Background <p>Appetite hormones and proinflammatory cytokines play a role in differentiating between bipolar I disorder (BD1) and bipolar II disorder (BD2). In this study, we developed a composite predictor of appetite hormones and proinflammatory cytokines to differentiate between BD1 and BD2.</p> Methods <p>Adult patients aged 20–59 with either BD1 or BD2 and experiencing a major depressive episode were included in the study. Cytokines such as C-reactive protein, interleukin-2, interleukin-6, and tumor necrosis factor-α and appetite hormones such as leptin, adiponectin, ghrelin, and insulin were evaluated as potential predictors through a classification and regression tree (CRT) to differentiate between BD1 and BD2.</p> Results <p>A composite predictor of adiponectin, leptin, and ghrelin was significantly more accurate (for BD1: area under the curve = 0.897; for BD2: area under the curve = 0.905, <i>P</i> &gt; 0.05) in differentiating between BD2 and BD1 than any single predictor (four appetite hormones and six cytokines). High levels of adiponectin and ghrelin and high and low levels of leptin (≤ 4430.8 and &gt; 10,957.2 ng/L) were associated with BD2, whereas low levels of adiponectin and ghrelin and intermediate levels of leptin were associated with BD1.</p> Conclusions <p>The composite predictor of appetite hormones showed potential for distinguishing between BD1 and BD2 during depressive episodes. However, given the exploratory nature of the analysis and the limited sample size, further studies are needed to validate the model’s utility in clinical settings and to better understand the pathomechanisms underlying BD subtypes</p>