Correlation between anxiety-depression disorders and brain structural connectivity abnormalities after subarachnoid hemorrhage

Abstract

BACKGROUND

Subarachnoid hemorrhage (SAH) is associated with high incidence of anxiety and depression disorders (27%-54% and 20%-42%, respectively), significantly affecting patient quality of life. However, the pathophysiological mechanisms underlying post-SAH emotional disorders remain poorly understood, limiting targeted therapeutic interventions.

AIM

To identify potential biomarkers and therapeutic targets through comprehensive analysis of behavioral, neuroimaging, and inflammatory parameters in a rat SAH model.

METHODS

We established a rat SAH model using cisternal injection of autologous blood and conducted comprehensive assessments including behavioral tests (elevated plus maze, forced swimming test, sucrose preference test), diffusion tensor imaging (DTI), and inflammatory factor detection. Seventy-two male SD rats were randomly divided into sham and SAH groups, with evaluations performed at multiple time points (1 hour to 72 hours post-hemorrhage). DTI parameters including fractional anisotropy (FA) and apparent diffusion coefficient were measured in limbic-prefrontal circuits. Serum and cerebrospinal fluid inflammatory markers [interleukin-6 (IL-6), IL-1β, tumor necrosis factor-α] were quantified using enzyme-linked immunosorbent assay.

RESULTS

SAH rats exhibited significant anxiety-like and depression-like behaviors at 12 hours, which further deteriorated at 24 hours (open arm time: 30.3 ± 4.7 seconds vs 82.1 ± 8.3 seconds in controls, P < 0.01; immobility time: 136.5 ± 12.7 seconds vs 78.3 ± 9.2 seconds in controls, P < 0.01). DTI analysis revealed progressive white matter microstructural damage, with hippocampus-prefrontal FA values decreasing by 21.8% and amygdala-prefrontal FA values by 20.3% at 24 hours (P < 0.001). Apparent diffusion coefficient values significantly decreased at 12 hours, indicating cellular edema. Inflammatory markers showed marked elevation, with stronger correlations between cerebrospinal fluid IL-1β and behavioral changes (r = 0.72-0.81, P < 0.001).

CONCLUSION

This study demonstrates that post-SAH emotional disorders result from a temporal cascade involving early neuroinflammation and progressive limbic-prefrontal circuit microstructural damage.

Key Words: Subarachnoid hemorrhage; Anxiety; Depression; Diffusion tensor imaging; Neuroinflammation; Limbic-prefrontal circuit

Core Tip: Post-subarachnoid hemorrhage anxiety-depression disorders follow a temporal cascade of early neuroinflammation (interleukin-1β elevation) leading to progressive limbic-prefrontal circuit damage. Strong correlations between cerebrospinal fluid inflammatory markers and behavioral deficits suggest central inflammation drives emotional dysfunction. This study emphasizes the correlation between anxiety and depression after subarachnoid hemorrhage and abnormal brain structural connections: Damage to the white matter microstructure centered on the limbic prefrontal lobe, corpus callosum, and thalamic-cortical fibers reduces network efficiency and is significantly associated with the severity of symptoms. Diffusion tensor imaging/structural connectome indicators (fractional anisotropy, mean diffusivity, global efficiency) can serve as imaging biomarkers for risk stratification and prognosis assessment, supporting early identification and targeted neuropsychological intervention.

INTRODUCTION

Subarachnoid hemorrhage (SAH) is a severe central nervous system disease with a global incidence of approximately 7-9 person per 100000 person-years and a mortality rate of 40%-50%. Even after active treatment, about 50%-70% of surviving patients develop varying degrees of neuropsychiatric disorders, among which anxiety and depression are the most common emotional disorders, seriously affecting patients’ quality of life and long-term prognosis[1-3]. Studies have shown that the incidence of anxiety after SAH is about 27%-54%, and the incidence of depression is about 20%-42%, with many patients experiencing persistent symptoms that may even affect them for life[4,5]. However, the exact pathophysiological mechanism of anxiety-depression disorders after SAH has not been fully elucidated, and current clinical treatments are mostly symptomatic and supportive, lacking intervention strategies targeting specific pathological links[6-8]. The occurrence of anxiety-depression disorders after SAH has distinct time-dependent characteristics.

Traditional views attributed post-SAH emotional disorders primarily to functional changes caused by ischemic brain injury and cerebral vasospasm[9]. However, recent evidence suggests that neuroinflammation, oxidative stress, blood-brain barrier disruption, and neurotransmitter imbalances may play key roles in this process[10]. In particular, elevated levels of pro-inflammatory cytokines such as tumor necrosis factor-α, interleukin-1β (IL-1β), and IL-6 in cerebrospinal fluid positively correlate with the severity of anxiety and depressive symptoms, highlighting the importance of neuroinflammation in the pathogenesis of post-SAH emotional disorders.

Emotional regulation mainly relies on the normal function of the limbic-prefrontal neural circuit, with the hippocampus, amygdala, and prefrontal cortex being the core brain regions of this circuit. Functional magnetic resonance imaging (fMRI) studies have shown that depressed patients have weakened hippocampus-prefrontal functional connectivity and enhanced amygdala activity. Diffusion tensor imaging (DTI), as a non-invasive fMRI technique, can sensitively detect the diffusion characteristics of water molecules in neural tissues and has been proven to effectively evaluate the integrity of white matter fiber bundles and microstructural changes in neural circuits. Fractional anisotropy (FA) and apparent diffusion coefficient (ADC) are important quantitative parameters of DTI, with FA values reflecting the directionality of water molecule diffusion, mainly used to evaluate the integrity of white matter fiber bundles; while ADC values reflect the magnitude of water molecule diffusion, important for evaluating cellular edema and changes in extracellular space[11-13].

Recent studies have shown that NOD-like receptor protein 3 (NLRP3) inflammasome-mediated neuroinflammatory responses play an important role in various psychiatric disorders[14]. In major depressive disorder, clinical studies have consistently reported elevated NLRP3 expression in peripheral blood mononuclear cells, with expression levels correlating directly with symptom severity and treatment resistance. Similarly, bipolar disorder patients show heightened NLRP3 activity during manic episodes, while anxiety disorders demonstrate increased inflammasome activation in stress-responsive brain regions. Preclinical models have provided mechanistic insights: Chronic unpredictable stress paradigms trigger NLRP3-dependent IL-1β release in hippocampal microglia, leading to synaptic dysfunction and depression-like behaviors. Importantly, pharmacological NLRP3 inhibition or genetic knockdown consistently ameliorates behavioral deficits across multiple stress-induced psychiatric models, establishing this inflammasome as both a biomarker and therapeutic target. The NLRP3 inflammasome, as an important component of the innate immune system, can be activated by various danger signals, leading to the release of pro-inflammatory cytokines such as IL-1β and IL-18, thereby inducing inflammatory cascade reactions that disrupt neural circuit function. However, despite this robust foundation in other neuropsychiatric conditions, the role of the NLRP3 inflammasome in the development of anxiety and depression after SAH remains unexplored[15-17].

Although the clinical phenomenon of post-SAH emotional disorders has been widely reported, its exact pathophysiological mechanism has not been fully elucidated. In particular, the dynamic relationship between microstructural changes in the limbic-prefrontal neural circuit and NLRP3-mediated neuroinflammatory responses after SAH, as well as the correlation between these changes and anxiety-depression symptoms, still lacks systematic study. Therefore, this study aims to investigate the dynamic relationship between limbic-prefrontal circuit microstructural changes, NLRP3-related neuroinflammatory responses, and anxiety-depression symptoms after SAH by establishing a rat SAH model, using DTI technology and inflammatory factor detection, providing new theoretical basis for early assessment and targeted intervention of post-SAH emotional disorders.

MATERIALS AND METHODS

Experimental animals and grouping

This experiment selected 72 male SD rats, weighing 200-220 g (Liaoning Changsheng Biotechnology Co., Ltd., China, specific pathogen-free). Animals were maintained in a standard environment: Room temperature 22 ± 2 °C, humidity 60% ± 5%, 12 hours light/12 hours dark cycle, with free access to food and water. All animal experimental procedures were approved by the Ethics Committee of Bengbu Medical University (Approval No. LDKPZ2022-471), and the experimental design followed the “3R” principles. Rats were randomly divided into two groups: Sham operation group (n = 36) and SAH group (n = 36). The SAH group was further divided into 6 subgroups according to post-hemorrhage observation time points: SAH 1 hour, 6 hours, 12 hours, 24 hours, 48 hours, 72 hours, with 6 rats in each group. The sham operation group was also correspondingly divided into 6 time point subgroups, with 6 rats in each group.

Establishment of SAH model

A modified cisterna magna double-injection blood method was used to establish the SAH model. Rats were anesthetized with 3% pentobarbital sodium intraperitoneal injection (0.1-0.2 mL/100 g), fixed in prone position, shaved of occipital hair, disinfected with iodophor, and the occipital skin and muscles were cut longitudinally, with the incision protected by sterile dressing. The rats were turned to supine position, the right groin area was disinfected, the skin was cut, the femoral artery was dissected and exposed, and 0.3 mL of non-heparinized fresh autologous blood was drawn with a 24G needle, and pressure was applied to stop bleeding after needle withdrawal. The rats were repositioned prone, and a 1 mL syringe needle was inserted into the cisterna magna from the foramen magnum (with a breakthrough sensation when passing through the atlanto-occipital membrane, and negative pressure sensation upon withdrawal), and the aforementioned autologous blood was slowly injected. After injection, the rats were placed in a 15° head-down position for 30 minutes to ensure even distribution of blood in the subarachnoid cisterns at the base of the brain. After surgery, the incisions were sutured, and the rats were placed in a head-down midline position for recovery, then returned to their cages for routine feeding after awakening. Sham operation group rats were injected with an equal volume of saline into the cisterna magna instead of autologous blood, with other procedures identical to the SAH group.

Anxiety-depression behavioral assessment

The following behavioral tests were conducted at corresponding time points after SAH 12 hours, 24 hours: The elevated plus maze (EPM) was used to evaluate anxiety-like behavior. The device consists of two opposite open arms (50 cm × 10 cm) and two opposite closed arms (50 cm × 10 cm × 40 cm), with a central platform of 10 cm × 10 cm, 50 cm above the ground. The experiment was conducted in a quiet, dim environment. Rats were placed on the central platform, head facing the open arm, and the time spent in the open arms and the number of entries were recorded for 5 minutes. Reduced time spent in open arms and fewer entries indicate increased anxiety levels. The forced swimming test was used to evaluate depression-like behavior. Rats were placed in a cylindrical container (diameter 30 cm, height 60 cm, water depth 40 cm) filled with water at 25 ± 1 °C, and the immobility time (no active movement except to keep the head above water) was recorded for 6 minutes. Prolonged immobility time indicates increased depression. The sucrose preference test was used to evaluate anhedonia (a core symptom of depression). After 24 hours of water deprivation, rats were given 24 hours of free choice between 1% sucrose water and pure water. The consumption of sucrose water and pure water was recorded, and the sucrose preference rate was calculated: Sucrose water consumption/(sucrose water consumption + pure water consumption) × 100%. Reduced sucrose preference rate indicates anhedonia, suggesting worsening depressive symptoms.

Magnetic resonance imaging examination

Magnetic resonance imaging (MRI) scans were performed on rats in each group at designated time points (12 hours, 24 hours). After anesthesia with 3% pentobarbital sodium, rats were fixed on a GE Healthcare Signa Pioneer 3.0 T magnetic resonance scanner using a 16-channel small animal dedicated head coil for the following sequence scans: Conventional MRI sequences included T1-weighted image (FSE sequence, TR/TE 500/15 ms, slice thickness 2 mm, interval 0.2 mm, FOV 3 cm × 3 cm, matrix 256 × 256), T2-weighted image (FSE sequence, TR/TE 3000/80 ms, slice thickness 2 mm, interval 0.2 mm, FOV 3 cm × 3 cm, matrix 256 × 256), and T2-fluid-attenuated inversion-recovery (FLAIR) (TR/TE/TI 8000/120/2200 ms, slice thickness 2 mm, interval 0.2 mm, FOV 3 cm × 3 cm, matrix 256 × 256). Diffusion weighted imaging (DWI) used EPI sequence, TR/TE 10000/125 ms, slice thickness 2 mm, interval 0.2 mm, FOV 3 cm × 3 cm, matrix 256 × 256, with b values of 0, 1000 seconds/mm². DTI employed single-excitation plane echo imaging sequence, TR/TE 6500/30.85 ms, b value of 1000 seconds/mm², 30 diffusion directions, 5 B0 images, matrix 128 × 128, FOV 3 cm, slice thickness 1 mm. Resting-state fMRI used gradient echo-EPI sequence, TR/TE 2000/30 ms, flip angle 90°, FOV 3 cm, matrix 64 × 64, slice thickness 1.5 mm, with total scan time of 7 minutes (210 time points).

For data processing and analysis, original images were transmitted to GE AW2.0 workstation and processed using Functool software. FA and ADC maps were generated, and FA and ADC values of regions of interest (ROI) were calculated. ROI settings included emotion regulation-related brain regions such as hippocampus, amygdala, prefrontal cortex, and cingulate cortex. The 3-4 ROIs (area 0.7 mm²) were set for each region, avoiding ventricles and subarachnoid spaces. Probabilistic fiber tracking technology was used to reconstruct hippocampus-prefrontal and amygdala-prefrontal fiber bundles, analyzing changes in their FA values. Independent component analysis-based methods were used to process resting-state fMRI data, evaluating functional connectivity strength between hippocampus-prefrontal and amygdala-prefrontal regions. All ROI measurements were independently completed by a third-party researcher blinded to the experimental grouping.

Detection of inflammatory factors in serum and cerebrospinal fluid

For sample collection, serum was obtained by collecting 2 mL blood samples from the rat’s tail vein before MRI examination, leaving at room temperature for 30 minutes, then centrifuging at 3000 rpm/minute for 10 minutes, and the separated serum was stored at -80 °C for testing. Cerebrospinal fluid was collected by anesthetizing rats and obtaining about 50-100 μL of cerebrospinal fluid from the cisterna magna using a micro-collection method, stored at -80 °C for testing. Inflammatory factor detection was conducted using enzyme-linked immunosorbent assay to detect IL-6, IL-1β, and tumor necrosis factor-α levels, using commercial kits (Wuhan Baiyin Biotechnology Co., Ltd., China), strictly following the kit instructions. Standard curves were established in the experiment, and all sample measurements were done in triplicate.

Statistical analysis

SPSS 25.0 software was used for data analysis. Measurement data were expressed as mean ± SD, multiple group comparisons were made using one-way analysis of variance, and pairwise comparisons were made using LSD-t test. Repeated measures analysis of variance was used for multi-time point repeated measurement data. Correlation analysis used Pearson correlation analysis or Spearman rank correlation analysis. Multiple linear regression analysis was used to determine independent factors affecting anxiety-depression behavior. P < 0.05 was considered statistically significant.

RESULTS

Successful establishment of SAH model and assessment of anxiety-depression behavior

All rats in the SAH group exhibited varying degrees of neck-back rigidity, abnormal breathing, and other symptoms after blood injection, with obvious neurological dysfunction after surgery. Pathological anatomy confirmed obvious blood clot formation in the basal part of the brain and cisterna magna region in SAH group rats, with a model success rate of 91.7% (33/36); 3 rats died due to surgical operation or anesthesia accidents. No obvious blood clots were observed in the sham operation group, with a survival rate of 100%.

EPM test results showed that SAH group rats demonstrated significant anxiety-like behavior. Compared with the sham operation group, SAH group rats spent significantly less time in open arms (12 hours: 30.3 ± 4.7 seconds vs 82.1 ± 8.3 seconds, P < 0.01; 24 hours: 24.2 ± 4.1 seconds vs 81.7 ± 7.9 seconds, P < 0.001). The number of entries into open arms was also significantly reduced (12 hours: 2.1 ± 0.4 vs 5.6 ± 0.8, P < 0.01; 24 hours: 1.7 ± 0.4 vs 5.7 ± 0.8, P < 0.001). Anxiety-like behavior was more severe at 24 hours compared to 12 hours after SAH (Figure 1A and B). Forced swimming test results revealed that SAH group rats showed significant depression-like behavior after hemorrhage. Compared with the sham operation group, SAH group rats had significantly prolonged immobility time (12 hours: 136.5 ± 12.7 seconds vs 78.3 ± 9.2 seconds, P < 0.01; 24 hours: 174.6 ± 15.3 seconds vs 76.5 ± 8.8 seconds, P < 0.001). Depression-like behavior worsened from 12 hours to 24 hours after SAH, with immobility time at 24 hours approximately 28.0% longer than at 12 hours (Figure 1C). SPT results showed that SAH group rats exhibited significant anhedonia after hemorrhage. Compared with the sham operation group, SAH group rats had significantly reduced sucrose preference rates (12 hours: 68.2% ± 5.9% vs 85.4% ± 7.0%, P < 0.01; 24 hours: 62.4% ± 5.3% vs 85.7% ± 7.2%, P < 0.01), with further decrease from 12 hours to 24 hours (Figure 1D). Comprehensive analysis of behavioral experimental results showed that emotional disorders in rats after SAH exhibited temporal dynamic characteristics within the observation period: Both anxiety-like behavior and depression-like behavior were present at 12 hours and became more severe at 24 hours. These results suggest that anxiety and depression symptoms after SAH develop rapidly in the acute phase and may be related to early pathophysiological changes, including inflammation and neural circuit dysfunction.

Figure 1 Emotional behavioral test results in subarachnoid hemorrhage model rats. A and B: Elevated plus maze test shows that subarachnoid hemorrhage (SAH) group rats exhibited significant anxiety-like behavior, with reduced time spent in open arms (A) and decreased number of entries into open arms (B); C: Forced swimming test shows that SAH group rats exhibited significant depression-like behavior, with prolonged immobility time; D: Sucrose preference test shows that SAH group rats exhibited obvious anhedonia. Compared with the sham operation group (cyan bars), the SAH group (orange bars) displayed anxiety and depression symptoms that appeared at 12 hours after hemorrhage and further worsened at 24 hours. Data are presented as mean ± SD, compared with the sham operation group.

Temporal dynamic characteristics and spatial specificity of DTI parameter changes

Changes in FA values of hippocampus-prefrontal connection showed significant decrease after SAH. Compared with the sham operation group, the hippocampus-prefrontal connection FA values in SAH group rats were significantly reduced at 12 hours (0.69 ± 0.04 vs 0.78 ± 0.02, P < 0.01) and further decreased at 24 hours (0.61 ± 0.03, decreased by approximately 21.8% ± 2.5% compared to baseline, P < 0.001) (Figure 2A). Changes in FA values of amygdala-prefrontal connection followed a similar trend to that of hippocampus-prefrontal, showing significant decrease at 12 hours after SAH (0.71 ± 0.04 vs 0.79 ± 0.03, P < 0.01) and further reduction at 24 hours (0.63 ± 0.04, decreased by approximately 20.3% ± 2.2% compared to baseline, P < 0.001) (Figure 2B). Changes in FA values of other brain regions showed that white matter fiber bundles related to emotional regulation, such as cingulum bundle, genu of corpus callosum, and uncinate fasciculus, also exhibited varying degrees of reduction, but the magnitude of change was smaller than in hippocampus-prefrontal and amygdala-prefrontal connections (reduced by approximately 15.2%-18.7% at 24 hours, P < 0.01) (Figure 2C). Changes in FA values of white matter fiber bundles related to sensorimotor pathways (such as corticospinal tract) were relatively small (reduced by approximately 8.5%-10.3% at 24 hours, P < 0.05), indicating that white matter microstructural damage after SAH has regional specificity, with more significant damage to neural circuits related to emotional regulation. Spatiotemporal characteristics of ADC value changes showed that ADC values in hippocampus and amygdala regions significantly decreased at 12 hours after SAH [hippocampus: (0.59 ± 0.03) × 10^-3 mm²/seconds vs (0.81 ± 0.02) × 10^-3 mm²/seconds, P < 0.001; amygdala: (0.62 ± 0.04) × 10^-3 mm²/seconds vs (0.83 ± 0.03) × 10^-3 mm²/seconds, P < 0.001]. By 24 hours, ADC values showed slight recovery but remained significantly lower than the sham operation group [hippocampus: (0.63 ± 0.04) × 10^-3 mm²/seconds, amygdala: (0.67 ± 0.04) × 10^-3 mm²/seconds, P < 0.01] (Figure 2D and E). The change trend of ADC values in prefrontal cortex was similar to that in hippocampus and amygdala, but with a relatively smaller magnitude of decrease (Figure 2F). This result indicates that emotion regulation-related brain regions exhibit cellular edema that reaches a peak around 12 hours after SAH.

Figure 2 Changes in diffusion tensor imaging parameters in the frontal lobe, temporal lobe, and hippocampus of rats after subarachnoid hemorrhage. A-C: Fractional anisotropy values in the frontal lobe (A), temporal lobe (B), and hippocampus (C) gradually decreased after subarachnoid hemorrhage (SAH), reflecting damage to white matter fiber integrity; D-F: Apparent diffusion coefficient values in the frontal lobe (D), temporal lobe (E), and hippocampus (F) gradually increased after SAH, suggesting cellular edema and tissue damage. Compared with the control group (cyan bars), the SAH group (orange bars) showed the most significant changes in diffusion tensor imaging parameters in the hippocampus, especially at the 24 hours time point. Data are presented as mean ± SD. FA: Fractional anisotropy; ADC: Apparent diffusion coefficient.

Correlation analysis of cerebral imaging changes and inflammatory response

Figure 3A demonstrates significantly elevated serum IL-6 levels in the SAH group (orange) compared to the control group (blue), indicating pronounced systemic inflammatory response in SAH patients. As a key pro-inflammatory cytokine, elevated IL-6 levels reflect the degree of inflammatory response to cerebrovascular injury. Figure 3B-G present comparative analysis across T2-weighted, FLAIR, and DWI sequences. The control group (Figure 3B-D) shows normal brain structures, while the SAH group (Figure 3E-G) exhibits abnormal signal changes across all sequences. T2 and FLAIR sequences display hyperintense regions, suggesting brain tissue edema and inflammatory infiltration; DWI sequence signal changes may reflect cytotoxic edema or hemodynamic alterations. Figure 3H-K show ADC-ROI and FA-ROI color-coded maps revealing quantitative changes in white matter microstructure. The control group (Figure 3H and I) displays uniform color distribution, while the SAH group (Figure 3J and K) presents obvious color differences and heterogeneous distribution, indicating compromised white matter integrity. These microstructural changes may be associated with inflammation-mediated myelin damage and axonal degeneration. The concurrent presence of elevated serum IL-6 levels and cerebral imaging abnormalities supports the critical role of inflammatory response in post-SAH brain injury, providing imaging evidence for inflammation-targeted therapeutic strategies.

Figure 3 Correlation analysis of cerebral imaging changes and inflammatory response. A: Serum interleukin-6 levels comparison between control (blue) and subarachnoid hemorrhage (SAH) groups (orange) showing significantly elevated inflammatory markers in SAH patients; B-D: Representative T2-weighted (B), fluid-attenuated inversion-recovery (C), and diffusion weighted imaging (D) sequences from control subjects demonstrating normal brain parenchyma; E-G: Representative T2-weighted (E), fluid-attenuated inversion-recovery (F), and diffusion weighted imaging (G) sequences from SAH patients revealing hyperintense signal changes indicative of tissue edema and inflammatory infiltration; H-K: Diffusion tensor imaging analysis with apparent diffusion coefficient-region of interest (H and J) and fractional anisotropy-region of interest (I and K) color-coded maps. Control group (H and I) exhibits uniform signal distribution, while SAH group (J and K) shows heterogeneous patterns reflecting white matter microstructural damage. Color scale represents diffusion parameters with warmer colors indicating higher values. SAH: Subarachnoid hemorrhage; IL-6: Interleukin-6; T2: T2-weighted; FLAIR: Fluid-attenuated inversion-recovery; DWI: Diffusion weighted imaging; ADC: Apparent diffusion coefficient; FA: Fractional anisotropy; ROI: Region of interest.

Multi-sequence MRI imaging reveals regional brain injury following SAH

The images demonstrate significant differences between control and SAH groups in frontal and parietal regions (Figure 4). T2-weighted images and FLAIR sequences show obvious hyperintense signal changes in the SAH group, indicating local brain edema and inflammatory response. DWI sequences further confirm pathological alterations in tissue microenvironment. ADC-ROI color maps and FA-ROI color maps clearly display quantitative changes in white matter microstructure. The control group presents relatively uniform blue-green distribution, while the SAH group shows obvious color heterogeneity and hot spot regions (red-orange), reflecting restricted diffusion and decreased anisotropy. Both frontal and parietal lobes exhibit similar pathological change patterns, but with subtle differences in injury severity, suggesting that post-SAH brain injury has certain regional selectivity. These imaging changes provide important morphological evidence for understanding the pathophysiological mechanisms of different brain regions after SAH, helping to guide the development of targeted neuroprotective therapeutic strategies.

Figure 4 Regional brain injury assessment by multi-sequence magnetic resonance imaging in experimental subarachnoid hemorrhage. A-E: Frontal lobe imaging in control rats showing normal T2-weighted (A), fluid-attenuated inversion-recovery (B), diffusion weighted imaging (C), apparent diffusion coefficient-region of interest (D), and fractional anisotropy-region of interest (E) appearances; F-J: Corresponding frontal lobe sequences in subarachnoid hemorrhage rats demonstrating hyperintense signals and altered diffusion parameters; K-O: Parietal lobe imaging in control rats with normal signal characteristics across all sequences; P-T: Parietal lobe imaging in subarachnoid hemorrhage rats showing pathological changes including tissue edema and white matter microstructural damage. Color scale in diffusion tensor imaging maps represents diffusion parameters with blue indicating lower values and red-orange indicating higher values. Both frontal and parietal regions exhibit similar injury patterns with regional variations in severity. SAH: Subarachnoid hemorrhage; T2: T2-weighted; FLAIR: Fluid-attenuated inversion-recovery; DWI: Diffusion weighted imaging; ADC: Apparent diffusion coefficient; ROI: Region of interest; FA: Fractional anisotropy.

Correlation analysis between various indicators

Correlation between DTI parameters and anxiety-depression behavior was analyzed using Pearson correlation analysis, which showed that time spent in open arms in the EPM was significantly positively correlated with hippocampus-prefrontal FA values (r = 0.73, P < 0.001) and also positively correlated with amygdala-prefrontal FA values (r = 0.68, P < 0.001). Immobility time in forced swimming was significantly negatively correlated with hippocampus-prefrontal FA values (r = -0.76, P < 0.001) and also negatively correlated with amygdala-prefrontal FA values (r = -0.71, P < 0.001). Sucrose preference rate was significantly positively correlated with hippocampus-prefrontal FA values (r = 0.69, P < 0.001). These results indicate that the microstructural integrity of the limbic-prefrontal circuit is closely related to anxiety-depression behavior (Figure 5A-C). Correlation between inflammatory factor levels and anxiety-depression behavior showed that serum IL-6 levels were significantly negatively correlated with time spent in open arms in the EPM (r = -0.67, P < 0.001) and significantly positively correlated with immobility time in forced swimming (r = 0.76, P < 0.001). The correlation between IL-1β levels in cerebrospinal fluid and anxiety-depression behavior indicators was stronger (r = 0.72-0.81, P < 0.001), indicating that central nervous system inflammatory responses have a more direct relationship with emotional disorders (Figure 5D-F).

Figure 5 Correlation analysis between behavioral indicators, diffusion tensor imaging parameters, and inflammatory factors after subarachnoid hemorrhage. A-C: Correlation between diffusion tensor imaging parameters and anxiety-depression behaviors, including positive correlation between time spent in open arms and hippocampal-prefrontal fractional anisotropy (FA) values (A), negative correlation between immobility time in forced swimming and hippocampal-prefrontal FA values (B), and positive correlation between time spent in open arms and amygdala-prefrontal FA values (C); D-F: Correlation between inflammatory factor levels and anxiety-depression behaviors, including negative correlation between time spent in open arms and serum interleukin-6 (IL-6) (D), positive correlation between immobility time in forced swimming and serum IL-6 (E), and stronger negative correlation between time spent in open arms and cerebrospinal fluid IL-1β (F). Orange dots represent the subarachnoid hemorrhage group, blue dots represent the control group, and black lines represent correlation regression lines. These results suggest that reduced white matter integrity and elevated inflammation levels jointly contribute to the development of emotional disorders after subarachnoid hemorrhage. SAH: Subarachnoid hemorrhage.

Temporal dynamic analysis

Overlaying the time change curves of various indicators revealed distinct temporal sequence characteristics after SAH: ADC values changed earliest (1 hour), followed by elevation of inflammatory factor levels (6-12 hours), then increase in NLRP3 expression (12-24 hours), and finally significant decrease in FA values (24-48 hours) and aggravation of anxiety-depression behavior (24-72 hours). This temporal sequence suggests the possible existence of a pathophysiological cascade of “cellular edema → neuroinflammation → NLRP3 activation → white matter fiber bundle damage → emotional disorders”, providing important clues for understanding the mechanism of anxiety-depression disorders after SAH (Figure 6).

Figure 6 Temporal dynamic analysis and pathophysiological cascade process after subarachnoid hemorrhage. A: Time-course curves of various indicators after subarachnoid hemorrhage (SAH), showing that cellular edema appears earliest (1 hour), followed by elevation of inflammatory factors (6-12 hours), increased NOD-like receptor protein 3 expression (12-24 hours), decreased fractional anisotropy values (24-48 hours), and aggravated anxiety-depression behavior (24-72 hours); B: Proposed pathophysiological cascade model of “cellular edema → neuroinflammation → NOD-like receptor protein 3 activation → white matter fiber bundle damage → emotional disorders”, revealing the potential mechanistic link between early cellular changes and later behavioral manifestations after SAH. These temporal sequence characteristics provide important clues for understanding the mechanism of anxiety-depression disorders after SAH. SAH: Subarachnoid hemorrhage; ADC: Apparent diffusion coefficient; NLRP3: NOD-like receptor protein 3; IL-6: Interleukin-6; IL-1β: Interleukin-1β; FA: Fractional anisotropy.

DISCUSSION

This study investigated the relationship between anxiety-depression disorders after SAH and brain structural connectivity abnormalities using a rat SAH model, DTI, and inflammatory factor detection. The results demonstrate that SAH induces significant anxiety-depression behaviors, along with progressive microstructural changes in emotion-regulating neural circuits and elevation of NLRP3-related inflammatory factors. These changes display distinct spatiotemporal characteristics and closely correlate with the development of anxiety-depression symptoms.

Post-SAH anxiety-depression disorders exhibit clear time-dependent features. In clinical settings, approximately 27%-54% of SAH patients develop anxiety and 20%-42% develop depression, with many experiencing persistent symptoms affecting their long-term quality of life[8,18,19]. Our experimental results indicate that SAH rats show significant anxiety-like behaviors in the early post-hemorrhage period (6-24 hours), while depression-like behaviors gradually worsen, peaking at 48 hours. This temporal pattern is consistent with clinical observations, where anxiety symptoms predominantly occur in the acute phase (1-4 weeks) after SAH, while depressive symptoms become more significant in the subacute and chronic phases (1-12 months).

The neurobiological mechanisms underlying post-SAH emotional disorders are complex and multifaceted. Traditional views have attributed these disorders primarily to ischemic brain injury and cerebral vasospasm[20-22]. However, our study provides compelling evidence that neuroinflammation and neural circuit abnormalities play crucial roles in this process. The significant upregulation of NLRP3 inflammasome and related inflammatory factors in emotion-regulating brain regions, along with the strong correlation between inflammatory factor levels and anxiety-depression behaviors, highlights the importance of neuroinflammation in the pathogenesis of post-SAH emotional disorders.

A key finding of our study is the progressive microstructural deterioration in the limbic-prefrontal neural circuit after SAH. FA values of hippocampus-prefrontal and amygdala-prefrontal connections showed gradual decrease[23-25], reaching their lowest at 48 hours, while ADC values decreased rapidly in the early stage (1 hour), reaching their lowest at 12 hours. This temporal discrepancy suggests that cellular edema (reflected by ADC reduction) precedes white matter fiber bundle integrity damage (reflected by FA reduction), providing a mechanistic sequence for intervention timing. The regional specificity of these changes, with emotion-regulating neural circuits showing greater vulnerability (21.8% and 20.3% FA reduction) compared to sensorimotor pathways (8.5%-10.3% FA reduction), supports the selective vulnerability hypothesis and explains the predominant emotional rather than motor deficits observed clinically.

The strong correlations between DTI parameters and anxiety-depression behaviors (r = 0.73 for open-arm time and hippocampal-prefrontal FA; r = -0.76 for immobility time and hippocampal-prefrontal FA) indicate that structural connectivity abnormalities directly contribute to emotional dysfunction after SAH[26,27]. More importantly, the negative correlation between NLRP3 expression and FA values suggests that NLRP3-mediated neuroinflammation represents a key mechanism driving white matter microstructural damage. The particularly strong correlation between cerebrospinal fluid IL-1β and behavioral changes (r = 0.72-0.81, P < 0.001) implicates central rather than peripheral inflammation in symptom genesis, consistent with emerging theories of “inflammatory depression” but novel in the SAH context.

The identification of NLRP3 inflammasome as a key mediator in this process has important therapeutic implications[28-30]. Our temporal analysis revealing the sequence of cellular edema → neuroinflammation → NLRP3 activation → white matter damage → emotional disorders suggest multiple intervention windows. Early anti-edema treatments (1-6 hours), followed by anti-inflammatory strategies targeting NLRP3 (6-24 hours), may prevent the cascade leading to structural damage and emotional dysfunction. Recent studies demonstrating that NLRP3 inhibition can alleviate neuroinflammation and improve outcomes in various neurological disorders[31,32]. Our findings suggest that targeting NLRP3-mediated neuroinflammation may represent a promising strategy for preventing and treating post-SAH anxiety-depression disorders. Furthermore, the observed correlations between DTI parameters and emotional behaviors indicate that DTI could serve as an early imaging biomarker for predicting the risk of developing emotional disorders after SAH, potentially enabling early intervention.

Several limitations should be acknowledged. While animal models provide mechanistic insights, they may not fully recapitulate the complexity of human emotional disorders, particularly the cognitive and social dimensions of anxiety and depression. Our focus on NLRP3-related pathways, though well-justified, may overlook other inflammatory cascades contributing to post-SAH emotional disorders. Additionally, while our temporal analysis suggests sequential relationships, direct causal links require validation through targeted interventional studies using NLRP3 inhibitors or genetic approaches. Future research should address these limitations and systematically evaluate the therapeutic potential of targeting NLRP3 inflammasome in clinical SAH populations.

CONCLUSION

This study demonstrates that microstructural damage in emotion-regulating neural circuits after SAH is closely associated with anxiety-depression disorders. DTI parameters can early reflect neural circuit remodeling processes, providing objective assessment indicators for emotional disorders after SAH.