The purpose of this study was to assess the effect of a typical micronutrient supplement on specific blood and tissue nutrient levels as the primary outcome and on general nutritional status (assessed by MNA) as the secondary outcome in cognitively impaired, independently living older adults in a routine setting.
We did not see overt malnutrition in our patient population; however, several epidemiological studies have shown that in cognitive impairment and AD, lower levels of micronutrients within the normal range are observed than among controls, suggesting that there is an additional need for specific micronutrients. For zinc, we observed a micronutrient deficiency at baseline only in men. Our data show, however, that a significant improvement in selected micronutrients can be achieved after taking a micronutrient supplement for 2 months.
Effect of vitamin supplementation on micronutrient concentrations in blood and intracellular tissue
Our data are consistent with those of others in showing decreased zinc levels in men with mild cognitive impairment . This could be explained by the fact that men have a higher daily zinc requirement (10 mg/day) than women (7 mg/day). Nonetheless, after supplementation, plasma zinc levels did not increase. One possible reason could be decreased zinc absorption efficiency in elderly persons . Inasmuch as a single capsule contains only 3.75 mg of zinc (54% of the daily requirement for women and only 38% for men), this amount might not be sufficient. A second explanation might be that the study period (2 months) might have been too short for zinc regarding its turnover time. One study has shown that improving the zinc status of older adults (> 65 years) with low zinc levels by using a supplement (7 mg zinc) for 1 year resulted not only in a decreased incidence and duration of pneumonia, but also in a decreased number of new antibiotic prescriptions and decreased total days of antibiotic use . In addition, normal baseline serum zinc concentrations were associated with a reduction in all causes of mortality.
In our study, the micronutrient status of vitamin E, β-carotene, and lycopene in BMC was within normal limits in most subjects before and after supplementation. With regard to vitamin C, only seven BMC concentrations were above detection levels despite normal vitamin C values in plasma at baseline. After intervention, vitamin C was above detection levels in all patients, indicating that 150 mg (250% of recommended dietary allowance [RDA]) is sufficient. The plasma and BMC concentrations of β-carotene and lycopene were significantly correlated before and after supplementation. Similarly, a study by Peng et al. also found a correlation between these micronutrients in plasma and BMC . On the other hand, Pateau et al. found no correlation between the lycopene concentration in plasma and BMC . Because the recommended daily amount of β-carotene is 2–4 mg and no recommendation for lycopene exists, and because the supplement contained only 0.5 mg of carotenoids (<25% RDA), our data may support the need for increased levels of carotenoids and lycopenes in supplements. Plasma and BMC concentrations of vitamin E were significantly correlated before supplementation, similar to the findings reported in Peng et al. . Following supplementation, this correlation was lacking, indicating that BMC were completely saturated after supplementation, whereas plasma concentration is not saturable. This may suggest that the level of vitamin E in the supplement with 200% RDA is more than sufficient.
Although the subjects had normal B vitamin levels in blood, 15 (51.7%) displayed increased Hcy levels before supplementation, indicating that blood levels of vitamin B6, vitamin B12, and folic acid inadequately reflect intracellular supply. We observed an improvement in B vitamin status and a subsequent decrease in plasma Hcy. Hcy is not only a risk factor for vascular disease, but also for cognitive disorders . Quadri et al. found an association between dementia and hyperhomocysteinemia . Many studies found that elderly people in particular have a deficiency in their supplies of vitamin B12 and folic acid . In our study, most blood levels for vitamins B1, B6, and B12 and for folic acid were normal before supplementation. After supplementation, these levels increased significantly. This is in line with the amount of vitamins contained in the supplement itself: B1 (200% RDA), B6 (>300% RDA), B12 (84% RDA), and folic acid (125% RDA). Whereas there was no significant change in the concentration of MMA, holo-TC increased significantly (P < 0.01), indicating that the intracellular status could still be optimized. Thus, an increase in the amount of vitamin B12 in the supplement may help to achieve better efficacy. However, analysis of the required amount of supplementation requires further study with several levels of supplementation.
Despite there being normal vitamin B1 concentrations in the blood, the low ETKA in 18 (51.4%) subjects indicated B1 deficiency in tissue. This finding was supported by an elevated TPP effect of more than 20% in some of our patients. Decreased ETKA might not be the best indicator in the elderly population owing to apoenzyme variations. In a study comparing ETKA and the TPP effect in healthy elderly subjects (70–82 years of age) with those in young subjects (19–37 years of age), researchers found that the older adults had a lower ETKA but a similar TPP effect and erythrocyte thiamine levels . Although the vitamin B1 concentration in blood increased significantly (P < 0.001) after supplementation in our study, ETKA did not increase. The TPP effect, however, decreased significantly, indicating improved intracellular vitamin B1 status and confirming the superiority of measuring the TPP effect.
Taken together, our results showed decreased Hcy levels, a decreased TPP effect, and improved antioxidant status after micronutrient supplementation. This might be helpful for improving cognitive function. The FACIT study, which showed beneficial effects on global cognitive function in men and women (aged 50–70 years), achieved decreased Hcy levels following 3-year supplementation with folic acid . A multivitamin, multimineral-enriched drink administered over a 6-month period decreased Hcy levels, increased B12 levels, and improved neuropsychological performance in frail elderly persons (N = 101; age > 65 years; BMI <25) . Supplementation with a multinutrient drink for 12 weeks improved memory (delayed verbal recall) in patients with mild AD . In our study, the period of supplementation was probably too short to have a positive effect on cognition as measured by MMSE.
Effect of vitamin supplementation on nutritional and health status
Micronutrient supplementation was associated with a significant increase in the MNA score in persons at risk for malnutrition. This improvement was mainly attributable to improved self-perception of general health status after micronutrient supplementation. The five dropouts in the MNA analysis may also have affected this result, especially considering the open-label character of the study. The MNA is a well-established assessment tool for the identification of patients at risk for malnutrition regardless of cognitive condition  and has also been used in interventional studies for follow-up evaluation [41, 42]. The MNA questions that showed significant intercorrelations in the study by Soini et al.  formed logical patterns between anthropometrics, decline in food intake with weight loss, self-perceived nutritional status, the ability to eat, and the use of fruit and vegetables. However, the results of the MNA should not be overemphasized, because the aim of this supplementation was not the improvement of energy and protein deficiency. Other studies have shown that subjective assessment of health does have predictive value . Christensson et al.  concluded that “self-experienced health status” has the greatest predictive value in MNA classification. One reason they discussed was the possibility of a reduced incidence of infections. Barringer et al.  found a significantly reduced incidence of infections in patients with diabetes (age > 65) after 1 year of multivitamin and mineral supplementation. In Christensson et al. , quality of life was assessed as a secondary endpoint by the Medical Outcomes Study 12-Item Short Form, which showed no difference before and after supplementation. A significant risk reduction for infections was shown after 18 months of multivitamin supplementation in a subgroup of patients without dementia . We did not assess infection rate in the present study. We propose that in future trials, an assessment for infections and for comprehensive quality of life should be incorporated.
Strengths and limitations
The main strength of this study is its comprehensive approach for the determination of blood and intracellular nutrient status, as well as assessment of metabolic markers before and after supplementation, in addition to general nutritional assessment. To our knowledge, the comprehensive and innovative methods used herein for micronutrient analysis are novel.
One limitation is that we did not analyze all of the micronutrients contained in this trial supplement. In addition, the generalizability of our results is limited because of the sample size. Moreover, without parallel controls, the primary outcome could not be separated from a time effect and the secondary outcome could not be separated from a placebo effect. A major limitation regarding self-assessed health status is the open-label character of this study, which could have biased the subjective assessment of the participants.