Abstract
Background:
Bipolar depression is frequently misclassified as unipolar depression during the early course of illness, which may delay appropriate treatment and worsen prognosis. Identifying objective and accessible biological indicators that support early differentiation remains a major clinical challenge.
Methods:
We conducted a retrospective case-control study based on hospitalized patients from a large psychiatric center in Beijing between 2015 and 2020. Patients were classified as unipolar depression or bipolar depression according to longitudinal diagnostic information derived from repeated admissions and ICD-10 criteria. To reduce demographic confounding, propensity score matching was applied for age and sex. Peripheral thyroid-related hormones and routinely available immune-inflammatory indices obtained at first hospitalization were analyzed. Logistic regression models were constructed to evaluate endocrine markers, immune-inflammatory markers, and their combined contribution to diagnostic discrimination. Model performance was examined using receiver operating characteristic analysis.
Results:
After matching, patients with bipolar depression exhibited lower levels of several thyroid hormones and selected inflammatory indices, alongside higher monocyte-related ratios, compared with unipolar depression. In multivariable analyses, total triiodothyronine showed an inverse association with bipolar depression, whereas monocyte-related inflammatory measures demonstrated positive associations. A combined model integrating endocrine and immune-inflammatory indicators demonstrated improved discriminative performance compared with models based on either domain alone.
Conclusions:
Preliminary findings suggest that integrating peripheral endocrine and immune-inflammatory information may enhance early differentiation between bipolar and unipolar depression. These results highlight the potential value of multimodal, low-cost biomarkers derived from routine blood tests in supporting clinical decision-making for mood disorders. Further validation in prospective and multi-center studies is warranted.