Changes in Serum GDF-15 in Relation to the Onset of Post-Stroke Depression After Acute Ischemic Stroke and the Development of a Comprehensive Prediction Model

Abstract

Background: This study aimed to investigate the relationship between serum growth differentiation factor-15 (GDF-15) levels and the development of post-stroke depression (PSD) following acute ischemic stroke (AIS) and to construct an integrated predictive model.

Methods: This retrospective study enrolled 302 patients with AIS admitted to Siyang County Hospital of Traditional Chinese Medicine between January 2023 to October 2025. Among them, 100 patients developed PSD, and the remaining 202 patients did not. Serum GDF-15 levels at admission, general clinical data and disease-related information were compared between groups. Logistic regression identified independent risk factors for PSD, and a nomogram-based prediction model was developed. Model performance was assessed using receiver operating characteristic (ROC) analysis.

Results: Compared with the non-PSD group, the PSD group had older age, higher National Institutes of Health Stroke Scale (NIHSS) scores, and a greater proportion of patient with a family history of psychiatric disorders and frontal lobe infarction (p < 0.05). Serum level of high-sensitivity C-reactive protein (hs-CRP), homocysteine, uric acid, Interleukin-6 (IL-6), neurofilament light chain (NfL) and GDF-15 were elevated, whereas left ventricular ejection fraction was lower in the PSD group (p < 0.05). Logistic regression showed that higher NIHSS scores (odds ratio (OR) = 1.230, 95% confidence interval (CI): 1.045–1.447), family history of psychiatric disorders (OR = 7.760, 95% CI: 1.709–35.238), frontal lobe infarction (OR = 5.275, 95% CI: 1.574–17.678), and elevated level of IL-6 levels (OR = 1.319, 95% CI: 1.089–1.599), NfL (OR = 1.103, 95% CI: 1.018–1.195), and GDF-15 (OR = 1.021, 95% CI: 1.009–1.033) were independent risk factors for PSD in patients with AIS (p < 0.05). The nomogram achieved an area under the ROC curve of 0.863 (95% CI: 0.810–0.885) for predicting PSD in patients with AIS. 

Conclusions: Serum GDF-15 is closely associated with PSD onset after AIS. A comprehensive prediction model integrating GDF-15 with other clinical factors demonstrates high clinical utility for predicting PSD risk.