Higher serum lipocalin 2 is associated with post-stroke depression at discharge.

BACKGROUND AND AIMS: Post-stroke depression (PSD), as one of the common complications after stroke, seriously affects the physical and mental health and functional prognosis of patients. Previous studies have shown that the increase of inflammatory mediators is associated with the occurrence of PSD. Lipocalin 2 (LCN2), as an acute phase protein, is involved in the development of acute ischemic stroke (AIS), and its expression is up-regulated in patients with depression, suggesting that there is a potential correlation between serum LCN2 and depression. The aim of this study was to explore the relationship between serum LCN2 at admission and PSD at discharge.

METHODS: A total of 358 AIS patients were retrospectively included. All patients had fasting venous blood taken within 24 h of admission to detect serum LCN2. The patients were evaluated by 17-item Hamilton Depression Scale (HAMD) before discharge. Patients with HAMD score > 7 were diagnosed with PSD. The correlation between serum LCN2 and PSD was tested using binary logistic regression analysis.

RESULTS: In our study, 92 (25.7%) patients were diagnosed with PSD at discharge. According to the serum LCN2 value, the patients were divided into three layers (Tertile1 ≤ 105.24ng/ml; Tertile2: 105.24-140.12ng/ml; Tertile3 ≥ 140.12ng/ml), with T1 layer (the lowest levels) as a reference, after adjusting for multiple potential confounding factors, T3 layer (the highest levels) was independently associated with the occurrence of PSD (odds ratio [OR] = 2.639, 95% confidence interval [CI]: 1.317-5.287, P = 0.006). Similar results were found when the serum LCN2 was analyzed as a continuous variable. The optimal cut-off value of serum LCN2 at admission to predict PSD at discharge was 117.60ng/ml, at this threshold, the sensitivity was 77.2%, and the specificity was 53.4%.

CONCLUSIONS: High serum LCN2 levels at admission are an independent risk factor for PSD in patients with AIS at discharge.