Staging bipolar disorder: clinical, biochemical, and functional correlates

OBJECTIVE: There are several models of staging in bipolar disorder (BD), but none has been validated. The aims of this study were to empirically investigate clinical variables that may be useful to classify patients in clusters according to stage and study the association with biomarkers as biological validators. METHOD: This was a historical cohort study. Patients (n = 115) diagnosed with BD and not in an acute episode and first-degree relatives of patients diagnosed with BD (n = 25) were recruited. Sociodemographic, clinical, and functional data were collected. Serum cytokines, brain-derived neurotrophic factor, and biomarkers of lipid and protein oxidation were assessed. Cluster analysis was carried out to build a model of staging, and logistic regression was conducted to study associations between the model and biomarkers. RESULTS: Cluster analysis divided the sample into two equitable groups, denominated early and late stage, with empirical cutoffs for the Functioning Assessment Short Test score, number of episodes, age at onset of the disorder, and time elapsed since first episode. In the logistic regression, IL-6 was associated with late stage (P = 0.029). CONCLUSION: This study supports that clinical, functional, and biochemical variables may help to define a classification of staging in BD.