Efficacy and influencing factors of 3-N-Butylphthalide-trazodone combination in managing post-ischemic stroke depression

Abstract

BACKGROUND

Given the current insufficiency of effective prevention and treatment methods for post-ischemic stroke depression (PISD), the exploration of improved management strategies is essential.

AIM

To analyze the efficacy and influencing factors of the 3-N-Butylphthalide (NBP)-Trazodone (TRZ) combination in the management of PISD.

METHODS

A total of 122 patients with PISD (March 2023-March 2025) were enrolled. According to the treatment regimens they actually received, 56 cases receiving TRZ alone were assigned to the reference group, while 66 cases administered NBP + TRZ were allocated to the research group. The following parameters were comparatively analyzed: Therapeutic efficacy; safety (nausea, celialgia, headache, and drowsiness); National Institutes of Health Stroke Scale (NIHSS); Barthel Index; 17-item Hamilton Depression Rating Scale; and serum neurotransmitters [5-hydroxytryptamine (5-HT), dopamine (DA), and norepinephrine (NE)]. Subsequently, determinants of treatment efficacy were explored using uni- and multivariate analyses.

RESULTS

Compared with the reference group, the research group demonstrated: (1) A markedly higher overall treatment effectiveness rate; (2) A comparable total incidence of adverse reactions; (3) Significantly reduced NIHSS and HAMD-17 scores after intervention; and (4) Improved Barthel Index and higher 5-HT, DA, and NE levels after treatment. Comorbid hyperlipidemia, 5-HT, DA, and medication regimen were identified as independent determinants of treatment efficacy in patients with PISD.

CONCLUSION

The NBP-TRZ combination is effective in the management of PISD. Ineffective treatment is associated with clinical features such as hyperlipidemia, 5-HT < 60 ng/mL, DA < 122 ng/mL, and TRZ monotherapy.

Key Words: 3-N-Butylphthalide; Trazodone; Post-ischemic stroke depression; Therapeutic effectiveness; Influencing factors

Core Tip: For the optimization of treatment for post-ischemic stroke depression, this study verified the efficacy of the 3-N-Butylphthalide (NBP)-Trazodone (TRZ) combination and explored efficacy-associated determinants. Without increasing the total incidence of adverse reactions, the NBP-TRZ combination was significantly superior to TRZ monotherapy in therapeutic efficacy, while also significantly improving patients’ neurological function and daily living ability, alleviating depression, and regulating serum neurotransmitters. Comorbid hyperlipidemia, low 5-hydroxytryptamine, reduced dopamine, and TRZ monotherapy were associated with treatment ineffectiveness in such patients.

INTRODUCTION

Stroke is a significant cause of disability and mortality worldwide, with approximately 65% of cases classified as ischemic stroke (IS)[1]. Pathologically, IS is characterized by a sudden and pronounced reduction in regional cerebral perfusion, inducing major cerebral artery occlusion and neuronal death[2]. As a common emotional disorder in patients with IS, post-IS depression (PISD) is usually manifested by pessimism, irritability, apathy, anhedonia, mood fluctuations, and suicidal ideation[3]. IS-induced white matter damage may contribute to this condition by impairing mood regulation-related neural circuits or signaling pathways such as the VEGF/MAPK/extracellular signal-regulated kinase axis[4]. The risk of PISD is associated with factors, including age, Barthel Index, National Institutes of Health Stroke Scale (NIHSS) score, homocysteine level, and red blood cell distribution width[5]. Approximately 20%-65% of stroke patients develop post-stroke depression, which interferes with rehabilitation and increases stroke the risks of stroke recurrence and mortality[6]. Given the lack of highly effective prevention and treatment strategies for PISD, the exploration of improved therapeutic approaches is urgently required[7]. Trazodone (TRZ) is a triazolopyridine serotonin receptor antagonist and reuptake inhibitor antidepressant. In the treatment of adult major depressive disorder, it is commonly used because it effectively alleviates depressive symptoms such as anxiety, insomnia, irritability, and psychomotor agitation; however, it may also cause adverse effects such as drowsiness and headaches[8]. This medicine has also been shown to improve daily living ability in patients with post-stroke depression[9]. 3-N-Butylphthalide (NBP), a compound derived from celery seed, can reduce neuronal damage-induced ferroptosis and ameliorate brain injury by inhibiting the aryl hydrocarbon receptor-cytochrome P450 family 1 subfamily B member 1 signaling axis[10]. Its anti-inflammatory, antioxidative, anti-apoptotic, and microcirculation-protective properties enable it to significantly improve functional outcomes in patients with acute IS[11]. Additionally, a mouse study demonstrated that NBP effectively mitigated depressive behaviors following traumatic brain injury, suggesting potential therapeutic value in PISD management[12]. However, the efficacy and influencing factors of the NBP-TRZ combination in PISD management remain insufficiently reported in existing literature. Therefore, this study conducted a detailed investigation and analysis of this therapeutic strategy.

MATERIALS AND METHODS

Patient selection

Eligibility requirements: All patients met the diagnostic criteria for PISD[13], experienced their first episode, and were in the recovery phase. They had no allergic reactions to the study medications, had a 17-item Hamilton Depression Rating Scale (HAMD17) score ≥ 7[14], and had complete case date. Their cognitive and communication abilities were normal. Exclusion criteria: Patients with post-stroke coma, dysarthria, or other conditions were excluded. Individuals with blood diseases, acute or chronic infections, malignancies, immune dysfunction, vascular dementia, dysfunctional vital organs, severe heart disease, or arrhythmia were also excluded. Patients with epilepsy or mania were considered ineligible. Patients who had previously engaged in self-harm or suicidal behavior, or showed such tendencies, were not included. Individuals with a history of mental illness or those who were pregnant or lactating were also excluded.

A total of 122 patients with PISD treated in our hospital (March 2023-March 2025) were strictly selected according to the inclusion criteria. Among them, 56 cases receiving TRZ alone constituted the reference group, whereas 66 cases receiving NBP + TRZ constituted the research group.

Intervention method

Patients in the reference group received TRZ with an initial dose of 50 mg twice daily. Depending on the patient’s condition, the dosage was increased as appropriate after two weeks, without exceeding the maximum dose of 250 mg/day. Patients in the research group received NBP capsules (specification: 0.2 g), administered orally three times daily. Both groups completed a 14-day treatment course.

TRZ dose adjustment followed the principle that the attending physician conducted a comprehensive evaluation based on the degree of improvement in depressive symptoms (HAMD-17 score reduction rate) and drug tolerance two weeks post-administration. If the patient showed no obvious adverse reactions but showed no significant improvement in depressive symptoms (< 30% reduction in the HAMD-17 score), the dosage was increased to 75 mg/day during the following week. Thereafter, the dosing was gradually increased at a rate of 50 mg/week according to the patient’s condition and treatment response, with the maximum daily dose not exceeding 250 mg. If intolerable adverse reactions (e.g., severe nausea, abdominal distension, headache, and drowsiness) occurred during treatment, the dosage was reduced or restored to the previously tolerated dosage as appropriate. Both groups followed the same TRZ dose-adjustment principle.

Data collection and outcome measures

Efficacy: The therapeutic efficacy was classified as significantly effective (essential resolution of clinical symptoms plus > 80% reduction in the anxiety and depression scores), effective (certain resolution of clinical symptoms plus a 30%-79% reduction in the anxiety and depression scores), or ineffective (no change or worsening of clinical symptoms)[15]. Total effectiveness rate = (significantly effective cases + effective cases)/total cases.

Safety: The incidence and proportion of adverse events, including nausea, celialgia, headaches, and drowsiness, were recorded.

Neurological function: Neurological status was assessed using the NIHSS[16], which covers 13 items with a total score of 42 points (1-4: Mild impairment; 5-20: Moderate impairment; > 20: Severe impairment).

Self-care ability: Patient’s ability to perform daily self-care was evaluated using the Barthel Index (0-100 points)[17], in which higher scores indicate stronger self-care ability.

Depression severity: The HAMD-17 was used to assess depressive symptoms, with scores of 0-54. Score interpretation: < 7 = normal; 7-17 = mild; 18-24 = moderate; > 24 = severe.

Serum neurotransmitters: Fasting venous blood samples (3 mL) were collected pre- and post-treatment. Enzyme-linked immunosorbent assay (ELISA) was used to quantify serum 5-hydroxytryptamine (5-HT), dopamine (DA), and norepinephrine (NE) levels.

Statistical analysis

Data were processed and analyzed using SPSS 26.0. Measurement data (mean ± SD) were analyzed using the independent samples t-test for between-group comparisons and the paired t-test for within-group comparisons pre- and post-treatment. Counting data were expressed as n (%) and analyzed using the χ² test. Finally, efficacy-associated determinants were identified using uni- and multivariate (Logistic regression) analyses. Statistical significance was defined as P < 0.05.

RESULTS

General data

Baseline clinical comparability between the two groups was first confirmed via balanced distributions in age, sex (male), body mass index (BMI), stroke course, infarct site, hypertension, diabetes, and hyperlipidemia (P > 0.05; Table 1).

Table 1 General patient data, n (%).

Indicators	Reference group (n = 56)	Research group (n = 66)	χ2/t	P value
Age (years)	59.96 ± 6.89	57.67 ± 8.68
Male	35 (62.50)	40 (60.61)	0.046	0.830
Body mass index (kg/m2)	23.34 ± 2.27	22.74 ± 2.36	1.424	0.157
Stroke course (day)	20.36 ± 5.69	22.36 ± 6.27	1.831	0.070
Infarction site			0.529	0.767
Cortical and subcortical	28 (50.00)	35 (53.03)
Brainstem	15 (26.79)	14 (21.21)
Cerebellum	13 (23.21)	17 (25.76)
Hypertension	15 (26.79)	20 (30.30)	0.183	0.669
Diabetes	23 (41.07)	19 (28.79)	2.025	0.155
Hyperlipidaemia	11 (19.64)	15 (22.73)	0.172	0.679

Therapeutic effectiveness

A significant inter-group difference in the total effective rate was observed (P < 0.05), favoring the research group (89.39% vs 75.00%, P < 0.05; Table 2).

Table 2 Curative effect analysis, n (%).

Indicators	Reference group (n = 56)	Research group (n = 66)	χ2	P value
Significantly effective	25 (44.64)	39 (59.09)
Effective	17 (30.36)	20 (30.30)
Ineffective	14 (25.00)	7 (10.61)
Overall effectiveness	42 (75.00)	59 (89.39)	4.405	0.036

Safety

Safety evaluation primarily included the incidence of nausea, celialgia, headache, and drowsiness. As shown in Table 3, no significant difference was observed in the total incidence of adverse reactions between the groups (reference group: 8.93%; research group: 4.55%; P > 0.05).

Table 3 Safety analysis, n (%).

Indicators	Reference group (n = 56)	Research group (n = 66)	χ2	P value
Nausea	0 (0.00)	2 (3.03)
Celialgia	1 (1.79)	1 (1.52)
Headache	1 (1.79)	0 (0.00)
Drowsiness	3 (5.36)	0 (0.00)
Total	5 (8.93)	3 (4.55)	0.950	0.330

NIHSS, Barthel Index, and HAMD17 scores

Neurological deficits, daily living ability, and depression severity were assessed using the NIHSS, Barthel Index, and HAMD-17 tools, respectively (Figure 1). No significant differences were observed between the groups in baseline scores (P > 0.05). Post-treatment, the NIHSS and HAMD17 scores decreased, whereas the Barthel Index score increased in both groups (P < 0.05), with greater improvements observed in the research group than with the reference group (P < 0.05).

Figure 1 Analysis of National Institutes of Health Stroke Scale, Barthel Index, and 17-item Hamilton Depression Rating Scale scores. A: Pre- and post-treatment National Institutes of Health Stroke Scale scores across groups; B: Barthel Index scores pre- and post-treatment; C: Pre- and post-treatment 17-item Hamilton Depression Rating Scale in two groups. ^aP < 0.05, ^bP < 0.01 within-group comparison vs pre-treatment, ^cP < 0.05 vs the reference group. NIHSS: National Institutes of Health Stroke Scale; HAMD17: 17-item Hamilton Depression Rating Scale.

Serum neurotransmitters

Figure 2 visualizes the changes in serum neurotransmitters (NTs) levels (5-HT, DA, and NE) measured using ELISA. No significant differences were observed between the groups at baseline (P > 0.05). Post-treatment, serum levels of 5-HT, DA, and NE increased in both groups (P < 0.05), with greater increases observed in the research group (P < 0.05).

Figure 2 Comparative analysis of serum neurotransmitters. A: 5-hydroxytryptamine alterations (pre- vs post-treatment) in two groups; B: Dopamine changes pre- and post-treatment; C: Pre- and post-treatment norepinephrine levels. ^aP < 0.05 and ^bP < 0.01 vs pre-treatment value, ^cP < 0.05 vs the reference group. 5-HT: 5-hydroxytryptamine; DA: Dopamine; NE: Norepinephrine.

Analysis of factors influencing patients’ therapeutic effects

Univariate analysis (Table 4) showed no significant differences between the ineffective and effective groups in age, sex (male), BMI, stroke course, infarction site, hypertension, diabetes, NIHSS, Barthel Index, HAMD-17, or NE (P > 0.05). However, statistically significant differences were observed in hyperlipidemia, 5-HT, DA, and medication regimen.

Table 4 Determinants of patients' curative effects by univariate analysis, n (%).

Indicators	Ineffective group (n = 21)	Effective group (n = 101)	χ2	P value
Age (years)			1.107	0.293
< 60	9 (42.86)	56 (55.45)
≥ 60	12 (57.14)	45 (44.55)
Male	12 (57.14)	63 (62.38)	0.201	0.654
Body mass index (kg/m2)			0.164	0.685
< 23	11 (52.38)	48 (47.52)
≥ 23	10 (47.62)	53 (52.48)
Stroke course (day)			1.055	0.304
< 20	7 (33.33)	46 (45.54)
≥ 20	14 (66.67)	55 (54.46)
Infarction site			0.332	0.847
Cortical and subcortical	10 (47.62)	53 (52.48)
Brainstem	6 (28.57)	23 (22.77)
Cerebellum	5 (23.81)	25 (24.75)
Hypertension	7 (33.33)	28 (27.72)	0.268	0.605
Diabetes	10 (47.62)	32 (31.68)	1.956	0.162
Hyperlipidaemia	8 (38.10)	18 (17.82)	4.261	0.039
NIHSS (points)			1.418	0.234
< 16	6 (28.57)	43 (42.57)
≥ 16	15 (71.43)	58 (57.43)
Barthel Index (points)			0.391	0.532
< 48	13 (61.90)	55 (54.46)
≥ 48	8 (38.10)	46 (45.54)
HAMD17 (points)			3.103	0.078
< 16	7 (33.33)	55 (54.46)
≥ 16	14 (66.67)	46 (45.54)
5-HT (ng/mL)			4.989	0.026
< 60	13 (61.90)	36 (35.64)
≥ 60	8 (38.10)	65 (64.36)
DA (ng/mL)			7.112	0.008
< 122	15 (71.43)	40 (39.60)
≥ 122	6 (28.57)	61 (60.40)
NE (ng/mL)			1.706	0.192
< 20	12 (57.14)	42 (41.58)
≥ 20	9 (42.86)	59 (58.42)
Medication regimen			4.405	0.036
3-N-Butylphthalide + Trazodonee	7 (33.33)	59 (58.42)
Trazodone	14 (66.67)	42 (41.58)

NIHSS: National Institutes of Health Stroke Scale; HAMD17: 17-item Hamilton Depression Rating Scale; 5-HT: 5-hydroxytryptamine; DA: Dopamine; NE: Norepinephrine.

Further multivariate analysis (Table 5) confirmed hyperlipidemia, 5-HT, DA, and medication regimen as independent determinants of treatment efficacy (P < 0.05). Hyperlipidemia and TRZ monotherapy increased the risk of treatment failure, whereas higher levels of 5-HT (≥ 60 ng/mL) and DA (≥ 122 ng/mL) served as protective factors.

Table 5 Predictors of patients’ curative effects by multivariate modeling.

Variable	β	SE	Wald	P value	OR	95%CI
Hyperlipidaemia	1.394	0.602	5.369	0.020	4.030	1.240-13.102
5-HT (ng/mL)	-1.117	0.556	4.039	0.044	0.327	0.110-0.973
DA (ng/mL)	-1.592	0.589	7.315	0.007	0.203	0.064-0.645
Medication regimen	1.331	0.571	5.444	0.020	3.786	1.237-11.583

5-HT: 5-hydroxytryptamine; DA: Dopamine.

DISCUSSION

PISD typically occurs more than two weeks after a stroke, with patients usually presenting at least four additional depressive symptoms simultaneously[18]. To effectively alleviate clinical symptoms, facilitate rehabilitation, and improve outcomes in patients with PISD, this study comparatively analyzed the efficacy of TRZ monotherapy and NBP + TRZ combination therapy. The results confirmed that the NBP + TRZ combination provided higher therapeutic efficacy than TRZ monotherapy. TRZ exerts antidepressant effects mainly by antagonizing the 5-HT 2A and 5-HT 2C receptors, which partially explains its therapeutic mechanism in patients with PISD[19]. The mechanism by which NBP mitigates post-stroke depression may involve the inhibition of the toll-like receptor 4/nuclear factor κB signaling pathway, as well as effective downregulation of PANoptosis-related molecules (caspase-1/3/8, etc.) in the hippocampus, thereby inhibiting neuroinflammation and neuronal cell death[20]. Because the two drugs exert antidepressant effects through different pathways, their combined use produce synergistic therapeutic effects in patients with PISD.

Although the combination regimen included an additional medication (NBP), its safety profile was comparable to that of TRZ monotherapy, suggesting acceptable tolerability. Notably, three cases of drowsiness occurred in the TRZ monotherapy group, whereas none were observed in the combination therapy group. This may relate to the sedative effect of TRZ itself[21]. The co-administration NBP-TRZ may accelerate neurological recovery and indirectly improve overall alertness, thereby reducing the occurrence of drowsiness to some extent[22]. Nevertheless, validation in larger samples is required, particularly to further evaluate the effect of combination therapy on sedation. Scale-based evaluations further demonstrated that the NBP-TRZ combination significantly reduced neurological deficits, improving daily living ability, and alleviated depressive symptoms in patients with PISD. This effect may be because NBP can inhibit ferroptosis by activating the Nrf2/HO-1 signaling pathway, thereby reducing cerebral infarction volume, inhibiting lipid peroxidation and iron accumulation, and alleviating neurological dysfunction[23]. The neuroprotective effect of NBP in stroke patients has also been associated with its synergistic inhibition of inflammation and oxidative stress[24]. ELISA results further revealed that NBP + TRZ more effectively improved serum NT levels (5-HT, DA, and NE). This finding is consistent with the depletion of serum NT levels reported by Wu et al[25] in a mouse model of post-stroke depression. Reduced expression of 5-HT has been closely associated with the aggravation of post-stroke depression and has also been implicated in the therapeutic mechanisms of electroacupuncture therapy[26,27]. Moreover, higher levels of 5-HT, DA, and NE can reflect enhanced monoaminergic neuromodulation in the hippocampus[28]. Basic studies have also demonstrated that NBP protects dopaminergic neurons by inhibiting microglia-mediated neuroinflammation and promoting microglia transformation into the anti-inflammatory M2 phenotype in Parkinson’s disease models[29]. Although this study did not include animal experiments, these findings support the hypothesis that NBP may reduce post-stroke damage of monoaminergic neurons (e.g., raphe nucleus and locus coeruleus) through anti-inflammatory mechanisms, thereby promoting a favorable neural microenvironment for the synthesis and release of 5-HT, DA, and NE and enhancing the effects of TRZ. Finally, univariate and multivariate analyses confirmed hyperlipidemia and TRZ monotherapy as independent risk factors for treatment failure in PISD, whereas higher levels of 5-HT (≥ 60 ng/mL) and DA (≥ 122 ng/mL) served as independent protective factors.

From a clinical translational perspective, the identification of low 5-HT, reduced DA, hyperlipidemia, and monotherapy as risk factors for poor therapeutic efficacy may provide guidance for individualized treatment approaches. First, serum NT level quantification is a candidate auxiliary means for early identification of high-risk patients. For patients with low baseline NT levels, a more proactive combined treatment strategy may be considered. Second, in patients with hyperlipidemia, intensive lipid-lowering therapy may help to improve the efficacy of antidepressants. However, given the lack of relevant clinical applications, the efficacy and mechanisms of such strategies require further investigation.

This study also has several limitations. First, potential selection bias may exist because of the relatively limited sample size and the single-center design. Multi-centered, large-scale randomized controlled trials are needed to confirm result generalizability. Second, the final average daily dose of TRZ was not recorded in either group. Relevant supplementation in follow-up research can help avoid the interference of this confounding factor in the conclusions. Finally, the clinical application and mechanisms of intensive lipid-lowering therapy in the treatment of PISD were not explored and should be investigated in future studies.

CONCLUSION

In summary, NBP + TRZ demonstrates clear advantages over TRZ monotherapy in the treatment of PISD. It improves therapeutic efficacy, neurological function, daily living ability, and depressive symptoms without increasing the overall adverse reactions, while also effectively regulating serum NTs. Comorbid hyperlipidemia, 5-HT (< 60 ng/mL), DA (< 122 ng/mL), and TRZ monotherapy were associated with an increased risk of ineffective treatment in patients with PISD.