.9%; P = 0.003) compared with the placebo, but had no effect on plasma concentrations of angiotensin II and aldosterone.
   Nutrition, 2014 · checked 2026-09-24
8. btml-r2-05 · randomised parallel trial
   In mildly hypercholesterolemic subjects, daily consumption of traditionally prepared buttermilk for 12 weeks did not lower serum total-C or LDL-C concentrations, nor did it prevent the serum total-C and LDL-C raising effect of daily egg yolk consumption.
   Nutr Metab Cardiovasc Dis, 2015 · checked 2026-09-24
9. btml-r2-10 · outbreak investigation
   An enterotoxigenic strain affiliated to the phylogenetic group with the highest risk of food poisoning was isolated from the crème spinach and the strawberry buttermilk, and also from the stool samples of the one B. cereus positive outbreak case-patient, who ate both.
   Int J Food Microbiol, 2016 · checked 2026-09-24
10. btml-r2-11 · cross-sectional food quality survey
    The results of microbial tests on buttermilk and yogurt, grilled meat and chicken sandwiches, vegetables and salads showed that bacteria Escherichia coli and Staphylococcus aureus and all microorganisms were negative.
    J Health Popul Nutr, 2021 · checked 2026-09-24
11. btml-r2-12 · regulation
    when characterizing ingredients such as those in paragraphs (d) (6), (7), (8), and (9) of this section, and lactic acid-producing organisms are used the food may be named 'cultured buttermilk'.
    21 CFR 131.112, Cultured milk, eCFR current text · checked 2026-09-24
12. btml-r2-08 · secondary analysis of an RCT
    After adjustment for baseline lipid concentrations, season, and estimated total dairy PL intake, intervention-group allocation was associated with higher post-intervention TC (β = 0.485; 95% CI: 0.023-0.948) and LDL-C (β = 0.330; 95% CI: 0.024-0.636) than in the control group.
    Nutrients, 2026 · checked 2026-09-24