Efficacy Evaluation and Dose Prediction of a Novel Botanical Formula on Eye Fatigue and Dry Eye to Provide a Personalized Nutrition Solution. (29th May 2020)

• Record Type: Journal Article
• Authors: Kan, Juntao; Gu, Zhensheng; Du, Jun
• Is Part Of: Current developments in nutrition. Volume 4(2020)Supplement 2
• Page Start: 112
• Page End: 112
• Publication Date: 2020-05-29
• Subjects: Nutrition -- Periodicals; Nutritional Physiological Phenomena; Nutrition; Periodicals; Periodicals; Fulltext; Internet Resources; Periodicals; 612.3
• DOI: 10.1093/cdn/nzaa041_016
• Languages: English
• ISSNs: 2475-2991

Abstract

Abstract: Objectives: With the frequent use of video display units, eye fatigue is more and more common globally. We aimed to evaluate the protective effect of a novel botanical combination of lutein ester, zeaxanthin, extracts of black currant, chrysanthemum and goji berry on adults with eye fatigue in a randomized controlled trial, and predict personalized dose for each individual in real-life consumption. Methods: 360 participants were randomized to receive our formula with 3 different doses (containing 6 mg, 10 mg, and 14 mg of lutein, respectively) or placebo once daily for 90 days. Each participant had a total of 3 visits at baseline (V1), 45 days (V2) and 90 days (V3) throughout the study. Results: The formula intervention improved individual score of eye fatigue symptoms, including eye soreness, photophobia, blurred vision, dry eye, foreign body sensation and tearing. Compared with placebo, the formula with all 3 doses significantly decreased the total score of eye fatigue symptoms and increased the visuognosis persistence time at both V2 and V3. For Schirmer test, the formula with 10 mg and 14 mg of lutein improved the tear secretion at V3. For Keratography, first tear break time, average tear break time and tear meniscus height were significantly increased after formula intervention. The intervention of formula with all 3 doses significantly increased macular pigment optical density at V2 and V3, while optical coherence tomography showed that there was no significant difference in retinal thickness and retinal volume among all groups at both visits. Different machine-learning algorithms were used to predict personalized dose based on the eye-related indexes and other features, such as anthropometrics, physical activity, dietary habit, and blood biomarkers under 3 different dosages. Among them, XGBoost performed best with R = 0.649 for training set, R = 0.638 for test set, and R = 0.685 for external validation set. Conclusions: The formula improves eye fatigue, dry eye, and macular function without changing the structure, providing a nutritional alternative strategy. XGBoost could successfully predict dose to provide a personalized nutrition solution. Funding Sources: This study was supported by National Key R&D Program of China and Nutrition Research Foundation of Chinese Nutrition Society.