The results of our study clearly indicate no association between the intake frequency of different food products and MS in the population of Biancavilla. At present there is no clear advice to follow a specific diet for people with MS, as confirmed by data coming from the NARCOMS registry, which investigated the prevalence of 19 different diets among 6990 MS patients without recognizing a specific alimentary regimen among patients (20).
Furthermore, despite some evidence of a proinflammatory role of the Western-type diet has been reported (2, 21), results are not univocal. Evidence from the literature reported a moderate correlation between prevalence of MS and total meat intake (r = 0.61, p < 0.01), particularly pork consumption (r = 0.97, p < 0.001) (12), while an OR = 2.07 (1.18–3.63) and an OR = 1.38 (1.13–1.68) were reported for processed and unprocessed meat (13). However, other studies reported no association between meat and MS (22, 23). In order to solve the ongoing debate, the association between the intake of different red and processed meat intake and the development of chronic inflammatory diseases, including MS, during a follow-up period of 25 years, will be evaluated in an ongoing study developed out of a Danish open register-based cohort study involving 57’053 persons (24).
With regard to fish intake, there is some evidence in literature about a protective role towards MS. An OR = 0.69 (0.55–0.87) (23) and a OR = 0.77 (0.64–0.92) (25) were reported in literature for the consumption of 0.5–2.99 fresh fish servings per week between cases and controls. These results have been confirmed in a more recent study by Hedström and coworkers, which assessed an association between an increased risk of MS and a low intake of fish (OR 1.2, 95% CI 1.1–1.4) (26).As regards the association between the intake of milk and dairy products and the development of MS, such hypothesis had been speculated several years ago, also relying on the evidence of molecular mimicry between butyrophilin and myelin oligodendrocyte glycoprotein (MOG) (15, 27). However, such a molecular mechanism was found to have possibly a non-pathogenic role and rather to prevent the development of the experimental autoimmune encephalomyelitis (EAE) through an upregulation in IL-10 secretion and a reduction in Th1 proliferation (28). A worldwide study, conducted in 1992, reported a correlation between cow milk consumption and MS prevalence with r = 0.84 (p < 0.001) and between butter intake and MS with r = 0.62 (p = 0.01). However, results have been inconclusive and conflicting here too, with some studies reporting a significant lower intake of dairy products in cases compared with controls (29). Finally, despite many patients frequently use alternative therapies and dietary supplements, the role of dietary intervention as a complementary treatment on MS outcomes is unclear and the strength of the evidence existing in literature is limited by the small number of the cohort and the retrospective nature of the studies (30). At the current stage, the best option is probably still represented by the Mediterranean diet, which is recommended by the American Heart Association (31), particularly for its beneficial effect in decreasing cardiovascular risk and improving glycemic control (32).
According to our results, the high incidence of MS in the town of Biancavilla (18) is likely to be not related to dietary habits. However, our study has several limitations. First of all, its retrospective case-control design makes it prone to selection and recall bias, and possibly to reverse causality, which should all be taken into account in interpreting analysis (33). Nevertheless, case-control studies are the most suitable in order to investigate environmental factors, since cohort studies are expensive and rarely available for low-incidence diseases as MS, requiring a very large study population and a long follow-up time to collect enough cases. Secondly, some possible confounders may not have been considered nor obtained from the collected data, as BMI during childhood and adolescence, smoking status, alcohol intake and the consumption of some foods excluded from the questionnaire, as fruits, vegetables, poultry and sugar. Finally, the small simple size does not allow us to generalize our results, but is due to the population-based nature of the study (18).