Risk of early onset dementia (EOD) in relation to fruit and vegetable consumption: a case-control study

Introduction Early onset dementia (EOD) is a condition characterized by the onset of severe cognitive impairment before 65 year 1. There is a general agreement that clinical presentation and conse- quences on daily life are different than late-onset dementia, due to the occurrence when subjects are still involved within occupational activities as well as familiar responsibilities 2. Previous studies have raised the possibility that the risk factors may also differ between the two forms, including the role of environmental and life-style determinants such as diet 3. There is some epidemiologic evidence of an association between dietary habits and protective relations to cognitive decline and incident dementia 4 5. The aim of this study is to evaluate, for the first time to our knowledge, the association between diet and EOD risk. Materials and methods Using a case-control design, we recruited newly-diagnosed EOD cases from 2016 to 2019 in Modena province, Northern Italy. We recruited controls from care-givers of dementia patients referring the Neurology Units of Modena-Baggiovara and Carpi Hospi- tals. We investigated dietary habits through a self-administered semi-quantitative food frequency questionnaire, designed and validated to capture eating behaviors in Italy, and specifically developed as part of the European Prospective Investigation into Cancer and Nutrition (EPIC) study for the Northern Italy pop- ulation 6. Participants were asked to respond to 248 questions about 188 different food items, in order to assess frequency and quantity of daily consumption for each food item. We specifically focused on consumption of fruits and vegetables that were categorized into major and sub-group categories as previously reported 7. We computed the odds ratios (ORs) of EOD risk, and the corresponding 95% confidence intervals (CIs), according to increasing tertile of intake using an unconditional logistic regres- sion model and we also modeled the relation using restricted cu- bic splines with three knots (10, 50 and 90%). We implemented a model adjusted for sex, age (years), educational attainment (years), and total energy intake (kcal/day). Results We recruited 56 (men/women: 22/34) and 53 (men/women: 24/29) controls who agreed to participate and returned study ma- terial. Regarding food intake, cases showed a lower intake than controls of vegetables (119 g/day versus 148 g/day), pulses (19 g/day versus 22 g/day), potatoes (15 g/day versus 23 g/day), and dry fruits (2.6 g/day versus 4.0 g/day). Conversely cases showed a higher intake than controls of fresh fruits (256 g/day versus 240 g/day). In the analysis of EOD risk according to increasing tertile we found an inverse association with all vegetables overall considered and also for individual subgroup, particularly leafy vegetables and other vegetables (e.g. eggplant, zucchini or sweet pepper). Also, an inverse association was found for potatoes and slightly for legumes. Conversely, the higher tertile of intake of fresh fruits was associated with no difference in EOD risk, but in the stratified analysis for non-citrus and citrus fruits we found inverse association for the latter. Interestingly, for non-citrus fruits we found inverse association in the second tertile of consumption, and an opposite positive association in the third tertile. Finally, with increasing tertile of dry fruits intake, we found a strong inverse association with EOD risk. Similarly in the spline analysis, the increasing consumption of vegetables, particularly leafy vegetables showed a strong inverse association, as simi- larly also increased pulse intake. Regarding fruit consumption, fresh fruit overall considered showed ambivalent relation, while higher consumption of citrus fruits only, as well as dry fruits, showed a negative association with EOD risk. Conclusions In this study, we found an association of consumption of vegeta- bles and to some extent fruit with EOD risk in an Italian popula- tion. Despite the study limitations, such as the possible occur- rence of selection bias, our findings suggest that dietary habits may influence EOD etiology and onset.