Platelet’s Fatty Acids and Differential Diagnosis of Major Depression and Bipolar Disorder through the Use of an Unsupervised Competitive-Learning Network Algorithm (SOM)

Aim of the study: The main purpose of this work was to verify, through the study of three fatty acids of platelets (namely: Palmitic Acid—PA, Linoleic Acid—LA, Arachidonic Acid—AA) the ability to distinguish adult subjects with Major Depression (MD) from those with Bipolar Disorder (BD), using an artificial neural network (Self Organizing Map—SOM) and an indirect index of the viscosity of the membrane (B2), in agreement with previous results; secondly, the ability to understand any similarities between children and adults in the molecular characterization of mood disorders, both in general and in relation to a subset of individuals with suicidal ideation, indicated by the survey instruments used (SOM). Design: The study design, in order to achieve the objectives, has forecast the recruitment of three groups of subjects without regard to sex, age, food intake patterns, or pharmacological therapies: 1) Controls; 2) Subjects with Major Depression; 3) Subjects with Bipolar Disorder. They were provided for the following investigations: 1) Platelet Fatty Acids analysis; 2) Plasma and platelet serotonin levels; 3) Oxidative stress and inflammation markers. Moreover, the data of the fatty acids of platelets, previously obtained by a group of children were used, for comparison, in the SOM, with adults surveyed. Subjects participating in the study: All participants, volunteer, were recruited in the judgment of psychiatrists as they presented themselves to the office visit of the “Dipartimento di Salute Mentale”, ASUR 4, Fano, Italy. All subjects were submitted to a semi-structured interview based on DSM-IV-TR criteria. 21-item Hamilton Depression Rating Scale (HDRS), Hamilton Anxiety Rating Scale (HARS), Clinical Global Impressions Scale (CGI), Symptom Checklist-90 (SCL-90), and Hypomania/Mania Symptom Checklist (HCL-32) were also applied to each patient. Adult subjects who did not give consent, patients with a diagnosis other than an affective disorder and subjects in their first clinical episode were excluded. The study was double blind. Outcomes: The results obtained have confirmed the main objective of the study. It was possible to obtain, in fact, the recognition of individuals with Major Depression and Bipolar Disorder, using the SOM and the index B2. Value of the study: The methodology used in this study may be of utility, such as quantitative diagnostic support to the psychiatrist, in order to reduce the high error that occurs in the first diagnosis, with regard to mood disorders. Limitations of this study: The limitations of this study are mainly related to the number of controls. They should have been more numerous such as the number of suicidal ideations. About the comparison with children, the main limitation seems to be the lack of psychiatric diagnosis of children, so that any results just assume the value of hypotheses. Future Research: Because of the strong classificatory properties of the SOM, it would require a RCT in a larger sample of subjects with mood disorders with the aim of assessing whether it is possible to identify subgroups of subjects with respect to the classical psychiatric classification and corresponding to different therapies. The same is for suicidal ideation.