This retrospective study used data from the stroke neuropsychology patient database of the First Affiliated Hospital of Soochow University, which contains information on AIS patients who received care in the stroke unit. The study was designed according to the tenets of the Declaration of Helsinki and was approved by the Institutional Review Board (IRB no. 2023–397). Results are reported according to Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines (21). All participants included in the study provided written, informed consent.

A total of 747 consecutive patients with AIS (within 3 days of symptom onset) admitted to the First Hospital of Soochow University between March 2018 and February 2023 were screened. Inclusion criteria were as follows: aged 18–85 years, admitted within 3 days after onset of AIS (primary or recurrent stroke), and willing and able to complete the psychological evaluation. Exclusion criteria were severe aphasia, apoplexy, or unconsciousness hindering the completion of depression tests (n = 38); history of depression, anxiety, and other psychiatric impairment (n = 14); history of tumor or severe systemic disease (n = 22); and diagnosis of transient ischemic attack (n = 18). Another 22 patients were excluded for missing laboratory data (e.g., monocyte, lymphocyte, and neutrophil counts). Ultimately, 633 patients were enrolled in the study. After propensity score matching (PSM) between patients who received IVT and those who did not on select baseline characteristics, there were 116 patients in each group (Figure 1).

The primary independent variable was receipt of IVT within 4.5 h of stroke onset. The standard dose was 0.9 mg/kg and total dose was <90 mg. The first dose was 10% of the total dose, which was administered intravenously over 1 min; the remaining 90% of the dose was continuously delivered intravenously over 1 h using a microinjection pump according to 2021 European guidelines (16).

The primary outcome of the study was a diagnosis of depression in the acute stage after ischemic stroke. Early-onset PSD was evaluated with the 17-item Hamilton Depression Scale (HAMD17) 14–21 days after stroke onset (22) using the established threshold score of ≥7 (23).

Fasting blood samples were collected on the morning after admission and demographic characteristics and risk factors were recorded within 6 h after admission.

Clinical parameters known to influence the occurrence of PSD were assessed, including demographic characteristics such as age and sex as well as vascular risk factors such as history of hypertension, diabetes mellitus, smoking status, and heart disease (including atrial fibrillation and coronary artery disease). Stroke severity was evaluated using the National Institutes of Health Stroke Scale (NIHSS) (24) at admission. Stroke subtype was classified according to Trial of Org 10,172 in Acute Stroke Treatment (TOAST) as large artery atherosclerosis (LAA), small artery occlusion (SAO), cardioembolism (CE), and other types (other cause [SOD] and undetermined cause [SUE]) (25). Mechanical thrombectomy was also recorded. Laboratory tests included neutrophil-to-lymphocyte ratio (NLR) (26), monocyte-to-lymphocyte ratio (27), monocyte-to-high density lipoprotein ratio (MHD) (28), and platelet-to-lymphocyte ratio (PLR) (29). Estimated glomerular filtration rate (eGFR) was calculated using the Chronic Kidney Disease Epidemiology Collaboration formula (30) and was included as a covariate in the multivariate logistic regression model as in previous studies (31), with a cutoff value of 60 mL/min/1.73 m².

Descriptive analysis of the entire study population was performed. Categorical variables are presented as a number and percentage; normally distributed continuous variables are reported as mean and standard deviation, and data with a non-normal (skewed) distribution are presented as median and interquartile range. Differences between groups were evaluated with the chi-squared test or Fisher’s exact test for categorical variables and the Student’s t-test or Mann–Whitney U test for continuous variables.

To examine the association between IVT and early-onset PSD while minimizing potential bias and confounding effects, we used a logistic regression model with PSM, with a 1:1 nearest neighbor matching algorithm and caliper width of 0.2. The propensity score (PS) was generated based on the aforementioned clinical parameters that are known to influence PSD occurrence, and the degree of matching was assessed using standardized mean difference with a threshold <0.1 considered acceptable.

Outcomes of the 2 groups in the PS-matched cohort were compared using a 2-sided t-test, with estimated PSs used as weights. A weighted cohort was generated using a pairwise algorithm (PA) (32) and inverse probability of treatment weighting (IPTW) (33). The PA model Pairs individuals in the treatment and control groups based on PSs that estimate the probability of receiving the treatment, thereby creating matched confounding for unbiased comparisons (32). The IPTW model assigns weights to individuals based on their PSs, giving more weight to those with similar treatment propensities and adjusting for confounding (33); this method allowed a more accurate estimation of the effect of treatment on outcome. We calculated the E-value to determine the contribution of potential unmeasured confounding factors to the observed association between IVT and early-onset PSD (34). To further assess the stability of our results, subgroup and interaction analyses were conducted for age, sex, smoking, history of stroke, hypertension, diabetes mellitus, and heart disease.

Statistical analyses were performed using R version 4.3.1¹ and Free Statistics software version 1.8 (35). Comparisons were performed with a 2-tailed test with a significance threshold of p < 0.05.

The characteristics of the 633 study participants are presented in Table 1. The average age was 63.9 ± 11.6 years and 40.9% were female. Patients in the IVT group were younger and had lower rates of hypertension, diabetes mellitus, history of stroke, and heart disease than those in the non-IVT group. For TOAST stroke subtypes, the IVT group had a lower prevalence of LAA and SAO and higher prevalence of CE and SOD+SUE; additionally, a larger proportion received mechanical thrombectomy and eGFR and NLR were higher compared with the non-IVT group. There were no differences in sex ratio, smoking status, or NIHSS score at admission between groups.

Before PSM, the prevalence of PSD was 18.3% (22/120) in patients with IVT and 29.2% (150/513) in those without IVT. After PSM, the baseline characteristics of the 2 groups were almost balanced, and the prevalence of PSD was 19.0% (22/116) and 34.5% (40/116), respectively. Treatment with IVT was associated with significantly lower rates of early-onset PSD (p < 0.05; Figure 2).

IVT was associated with a significant reductions in the occurrence of early-onset PSD in both univariate and multivariate logistic regression analyses, after IPTW or using the PA model based on PS; odds ratios (ORs) ranged from 0.44 to 0.61 (all p < 0.05; Table 2). The E-value for the study population was 2.24; that is, unmeasured confounders would have to be more than 2 times more prevalent in the IVT group than in the non-IVT group to explain the observed association.

In subgroup analyses, there were no effects of interactions between age, sex, smoking status, hypertension, diabetes mellitus, history of stroke, or heart disease on the association between IVT and early-onset PSD using a fully adjusted model (Figure 3).

This study was not without limitations. First, patients with severe stroke experiencing disorders of consciousness or speech were excluded because their impairment prevented them from completing the HAMD. Consequently, the occurrence of PSD in patients with AIS may have been underestimated. Second, given that the study was hospital-based, it was difficult to assess depression scores prior to stroke onset and the potentially confounding effect of preexisting depression symptoms cannot be completely excluded. In future studies with larger cohorts, depression scores before and after stroke should be compared (46). Finally, large-scale multicenter studies are needed to confirm our findings and validate their applicability to clinical practice, and to elucidate the mechanisms by which thrombolysis mitigates early-onset PSD.