In addition to the formal systematic review of human studies, a narrative (non-systematic) approach was performed. In alignment with EFSA guidance, this narrative synthesis was used to integrate contextual data: (a) the characteristics of the fermented foods studied, (b) supportive evidence on the bioavailability of relevant compounds, (c) plausible mechanisms of action, and (d) the safety of fermented food consumption.

Characteristics of the fermented dairy products were systematically extracted into a harmonized table (Tables 2, 3), which included details on product type, origin, starter cultures, raw materials, manufacturing processes, chemical composition, microbial counts, and analytical methods.

Risk of bias was assessed only for studies meeting the PICO criteria. Among these, nine were parallel-group trials and five were crossover trials. In the overall assessment of the parallel-group trials (Figure 2), six of the nine studies were rated as having a high risk of bias (27, 30, 32, 33, 37, 68), two studies were judged to have some concerns (25, 35), and only one study was rated as low risk (34).

The domains primarily contributing to the high overall risk were the randomization process (absence of clear information) and the selection of the reported result. Eight of the nine studies raised some concerns regarding selective reporting, mainly due to the absence of or non-adherence to a pre-specified analysis plan. Bias due to deviations from intended interventions and missing outcome data was generally well controlled, with eight studies rated as having low risk in both domains. Notably, the measurement of outcomes was consistently rated as low risk across all nine studies, indicating reliable outcome assessment.

Five crossover trials were also assessed for risk of bias, incorporating an additional domain to assess bias arising from period and carryover effects. Four of the five trials were judged to have an overall high risk of bias (26, 29, 31, 33), while one raised some concerns (28). The randomization domain was the most problematic (e.g., the allocation sequence was not random), with four studies raising some concerns, and one study rated as high risk (33). Regarding period and carryover effects (Domain S), four studies were considered low risk, whereas one study (29) was rated high risk. Concerns regarding selective reporting were frequent and observed in four crossover trials. For missing outcome data, four studies were rated as low risk and one as high risk. Bias due to deviations from intended interventions and bias in the measurement of the outcome were consistently rated as low across all five studies (Figure 3).

These results indicate that while bias due to deviations from intended interventions and outcome measurement was generally well controlled in both parallel-group and crossover trials, issues related to the randomization process and selection of reported results were common. These methodological limitations underscore the need for improved rigor and transparency in the design and reporting of future trials.

Fermented dairy products may influence cardiovascular health through multiple interrelated mechanisms involving the dairy matrix, fermentation-derived microbial and biochemical processes, and host–microbe interactions.

The matrix of dairy products plays an important role in modulating their effects on lipid metabolism. For instance, cheese, due to its higher calcium and protein content, has been shown to attenuate the potential adverse effects of saturated fats found in dairy fat. When consumed in the form of cheese, dairy fat results in significantly lower LDL-c and total cholesterol levels compared to butterfat, suggesting that the food matrix itself has a significant effect on blood lipid profiles (46). Furthermore, the formation of calcium-fatty acid complexes and amorphous calcium phosphates in the duodenum is influenced by the dairy matrix, where fat in milk is present as small globules, while in cheese, fat is encapsulated by proteins such as casein. This difference in fat structure potentially facilitates a greater interaction between fat and calcium, leading to more efficient lipid metabolism (28, 69).

Fermentation processes in dairy products lead to the production of bioactive peptides and beneficial compounds that exert favorable effects on cardiovascular health. Fermented dairy products are rich in bioactive peptides, which are generated during the fermentation process and have been shown to exert antihypertensive and anti-inflammatory effects. These peptides can inhibit the angiotensin I-converting enzyme (ACE), thereby reducing blood pressure (11). Additionally, dairy products are rich in essential minerals, including calcium, potassium, and magnesium, which contribute to improved blood pressure regulation and overall cardiovascular health (11). In kefir, for example, bioactive peptides produced during fermentation may enhance the immune response by activating macrophages and increasing phagocytosis, which contributes to reduced inflammation and improved cardiovascular health (31). These peptides may also enhance insulin sensitivity and help regulate blood glucose levels, thereby indirectly reducing cardiovascular risk factors, such as metabolic syndrome. Fermented dairy, particularly kefir, may also contribute to the reduction of atherosclerosis-related inflammation. Studies indicate that kefir can lower the expression of cell adhesion molecules such as ICAM-1 and VCAM-1, which play key roles in the recruitment of monocytes during the early stages of atherosclerotic plaque development (42). This anti-inflammatory effect may be mediated by bioactive peptides, which can modulate immune cell activity and reduce chronic low-grade inflammation, a known risk factor for CVD (31).

The ability of probiotics to lower cholesterol can also be attributed to their production of short-chain fatty acids (SCFAs) during fermentation. SCFAs, including propionate and butyrate, can reduce cholesterol levels by blocking hepatic cholesterol synthesis and redirecting plasma cholesterol toward the liver. In addition, SCFAs can disrupt the enterohepatic circulation of bile acids by deconjugating bile salts, which further enhances cholesterol excretion (35). The presence of probiotics in yogurt (as detailed in Section 3.5) has also been linked to increased production of bile salt hydrolase, which facilitates the deconjugation of bile acids, promoting their excretion and contributing to cholesterol lowering (29).

Dairy consumption can also lead to an increase in SCFA production by gut microbiota, which plays a role in modulating cholesterol levels. Propionate, a predominant SCFA, has been shown to inhibit acetate’s cholesterol-generating effects, thereby reducing plasma cholesterol synthesis (70). However, studies suggest that kefir may not produce sufficient propionate to significantly affect cholesterol synthesis, indicating that the bacterial composition of fermented dairy products is an important factor in their hypocholesterolemic effects (25).

The cardiovascular health benefits of fermented dairy products are multifaceted, involving the dairy matrix, fermentation processes, probiotic strains, and bioactive components such as peptides, minerals, and SCFAs. The interaction between these factors contributes to the modulation of cholesterol levels, regulation of blood pressure, and reduction of inflammation, collectively supporting cardiovascular health.

Bioavailability is a critical factor in determining whether bioactive compounds found in fermented dairy products can exert physiological effects. According to EFSA, bioavailability encompasses the processes of release from the food matrix (bio-accessibility), absorption, distribution, metabolism, and elimination. In the context of this review, the focus is on evaluating whether compounds such as peptides, probiotics, SCFAs, and minerals present in fermented dairy products are accessible and available in the human body to potentially modulate blood lipid levels.

The definition of fermented milk (CAC/RCP 243, 2003) and cheese (CAC/RCP 283, 1978) have been clearly made by Codex Alimentarius. As shown in Table 2, yogurt and yogurt-based products (including yogurt drinks/beverages, smoothies and frozen yogurt) were the most frequently studied fermented milk products, examined in 30 of the included studies. Additional products evaluated (Table 3) comprised kefir (seven studies), cheese (10 studies) and other types of fermented milks (four studies). Most of the products used were either commercially available items (26 studies) or their source was unspecified (19 studies). A smaller number were developed in laboratory settings, more specifically three yogurts (71–73) and two kefir products (36, 42), while one study investigated a traditional homemade fermented milk product (koumiss) (43).

For yogurts and yogurt-based products, the most common starter cultures reported in the included studies were the protocooperation mix of S. thermophilus and L. delbrueckii subsp. bulgaricus, which are well specified by Codex Alimentarius Standard No. 243/2003. Alternative culture yogurts made by the starter culture of S. thermophilus and other Lactobacillus species were reported, including L. acidophilus (37, 40, 71, 74, 75), Lacticaseibacillus rhamnosus (37) and Lactiplantibacillus plantarum (76). In a few studies, Bifidobacterium infantis (75), B. lactis (72, 74), and Enterococcus faecium (37) were used. Kefir products were produced using defined starter cultures prepared from kefir grains (25, 36, 42), as indicated by Codex Alimentarius Standard No. 243/2003. They consisted mostly of species of the genera Leuconostoc and Lactococcus, L. kefiri, along with lactose-fermenting yeast (Kluyveromyces marxianus) and non-lactose-fermenting yeast (Saccharomyces unisporus). The starter cultures used for pitched kefir were previously isolated from traditional kefir grains, which naturally exhibit variations in the microbial composition and diversity (42). In addition to the above, the latter also included L. kefiranofaciens, Acetobacter pasteurianus and more stains of yeasts (Pichia fermentans, S. cerevisiae, Kazachstania unispora). No starter culture details were reported for cheese, except for one study where they did not add any starter cultures (51).

Despite the significant role of the origin of milk on the nutritional value and the organoleptic properties of fermented dairy products (77, 78), most of the included studies (43 out of 51) did not include this information; however, some studies reported it. Four yogurt samples (53, 73, 76, 79) one kefir sample (29) and one cheese sample (28) were made from cow’s milk, one yogurt sample (73) and one cheese sample (51) from ewe’s milk, and one cheese sample from goat’s milk (49).

When available, the manufacturing process of the fermented milk products involved the typical stages: pasteurization of milk (80–115 °C, 5–60 min), cooling at temperature optimum for fermentation (25–50 °C) depending on the type of microorganisms involved (e.g., mesophilic or thermophilic cultures), the addition of specific starter culture, incubation at an optimum temperature to control fermentation (35–42 °C, 7–24 h) resulting in acidification (pH of 4.2–4.7), with or without milk protein coagulation, pasteurization (optional), cooling (23–27 °C), packing and cold storage (2–6 °C). In the case of frozen yogurt production, post-fermentation addition of flavorings, stabilizers/emulsifiers and/or sugars is usually included, followed by freezing the mixture (78). The cheese-making process involves the following main general steps for all varieties of cheese: pasteurization of raw milk (optional), acidification by indigenous lactic acid bacteria, ‘backsloping’ culture or starter culture, acid or enzymatic (rennet) coagulation, optional post-coagulation processes (i.e., cutting, cooking (scalding), cheddaring, curd washing, stretching, molding/drainage, pressing, salting), ripening (maturation, optional), packing, storage. In two studies, reconstitution of milk by adding skimmed milk powder (9–12%) and/or gelatin (0.4% w/v) in water was reported to produce yogurt (71) and fermented milk product Gaio® (27). Additionally, in some cases, the fresh whole milk was subjected to a skimming process (33, 38, 52, 73, 79) or was supplemented with skimmed milk powder (75). In some cases, it was reported that regulatory production standards were followed, including French regulations for yogurt production (80) and PDO guidelines for ‘Queso de Murcia’ cheese production (49).

Out of the 51 included studies, 32 studies reported the quantitative determination of proximate composition, predominantly for yogurt, including moisture, total fat, total carbohydrate, crude protein, ash, dietary fiber, and energy. These reported values meet the required specifications for milk protein (min. 2.7%) and fat (less than 10% for fermented milk, kefir and koumiss and 15% for yogurt and alternate culture yogurt), as set by the Codex Alimentarius (CAC/RCP 243, 2003). Also, this standard stipulates a minimum of 0.6% acidity (lactic acid), a minimum of 10⁷ colony-forming units (CFU)/g of microorganisms (total microorganisms in the starter culture) and a minimum of 10⁶ (CFU)/g of labelled microorganisms. Notably, findings suggest that certain fermented dairy products may qualify for health-related nutrition claims; 17 yogurt-based products reported fat contents below the thresholds defined in Regulation (EC) No 1924/2006 for “low-fat” nutrition claims (i.e., ≤3 g/100 g for solid foods and ≤1.5 g/100 mL for liquids), indicating their potential eligibility for such labelling under EU regulations. Furthermore, four yogurts had saturated fatty acid content not exceeding the maximum of 1.5 g/100 g, meeting the criteria for a “low-saturated fat” claim. One kefir could meet the requirements for “low sugars” claim (i.e., ≤5 g/100 g for solids or ≤2.5 g/100 mL for liquids).

Despite the relevance of identifying and quantifying bioactive compounds, including peptides, exopolysaccharides, unsaturated lipids, vitamins and minerals, very few included studies addressed these issues. As reported in the study of Olmedilla-Alonso et al. (73), low-fat and whole ewe’s milk yogurts were richer in SCFA than whole cow’s milk yogurt (20.26–21.17 vs. 11.31 g/100 g fat). Both ewe’s milk yogurts had similar content of PUFA (2.45–2.75 g/100 g fat) and CLA (0.24–0.27 g/100 g fat) with those of cow’s milk yogurt, but a two-fold higher n-3 α-linolenic acid content (0.76–0.87 g/100 g fat) and a lower n-6: n-3 ratio (2.07–2.13). Data for minerals (calcium, magnesium and potassium) showed that ewe’s milk yogurts contained higher amounts of calcium (2012.2–2063.1 mg/kg) and magnesium (171.1–171.8 mg/kg) but lower amount of potassium (1243.0–1264.3 mg/kg) compared with the respective values for cow’s milk yogurt (1081.3, 84.3 and 1,380 g/kg, respectively). Calcium was reported in similar or even higher amounts in cheese samples (1,600 mg/kg for skimmed cow’s milk cheese, Gamalost® and 8,000 mg/kg for the Gouda-type cheese, Norvegia®) of the included studies (44) but lower amounts (~1,200 mg/kg) in kefir samples (29, 30). Magnesium and potassium were reported for Gamalost® (130 and 980 mg/kg) and Norvegia® (330 and 770 mg/kg) (44). Sodium content varied between fermented dairy products; it was reported to be 396 mg/kg of yogurt smoothie (72), 1,000 mg/kg of low-fat plain yogurt (34, 81), 240 mg/kg for Gamalost® and 4,020 mg/kg for Norvegia®. Between regular-fat and reduced-fat Danbo and cheddar cheeses higher contents of SFA, MUFA and PUFA were observed (16.75–20.25 vs. 8.25–14.00, 7.00–11.50 vs. 3.25–8.25 and 0.75–1.75 vs. 0.25–0.75 g/100 g cheese, respectively), possibly reflecting their difference in total fat content (47, 50). The four cheeses had similar levels of calcium and sodium (667.5–772.5 and 1500.0–1750.0 mg/100 g, respectively). Values for phenolic compounds and vitamins were rarely reported in the included studies. Total phenolic content of 2.1 mg/g dry weight was estimated in low-fat plain yogurt (81). In the study of Richelsen et al. (32), the content of vitamin E and C was reported in the fermented milk product Gaio® at concentrations of 5 mg/kg and 100 mg/kg, respectively. Peptides were not investigated extensively in the reviewed studies; Usinger et al. (27) stated that 100 mL of the fermented milk product Cardio04® contained 470 mg total peptides, 2.5 mg VPP and 1.1 mg IPP. In particular studies, attention was focused on cholesterol contents in yogurt (30–40 mg/kg or 50 mg/L) (37, 82), yogurt smoothie (44 mg/kg) (72), yogurt drink (50 mg/kg) (83), kefir (62 mg/L) (31), mozzarella cheese (600 mg/kg) (52) and the fermented milk product Gaio® (50 mg/kg) (32). Compositional data for some cheese samples were reported through detailed descriptions of their contribution to the macro-and micronutrients and key bioactive compounds of the intervention diets (28, 45, 46).

Quantification of the microbial populations in the final products was reported in 9 studies on yogurt (35, 37, 53, 71, 74, 75, 79, 80, 84), 1 study on yogurt smoothie (72), 3 studies on kefir (25, 41, 42) and 2 studies on the fermented milk product, Gaio® (32, 68). These studies demonstrated that the freshly prepared products typically contained between 10⁷ and 10¹⁰ CFU/g of starter culture microorganisms. Only a handful of studies examined the microbial stability of the product. In two studies on yogurt (71, 74), as well as in one on the fermented milk product Gaio® (68), the authors stated that bacteria count in the final product remained almost stable during 1-week storage at 5–7 °C. Georgakouli et al. (79) investigated the population of the added starter cultures (S. thermophilus and L. bulgaricus) and yeasts/molds throughout storage of yogurt for 2 months at 4 °C. The results differentiated between the two bacteria; population of S. thermophilus remained somewhat stable for 30 days (~9 log CFU/g) and then a drop of ~1 log was observed at the end of storage, whereas L. bulgaricus counts tended to slightly increase for 1 week (6–6.5 log CFU/g) and after 10 days a reduction was noticed to 4.5 log CFU/g. The latter generally attained a lower population during fermentation and storage compared to the former. It is important to note that in some mild acidifying yogurts, L. bulgaricus may be present but does not grow appreciably, and its metabolic activity can be reduced or absent (85). Although this feature is desirable in such yogurt types, it may affect the metabolome and nutritional properties of the final product (85). On the other hand, yeasts and molds presented the opposite trend and grew during storage from 1.5 to 7 log CFU/g.

A notable limitation observed in some included studies was the reliance on prior published data on compositional characteristics of the fermented dairy products used, which may introduce measurement bias. In the study of Santurino et al. (49), the physicochemical composition of goat cheese samples was not directly measured but inferred from earlier work (86). Also, the fatty acid composition of two cheeses samples were either supplied from the manufacturer (52) or in the case of Pintus et al. (51) inferred from previous publication (87). Oosthuizen et al. (38) obtained the fatty acid composition data for frozen yogurt from South Africa Food Composition Tables. Pražnikar et al. (29) described the composition of the microbial population of kefir based on previous research by Vardjian et al. (88). Similarly, in the study of Richelsen et al. (32), the microbial counts were inferred from previous investigations. The latter approach may be justified when the nutrients of interest are well-characterized and stable. For example, the study by Vardjian et al. (88) reported, demonstrated high microbial stability, as well as similar counts and composition of lactobacilli and yeasts in kefir grains and beverage, over 10 weeks of propagation across two separate laboratories, which yielded comparable results. Their findings support the view that a stable microbiome of kefir grains ensures reproducible and constant product quality. Also, compositional data for some cheese samples were reported through detailed descriptions of their contribution to the macro-and micronutrients and key bioactive compounds of the intervention diets (28, 45, 46).

In the included studies, despite the complex and diverse microbiota of fermented dairy products-particularly kefir, which is traditionally produced using kefir grains-detailed compositional characterization was largely lacking. Previous research has shown that the microbial composition of kefir can vary significantly depending on factors such as the type of milk used, geographical origin, and fermentation conditions (89, 90). Among the studies reviewed, only one employed quantitative Real-Time Polymerase Chain Reaction for microbial profiling of kefir (41), and only one reported species-level identification (29). As most products evaluated were commercially available, sensory acceptability was presumed; however, no specific sensory data were provided, except for a brief description of the organoleptic properties of a cheese product (49).

The insufficient reporting of batch-to-batch variation and the variability in analytical techniques in the included studies highlight the fact that rational evaluation and control of product quality consistency are essential to ensure efficacy and safety. Without sufficient data on batch-to-batch variation, a significant gap remains, particularly in studies that reused the same products by referencing earlier research. Moreover, compositional characterization is often lacking in the existing literature. Therefore, this review highlights the importance of addressing these issues in future clinical trials to ensure reproducibility and accurate interpretation of results.

This comprehensive approach enabled us to uncover the “tip of the iceberg” regarding the current state of knowledge in this field. More specifically, although insights from the literature allowed us to identify key knowledge gaps in product characterization, as reflected in Tables 1, 2, our synthesis reveals a critical limitation in the field: a lack of standardized, comprehensive profiling of fermented dairy products used in clinical and observational studies. Despite their long history of consumption and cultural relevance, many of the fermented dairy foods included in our review were insufficiently characterized in terms of microbial composition, biochemical properties, and production parameters.

To advance the field and enhance scientific rigor in future studies on fermented dairy products, we suggest implementing comprehensive product characterization and study design. This should include specifying the origin and type of milk, explicitly listing the starter cultures, identifying microbial species and strains (e.g., through metagenomic approaches), as well as quantifying viable counts both at production and at the end of shelf-life. Furthermore, studies should analyze and provide detailed report of macronutrients (e.g., fat, protein, carbohydrate), micronutrient (e.g., sodium, calcium, magnesium), and key bioactive components (e.g., SCFAs, peptides, CLA, PUFA, vitamins) for the final products using validated analytical methods. Clear documentation of processing parameters (e.g., pasteurization parameters, fermentation time/temperature, and storage conditions), along with assessment of batch-to-batch variation, particularly microbial stability, is essential to ensure reproducibility. These recommendations align with the EFSA guidance for substantiating health claims and incorporating such standardized information in future studies will enable more meaningful comparisons and support the identification of bioactive components driving cardiovascular effects.

As outlined above, the primary aim of our systematic review is to evaluate whether consumption of fermented dairy foods confers potential health benefits on blood lipid profiles in healthy adults. To address this, we have detailed relevant studies and their outcomes in the section titled “Identification of pertinent human efficacy studies” Both sections have been organized in accordance with EFSA guidelines to ensure clarity and consistency. To avoid redundancy, we do not reiterate detailed descriptions of the individual studies here. Instead, this section focuses on providing targeted, concise information aligned with EFSA criteria, as applied in the PIMENTO WG3 framework for structuring our manuscripts.

Twelve of the 51 included studies (24%) explicitly reported at least one adverse effect associated with the consumption of yogurt (34, 83, 92–94), kefir (30, 31, 36), cheese (48, 49), and other fermented milk products (27, 68).

Most of the reported adverse effects were mild, self-limited gastrointestinal symptoms. The most frequently reported side-effects were gastrointestinal complaints (borborygmi, loose stools, obstipation) (27, 34, 68, 83, 92), abdominal bloating, cramping and/or distension (25, 27, 30, 34, 48), nausea, vomiting, diarrhea, halitosis, and/or constipation (25, 31, 48, 49, 94). Severe bloating was reported only in the study of Fathi et al. (30) and led to the discontinuation of intervention in 17 participants before the first follow-up visit. Only one study of Jauhiainen et al. (48) mentioned adverse effects, headache and tiredness. No gastrointestinal side effects or systemic reactions were reported, and no participant withdrawals were attributed to safety-related concerns for the rest of the studies (39 studies) (26, 28–30, 32, 33, 35–47, 50–53, 71–76, 79–82, 84, 95–99).

The self-limited gastrointestinal symptoms likely align with the intake of live microbial cultures and fermentation by-products (e.g., short-chain fatty acids, carbon dioxide) formed during bacterial metabolism. These effects are generally dose-dependent and appear to reflect an adaptive colonization phase by microorganisms (probiotic organisms) rather than intrinsic toxicity (25, 29–31, 36). However, no information concerning the association between the fermentation process and the reported adverse effects was provided in the studies included in this review. Also, the vast majority of the included studies implicitly specified subgroups recommended avoiding the consumption of fermented dairy products by excluding participants with a self-reported history of lactose intolerance, allergy or other adverse effects to the dairy products (30, 51, 84, 93).

Only one of the 51 studies (2%) cautioned against excessive intake of whole ewe’s milk yogurt, stating that “The energy value of whole ewe’s milk yogurt may be a problem if the energy content of the diet is not controlled” (73). No other studies issued overconsumption warnings.

No additional restrictions related to the safety of dairy-fermented foods were reported in any of the reviewed studies. Concerning the overall evidence qualification, given the consistent demonstration of only mild, self-limited gastrointestinal effects in less than half of the studies, together with the large proportion of studies reporting no safety concerns, the safety evidence of fermented dairy products in healthy adults is reasonably evaluated as convincing and sufficient (22).