Test food
In previous studies, we reported that 300 mg of egg yolk choline per day may improve verbal memory, one of the typical cognitive functions [17]. Therefore, the test food contained 300 mg of egg yolk choline as lecithin derived from egg yolk (Kewpie Corporation, Tokyo, Japan) for the choline group, and contained yolk oil free from egg yolk choline (Kewpie Corporation, Tokyo, Japan) for the placebo group. Both test foods were soft capsule-shaped supplements manufactured by Aliment Industry Co., Ltd. (Yamanashi, Japan), and the taste, appearance, and nutrient contents were adjusted to be equivalent. Table 1 shows the nutrient composition of each test food. Participants consumed seven capsules per day with water after breakfast.
Table 1
Nutrient composition of each capsules (/7 capsule)
| | | Placebo | | Choline |
Energy | (kcal) | | 21 | | 22 |
Protein | (g) | | 0.7 | | 0.7 |
Fat | (g) | | 1.5 | | 1.5 |
Carbohydrates | (g) | | 0.4 | | 0.4 |
Salt equivalent | (g) | | 0.0 | | 0.0 |
Subjects
The participants were 60 middle-aged and elderly Japanese males and females aged 60–80 years who were not demented but had been aware of their forgetfulness or had been pointed out by others and who were judged not to be ill by an investigator and with 26 points or more on the Mini-Mental State Examination-Japanese (MMSE-J) screening test [18, 19]. Exclusion criteria included the following: a history of psychiatric disorders, cerebrovascular diseases, or other serious diseases; a Geriatric Depression Scale Short Version-Japanese (GDS-S-J) of ≥ 6 points in the screening test [20]; use of medicines that may affect the central nervous system, healthy foods that may affect cognitive function, and other medicines that may affect the test results; regular healthy food consumption; impaired in vision and hearing; impaired daily living activities; excessive smoking and alcohol consumption; extremely irregular dietary habits; allergies to the test food ingredients; participation in other studies; and those who were judged by the investigator to be inappropriate as participants.
The sample size was determined based on a previous study [17]. The number of samples needed was ≧ 39 when type 1 risk α is 0.05 and power 1-β is 80%. The number of participants in this study was 60, 30 per group, to ensure this number of people. Sample size calculations are according to Cancer research and biostatistics statistical tools (https://stattools.crab.org/). The investigator fully explained the purpose and content of the study to the participants and provided written consent based on the participants free will before conducting the study.
Study design
The study was conducted at KSO Inc. (Tokyo, Japan) under the supervision of a physician from August to December 2021. This 12-week, randomized, double-blind, placebo-controlled, parallel-group study divided the participants into the placebo and choline groups by the block-randomization method. Allocation-adjustment factors were age, sex, egg consumption, plasma-free choline level, Cognitrax short version total and verbal memory scores, MMSE-J score, and GDS-S-J score. The reason why verbal memory score was added to the adjustment factor is that the function was confirmed by previous research [17]. The allocation was conducted by the head of statistical analysis not involved in laboratory tests, and the allocation table was sealed by the controller and kept strictly sealed until the end of the study. Additionally, subjects, interventionists, and outcome assessors were all blinded.
The Cognitrax test [21] and Trail Making Test (TMT) [22] were performed for the assessment of cognitive function. MOS 36-Item Short-Form Health Survey (SF-36) [23, 24] and the Simplified Japanese Version of the WHO-Five Well-Being Index (WHO-5) [25] were performed to assess the quality of life (QOL) of participants before and after intake of capsules for 6 and 12 weeks. The cognitive function test was performed in a private room with one clinical psychologist per participant and no other interference so that an accurate test could be performed. Additionally, blood was withdrawn to analyze the plasma choline level at each time point. Dietary surveys with brief-type, self-administered diet history questionnaires (BDHQs) were performed [26, 27] and blood and urine samples were taken for safety evaluation before and 12 weeks after intake.
The study was conducted after obtaining approval from the Nihonbashi Cardiology Clinic Institutional Review Board (approval number: NJI-021-07-01, approval date: July 27, 2021) following the Declaration of Helsinki (the World Medical Association) and the ethical guidelines for bioscience and medical research in the spirit for human subjects (Ministry of Education, Culture, Sports, Science and Technology, Ministry of Health, Labour and Welfare, and Ministry of Economic, Trade, and Industry). The study protocol was pre-registered with the Clinical Trials Registry System (UMIN-CTR) (UMIN 000045050).
Primary endpoint
The Cognitrax test (Health Solution Inc., Tokyo, Japan), a computer-based cognitive test with a Japanese version developed by CNS Vital Signs, LLC. (USA) [21], was administered as the primary endpoint. Cognitrax comprised 7 test items (verbal memory [VBM], visual memory [VIM], finger tapping [FTT], symbol digit coding [SDC], Stroop test [ST], shifting attention [SAT], and continuous performance [CPT]), and the results were shown as 11 domain scores (composite memory, VBM, VIM, psychomotor speed, reaction time, complex attention, cognitive flexibility, processing speed, executive function, simple attention, and motor speed).
Secondary and other endpoints
TMT part A (TMT-A) and part B (TMT-B) were performed to assess attention and executive function. Participants were asked to draw a line from 1 to 25 in order in TMT-A and from 1 to 13 with a number and letters alternating with 12 letters after a number and letters in TMT-B, both in order as early as possible. The time taken to perform the test and the number of errors were assessed.
SF-36 consisted of 36 questions to measure eight health concept subscales (physical functioning, physical role, bodily pain, general health, vitality, social functioning, emotional role, and mental health). Additionally, physical component summary scores (PCS), mental component summary scores, and role/social component summary scores were calculated from eight subscales. WHO-5 consisted of five questions (cheerful and in good spirits, calm and relaxed, active and vigorous, wake up feeling fresh and rested, things that interest me) to assess the condition in the last 2 weeks, and each score and total score were calculated.
Plasma-free and fat-soluble choline levels were analyzed, respectively. 100 µL of plasma was mixed with 900 µL of methanol. Following sonication for 10 minutes, the samples were centrifuged at 15,000 rpm for 10 minutes at 4°C, and the supernatant was aliquoted. Filter filtration was performed immediately before the analysis, and plasma-free choline levels was performed by LC/MS/MS quantitative analysis (Waters, QTRAP4500, 40°C, A: ammonium acetonitrile /B:5 mM acetic acid [pH 4.0]), and fat-soluble choline levels was performed by LC/MS/MS (BRUKER, micrOTOF-QⅡ, 30°C, A: ammonium acetonitrile /B:5 mM acetate [pH4.0]). The analyses were conducted by the specialist in charge of the analyses of Kewpie Corporation (Tokyo, Japan).
BDHQ dietary surveys were conducted twice before and 12 weeks after intake to ascertain dietary status and variations before and during the study. BDHQ were administered self-answer type survey indicated the meals in the past month.
As a safety evaluation, we performed physical measurement (height, weight, and body mass index), physical examination (systolic blood pressure, diastolic blood pressure, and pulse), hematological tests (leukocyte count, erythrocyte count, hematocrit, and platelet count), blood biochemistry tests (total protein, albumin, total bilirubin, indirect bilirubin, alkaline phosphatase, aspartate transaminase, alanine transaminase, L-lactate dehydrogenase, gamma-glutamyl transpeptidase, total cholesterol, triglycerol, high-density lipoprotein-cholesterol, urea-nitrogen, creatinine, sodium, potassium, chloride, blood glucose, and hemoglobin A1C), urinalysis (protein, glucose, and occult blood reaction), and medical interviews. Blood and urine analysis were conducted at BML, Inc. (Saitama, Japan). Blood collection was performed after fasting for 4 h or more. The participants completed daily life diaries to investigate the status of consumption of the test food and the presence of adverse events.
Statistical analysis
The test results are presented as the mean ± standard error of the amount of change, and a hazard ratio of < 5% was judged as significantly different. Statistical analysis was performed using the computer software IBM Statistical Package for the Social Sciences version 28.0 (IBM Japan Co., Ltd. (Tokyo, Japan)). Cognitrax, TMT-A and B, and plasma choline levels were compared between the choline and placebo groups at different time points by a Student’s t-test. Mann-Whitney’s U test was used to compare SF-36 and WHO-5 between the choline and placebo groups at different time points. To evaluate food survey and safety assessments, the data obtained before and 12 weeks after were compared by Paired t-test.