A randomized crossover trial of the effects of fermented milk on carotenoid absorption in healthy participants

2.1 Ethical approval

This trial was approved by the Chiyoda Paramedical Care Clinic Ethics Review Board (September 15, 2023, approval number: 23091502) which was composed of third parties not involved in the trial. Our research was conducted in accordance with the ethical principles stipulated by the Helsinki Declaration, for medical research involving human subjects (as announced by the Ministry of Education, Culture, Sports, Science and Technology and the Ministry of Health, Labour and Welfare) and was registered in the UMIN clinical registry system (trial ID: UMIN000052407, registration date: October 4, 2023). This trial was performed at the Chiyoda Paramedical Care Clinic (Tokyo, Japan). A screening test was conducted on October 12, 2023, and the intake tests were done on November 9 and 23 of 2023. Our clinical trial ended on March 31, 2024.

2.2 Study protocol and screening of study participants

The investigating clinicians thoroughly explained details of the study to the prospective participants based on the informed consent document. After confirming that the participants fully understood the information, the clinicians obtained written consent. Consenting participants subsequently underwent a screening test. This included a physical examination, physical assessment (height, weight, body mass index [BMI], blood pressure, and pulse rate), blood analyses (hematological, biochemical, and infectious disease tests), and measurement of plasma carotenoid concentrations. The participants were enrolled in the study if they met the study selection criteria (described below) and did not have any characteristics that would disqualify them according to the exclusion criteria.

2.3 Sample size calculations

The number of research participants was calculated from a preliminary trial (UMIN trial ID: UMIN000030890) which used the same fermented milk planned for consumption in this study. Using the difference in mean values and standard deviation, the effect size (Cohen’s d) for this study was estimated to be 0.91. In an experiment with 10 participants, the possibility of detecting a difference with an effect size of 0.91, at a significance level of 5%, gave us an 80% chance of detecting an effect when one existed. We determined the number of cases required to evaluate the difference in blood carotenoid concentrations for this study.

2.4 Selection and exclusion criteria

The selection criteria for study participants included those who had received a thorough explanation about the purpose and content of the research and were able to understand and voluntarily provide their consent in writing by exercising their free will. Participants were to be male subjects aged between 20 and 35 years at the time consent was provided, and individuals with a body mass index of 18.5 to <25.0 kg/m² at the time of screening.

The exclusion criteria disqualified from study participation those who [1] had consumed medicines, non-medicinal products, supplements, functional foods, or carotenoid-enhanced products at least twice per week in the month prior to the screening test, and those planning to continue consuming these products during the test period, [2] had severe or progressive illness or symptoms, [3] were diagnosed by a trial supervising physician, or trial sub-physician, as having abnormalities in lipid metabolism, glucose metabolism, liver function, or renal function, [4] had food allergies, [5] were intolerant to lactose, [6] had previously undergone gastrointestinal surgery, [7] had smoked within the month prior to the screening test, [8] had donated 200 mL of blood within the month prior to the initial screening or who, within three months prior, had donated 400 mL of blood, [9] participated in another clinical trial or a monitoring study within the month before the initial screening, or who were planning to participate in another clinical trial during the study period, or [10] were physicians or co-investigators responsible for the trial.

2.5 Allocation of test foods

The study food allocation manager, who belongs to an organization independent of other study personnel, randomly assigned the 10 selected study participants into two groups of 5 people using plasma β-carotene concentration as an allocation factor. Group A received carotenoids alone and group B received carotenoids + fermented milk.

2.6 Preparation of test drinks

The test drink product comprised 50 g of a carotenoid-containing drink and 50 g of fermented milk. The carotenoid drink was prepared by adding a carotenoid preparation to water. The carotenoid preparation included 5.0 mg of β-carotene, 4.0 mg of lycopene, 2.3 mg of lutein, 6.2 mg of astaxanthin and, 5.6 mg of crocetin (per 50 g serving).

The fermented milk was prepared by adding to a milk base product (concentrated skim milk), fermented milk starters (Lactobacillus delbrueckii subsp. bulgaricus OLL1251 and Streptcoccus thermophilus OLS3290). Fermentation continued until the count of each bacterium reached 1×10⁷/g. The nutritional composition of the fermented milk was as follows, per 100 g serving: energy, 35 kcal; carbohydrates, 4.9 g; proteins, 3.6 g; fat, 0.1 g; ash, 0.8 g; water content, 90.6 g; β-carotene, 0.01 mg; and exopolysaccharide, 8.0 mg.

2.7 Study design

The study was conducted as a two-group, two-period randomized crossover trial ( Figure 1). A 13-day washout period was required between each test day. The study subjects consumed prescribed meals (breakfast, lunch, and dinner) the day before the test. The nutritional components of the meal were 2280 kcal of energy, 332 g of carbohydrates, 57 g of proteins, and 67 g of fats. Only water was permissible as a drink. Participants were instructed to finish their meal before 21:00 h on the day before a test.

Figure 1. Flow chart of clinical trial.

On the day of the test, the participants attended the clinic without eating or drinking and were prohibited from consuming anything other than the designated food and drink items until after completion of all tests. Upon arrival at the facility, prior to consuming the test drink a physical examination, a compliance check, and a physical examination (including weight, BMI, blood pressure, and pulse rate) were performed. Administration of the test drink (50 g carotenoids + 50 g water, or 50 g carotenoids + 50 g fermented milk) was initiated between 9:00 and 10:00 h, followed by completion within 20 min of the start. Instructions were provided to ensure that ingestion times of the test drink did not differ significantly across the three periods. Blood was collected in ethylenediaminetetraacetic acid (EDTA) tubes and inverted gently several times before the test and, 2, 4, 6, and 8 h after ingestion of the test drink. Each collected blood sample was separated into plasma (1730 × g, for 10 min at 4°C) and was kept at ˗80°C until measurement. To avoid photodecomposition of carotenoids in the blood, the collection tubes were maintained in darkness until analyzed.

To ensure the safety of the research participants, water was provided from the time participants consumed the test drink through until just before the last 8-h blood collection. After the 4-h blood collection participants could consume an offered snack (300 g of rice with salt, 880 kcal, carbohydrates 203 g, and protein 13 g).

2.8 Restrictions and prohibitions during the study period

During the study period, the participants were asked to adhere to the following restrictions and prohibitions related to food, beverages, and other indulgences, such as: [1] refrain from making significant changes to one’s usual diet (in terms of meal quantity, meal content, snacking habits, etc.); [2] while the intake of beverages such as green tea and coffee remained unrestricted in terms of quantity and frequency, there could be no change in the volume or type of these items consumed; [3] no extreme dieting or overeating of any food ingredients, dishes, or drinks that were rich in carotenoids (i.e., green and yellow vegetables, seaweed, etc.); [4] alcohol consumption was generally permissible provided it did not exceed the participant’s usual amount or frequency, although, it was prohibited from the day prior to the start of the test day until the testing ended; and [5] smoking during the study period was prohibited.

Medicines, health foods—including functional foods, special use/indicated foods, and carotenoid fortified foods—quasi-drugs, and supplements were prohibited because of the potential risk of affecting the test results. Participants were also asked to refrain as much as possible from making significant lifestyle changes.

2.9 Evaluations — outcomes

We measured plasma concentrations of the carotenoids (i.e., β-carotene, α-carotene, lycopene, lutein, zeaxanthin, β-cryptoxanthin, astaxanthin, and total carotenoids) in the triacylglycerol-rich lipoprotein (TRL) fraction, and crocetin in all fractions. Total carotenoids were calculated by adding together β-carotene, α-carotene, lycopene, lutein, zeaxanthin, β-cryptoxanthin, and astaxanthin. We evaluated the change from pre-ingestion value to zero (delta value) and the incremental area under the curve (iAUC) as indicators of bioavailability.

The TRL fraction is a known indicator of postprandial carotenoid contents in plasma (van Vliet, Schreurs, and van den Berg 1995). Because of the high concentration of endogenous carotenoids present in blood, and thus the data from the TRL fraction was chosen for analysis. Crocetin was the carotenoid of choice for the kinetics analysis since it could be assessed in all plasma fractions.

2.10 Safety analysis

Symptoms, severity, outcome, and frequency of adverse events and side effects that occurred from the start to the end of the study period by all study subjects who had consumed the study substance at least once were recorded, and the association with the study substance was evaluated for all events.

2.11 Analysis of plasma carotenoids

To measure plasma carotenoids, 500 μL of plasma layered with 500 μL of saline solution with a density of 1.006 g/mL was ultracentrifuged at 15,500 × g, for 30 min, at 20°C to obtain a TRL fraction. The blood concentration of carotenoids in the total fraction and TRL fraction were analyzed by extracting carotenoids in accordance with previous reports, and using high-performance liquid chromatography (HPLC) (Morifuji et al. 2020).

The measurement of crocetin was conducted using the following methodology (Umigai et al. 2011): 50 μL of plasma was mixed with 200 μL of chilled methanol. After centrifugation at 12,000 rpm for 10 min, at 4°C, the supernatant was collected. The pellet was washed with a volume of 80% methanol, and the supernatant was collected following centrifugation. The collected supernatant was dried using a desiccator, then re-dissolved in 85% methanol to prepare the sample. Crocetin was quantified using HPLC (LC-20AT, SPD-M20A). All the analyses were performed on a 4.6-mm I.D. x 150-mm column (COSMOSIL Packed Column 5C18-MS-II). The mobile phase A consisted of 0.5% acetic acid and the mobile phase B methanol. The total running time was 5 min. The eluent flow was 1.0 mL/min, and the column temperature was set at 40°C. The mobile phase A was mixed with mobile phase B in a ratio of 15% to 85%, respectively, and eluted isocratically. The absorption wavelength was set to 425 nm.

2.12 Analysis of study drinks

Nutrients and carotenoid measurements were performed by Japan Food Research Laboratories, Tokyo. The moisture content was measured using the vacuum drying method, protein content was determined using a combustion procedure, fats were extracted for measurement using the method of Röse-Gottlieb, ash content was evaluated using a direct ashing method, and carotenoids were measured using High performance liquid chromatography (HPLC).

2.13 Statistical analysis

The per-protocol set (PPS) analysis was used for participants who fulfilled the compliance requirements. The changes in plasma carotenoid concentrations over time were analyzed using a two-factor repeated-measures ANOVA post-hoc paired Student’s t test. The iAUC over 0–8 h was calculated using the trapezoidal rule. Postprandial carotenoid concentrations were baseline-corrected using fasting values. The iAUC data for plasma carotenoid concentrations were analyzed using the paired Student’s t-test. Differences between groups were significant at P < 0.05.

3.1 Study flow

The flow chart related to enrolment, allocation, tracking, and data analysis of the study is shown in Figure 1. A screening test was conducted on 21 participants who consented to participate in the study to determine their eligibility. Four participants met the exclusion criteria. A further 3 participants withdrew from the study for personal reasons. Because an analysis of changes in the blood concentration of β-carotene could potentially produce errors when the intrinsic plasma carotenoid concentration is high, it was decided to select 10 subjects for the study, based on their lowest whole blood plasma β-carotene concentrations during the screening test, and allocated them to one of two groups (n = 5).

One individual discontinued study participation due to poor health condition during the experiment. The remaining nine selected study participants completed the entire research schedule. No participants had a compliance violation during the research study. The safety analysis set (SAS) and the full analysis Set (FAS) comprised all ten participants. The number of participants in the per protocol set (PPS) was nine.

Table 1 presents the characteristics of the study participants measured at baseline (PPS).

3.2 Outcomes

Figure 2 shows the changes for the TRL fractions of β-carotene, lycopene, lutein, astaxanthin, total carotenoids, and all fractions of crocetin. The plasma β-carotene concentration in the TRL fraction 4, 6, 8 h post-consumption significantly increased with the co-ingestion of carotenoids and fermented milk compared with the ingestion of carotenoids alone.

Figure 2. Change from 0 h in the plasma triacylglycerol-rich lipoprotein (TRL) fraction of (A) β-carotene, (B) lycopene, (C) lutein, (D) astaxanthin, (E) total carotenoid concentration, and (F) all fractions of plasma crocetin.

Blood was collected before ingestion and 2, 4, 6, and 8 h after ingestion of the test drinks (containing carotenoids or carotenoids + fermented milk).

Total carotenoids were calculated by adding together β-carotene, α-carotene, lycopene, lutein, zeaxanthin, β-cryptoxanthin, and astaxanthin.

Mean ± standard error (N = 9: per protocol set analysis)

* P < 0.05 versus a carotenoid mixture at a given time (using a repeated-measures two-factor ANOVA post-hoc paired Student’s t test for analysis).

In the TRL fraction—when comparing the concurrent intake of carotenoids and fermented milk with the intake of carotenoids alone—the plasma lycopene concentration became significantly higher 6 and 8 h after consumption of the test drink; the plasma lutein concentration became significantly higher after 6 and 8 h, the plasma astaxanthin concentration significantly increased 6 h post-consumption, and the total carotenoids concentration significantly increased 6 and 8 h after intake. The crocetin concentration in the all fraction of plasma was significantly higher 2 h after the carotenoids and fermented milk compared with the ingestion of carotenoids alone.

Moreover, the iAUC for the TRL fraction of the plasma β-carotene, lycopene, lutein, astaxanthin, total carotenoids and all fraction of plasma crocetin concentrations significantly increased with the co-ingestion of carotenoids and fermented milk compared with consumption of carotenoids alone ( Figure 3).

Figure 3. Incremental area under the curve (iAUC) for triacylglycerol- rich lipoprotein (TRL)-fraction of (A) plasma β-carotene, (B) lycopene (C) lutein (D), astaxanthin, (E) total carotenoid concentration, and (F) all fractions of plasma crocetin.

Blood samples were collected before ingestion and 2, 4, 6, 8 h after ingestion of the test drinks (containing carotenoids or carotenoids + fermented milk).

The iAUC data for plasma carotenoid concentrations were analyzed using the paired Student’s t test. Differences between groups were considered significant at P < 0.05.

Mean ± standard error (N = 9, per protocol analysis set).

Total carotenoids were calculated by adding together β-carotene, α-carotene, lycopene, lutein, zeaxanthin, β-cryptoxanthin, and astaxanthin.

3.3 Safety evaluation

No adverse events considered causally related to the consumption of the test drink occurred in this study.