Long-term outcome and predictors in cerebral venous sinus thrombosis: a prospective cohort study

Abstract

Background

Although cerebral venous sinus thrombosis (CVT) is a rare disease with an overall favorable prognosis, 3%-15% CVT patients may experience death or severe disability during acute phase from symptom onset. We examined clinical characteristics in CVT patients and identify predictors associated with 12-month functional outcomes in patients with newly diagnosed CVT.

Methods

In this prospective cohort study, 61 patients with newly diagnosed CVT between Jan 2024 and Aug 2025 were enrolled and followed up for 12 months. The primary outcome was functional disability at 12 months, defined by a modified Rankin Scale (mRS) score ≥ 2. Clinical, radiological, and demographic variables were assessed as potential predictors. Logistic regression was used to identify independent predictors of poor 12-month outcome.

Results

At 12 months, fifty-seven patients (93.4%) had a favorable outcome (mRS < 2), while 4 patients (6.6%) had a poor outcome (mRS 2–4). Poor outcome was associated with higher admission NIH Stroke Scale (NIHSS) scores, impaired consciousness, brain parenchymal infarction, and extensive thrombosis. In multivariable analysis, admission NIHSS score was independently associated with poor outcome (odds ratio 1.21 per point; 95% CI 1.052–1.396). An NIHSS score ≥ 9 had a sensitivity of 75.0% and specificity of 93.0% for predicting disability at 12 months.

Conclusion

An NIHSS score ≥ 9 at admission independently predicts poor 12-month outcome in CVT and may guide early risk stratification and management decisions.

Trial registration

This is a secondary analysis of prospectively collected data from the registered clinical trial (URL: https://www.chictr.org.cn; No. ChiCTR2400080136; Registration date: Jan 22, 2024).

Introduction

Cerebral venous sinus thrombosis (CVT) is a rare form of cerebrovascular disease that accounting for approximately 0.5-3.0% of all stroke cases [1]. It is mostly associated with one or more hereditary or acquired thrombophilia, and manifests as nonspecific symptoms of elevated intracranial pressure, brain parenchymal damage, and epileptic seizures resulting from impaired venous drainage [2, 3]. The overall prognosis in CVT patients is commonly favorable, with an estimated functional independency in about 80%. However, 3%-15% CVT patients may experience death or severe disability from the onset, particularly within the first 30 days [3, 4]. It is crucial to distinguish high-risk patients from those at lower risk and to identify variables associated with poor outcome, thereby assisting clinician in decision-making intervention.

Predictors associated with neurological outcome were proposed from clinical CVT studies, including demographic, symptomatic, radiological, and thrombophilic variables, and some risk scoring models have been developed to predict prognosis in CVT patients [5–8]. However, CVT is often implicated in multifactorial etiologies and exhibits geographic and demographic differences across these studies [9, 10], the correlation between the variables and the prediction of outcome is not invariably constant and has not widely established yet. Moreover, recent epidemiological trends demonstrate a decline of CVT during pregnancy and the postpartum period, with a simultaneous rise in incidence among men and older adults [11, 12]. These shifting demographic patterns could influence the clinical presentation and long-term prognosis of CVT.

This single-center, prospective, observational cohort study aims to evaluate the clinical characteristics of CVT, identify predictors associated with long-term outcome in CVT patients, and develop a prognostic risk model to facilitate early risk stratification.

Methods

Study population

This is a secondary analysis utilizing data from the entire prospective cohort embedded within a registered clinical trial protocol (URL: https://www.chictr.org.cn; No. ChiCTR2400080136). No matching or subgroup selection was performed for the purpose of this analysis. Adult patients with newly diagnosed CVT in the Guangdong region, China, were enrolled in the Department of Neurology, First Affiliated Hospital, Sun Yat-sen University. Patient enrollment and follow-up were conducted from Jan 2024 to Aug 2025. Inclusion criteria required participants to be: (1) aged 18–80 years; (2) newly presented at least one clinical feature of CVT; and (3) CVT diagnosis confirmed by CT venography (CTV), MR venography (MRV), or digital subtraction angiography (DSA) at admission. Exclusion criteria were: (1) age < 18 years or > 80 years; (2) non-first-time diagnosis of CVT; (3) severe neurological comorbidities that could confound clinical assessment. (4) primary symptoms attributable to other disorders such as encephalitis, metabolic encephalopathy, vascular malformations, moyamoya disease, aneurysms, severe infection/trauma, malignancy or chronic debilitating diseases; (5) loss to follow-up or death prior to data collection; (6) incomplete medical records lacking essential history, imaging, or laboratory data; or (7) pregnancy or lactation. The study was approved by Ethics Review Committee of the First Affiliated Hospital of Sun Yat-sen University (No. [2024]032), and was conducted in accordance with the principles of the Declaration of Helsinki, adhering to the STROBE guidelines. All participants gave written informed consent for this study.

Clinical evaluation

We collected demographic information, etiology, medical history, and clinical features—including mode of onset, neurological symptoms, acute and maintenance treatments, comorbidities and follow-up outcome. Onset was classified as acute (< 48 h), subacute (2–30 days), or chronic (> 30 days). Documented symptoms included headache, nausea/vomiting, seizures, focal deficits, mental status changes, impaired consciousness, visual disturbances, and cranial nerve involvement. Lumbar puncture was performed when needed to rule out alternative diagnoses. Consciousness level and neurological severity were assessed at admission using the Glasgow Coma Scale (GCS) [13] and the NIH Stroke Scale (NIHSS) [14]. Impaired consciousness was defined as a GCS score < 15 or an NIHSS item 1a (level of consciousness) score ≥ 1 to capture potential consciousness alterations. Follow-up was conducted at discharge, 3, 6, and 12 months. Outcome were assessed at 3 months (short-term) and 12 months (long-term) using the modified Rankin Scale (mRS), where scores 0–1 indicated favorable outcome (independent) and 2–6 indicated poor outcome (dependent or deceased) [15]. Functional outcome was determined through direct or telephone interviews by investigators.

Imaging

All patients underwent initial brain CT/CTV or MR/MRV upon admission. Imaging evaluations were performed independently by two experienced radiologists blinded to clinical information, with any discrepancies resolved through joint consensus review to yield a final determination. Typical radiological features of CVT — such as venous sinus filling defects, dense sinuses, or collateral venous circulation — were identified to differentiate from congenital hypoplasia. In cases with suspected vascular abnormalities, DSA in the venous phase was conducted to confirm the diagnosis [16]. The number of thrombosed veins and secondary brain parenchymal lesions (e.g., ischemic or hemorrhagic infarction, intracranial hemorrhage) were documented. Follow-up CT or MRI was performed at 3 and 12 months to evaluate recanalization and thrombotic recurrence. Recurrence was defined as radiographic evidence of thrombus progression or new thrombosis accompanied by clinical symptoms. Recanalization was graded using the Qureshi scale: grade 1 (no or minimal recanalization), grade 2 (partial recanalization in multiple sinuses), or grade 3 (complete recanalization of all involved sinuses) [17].

Statistical analysis

Data analyses were performed using SPSS Statistics (version 26; IBM Corp.). Continuous variables following a normal distribution are presented as mean ± standard deviation (SD) and compared using Student’s t-test; non-normally distributed data are reported as median (interquartile range) and analyzed with the Mann-Whitney U test. Categorical variables are summarized as frequencies and percentages, and compared with Pearson’s chi-square test. Comparisons were made between patients with favorable long-term outcome (mRS < 2) and those with poor outcome (mRS ≥ 2). Variables with a p-value < 0.05 in univariate analysis were entered into a binary logistic regression model, with results reported as odds ratios (OR) and corresponding 95% confidence intervals (CI). The model’s discrimination was assessed using receiver operating characteristic (ROC) curve analysis, and calibration was evaluated with the Hosmer–Lemeshow goodness-of-fit test. A two-sided p-value < 0.05 was considered statistically significant.

Results

A total of 100 adult CVT patients were initially screened. Among them, 39 were excluded for the following reasons: comorbid metabolic encephalopathy or severe peripheral neuropathy (n = 2), loss to follow-up (n = 1), incomplete data (n = 1), and non-first-time diagnosis (n = 35). Ultimately, 61 patients with newly diagnosed CVT were included in the analysis (Fig. 1). The enrolled patients received standard anticoagulation therapy using low-molecular-weight heparin or rivaroxaban within 48 h after diagnosis, and two patients (3.3%) underwent decompressive craniectomy due to clinical deterioration. At 3 months, 20 patients (32.8%) had a mRS score ≥ 2, decreasing to 4 patients (6.6%) at 12 months. Totally, 57 patients (93.4%) achieved a favorable outcome (mRS < 2), whereas 4 patients (6.6%) had a poor outcome (mRS 2–4). Headache, visual issues, or cognitive problems were the main residual symptoms in these four patients with poor outcomes. No major bleeding complications or deaths were observed during the 12-month follow-up (Fig. 2).

Fig. 1.

Flow chart of patient enrollment. Sixty-one patients with newly diagnosed CVT were enrolled and completed 12-month follow-up

Fig. 2.

Distribution of modified rankin scale (mRS) scores at discharge, 3 months, 6 months, and 12 months after discharge. Fifty-seven patients (93.4%) achieved a favorable outcome (mRS < 2), whereas 4 patients (6.6%) had a poor outcome (mRS 2–4) during the 12-month follow-up period

Clinical features

The mean age of the patients was 43.2 years (SD 17.1), and 36 (59.0%) of them were male. Subacute onset (2–30 days) occurred in 42 (68.9%) patients, while 3 (4.9%) had acute and 16 (26.2%) had chronic presentations. Most patients (n = 51, 83.6%) had identifiable prothrombotic risk factors, including infection (31.2%), thrombocytosis (26.2%), and polycythemia (21.3%). Headache was the most common presenting symptom (83.6%), followed by nausea/vomiting (55.7%), focal neurological deficits (29.5%), and visual disturbances (26.2%). Seizures were observed in 10 patients (16.4%). Compared with patients with favorable long-term outcome, those with poor long-term outcome were more seriously ill at admission (median 14.0 vs. 0.0 by NIHSS, p = 0.007; median 10.0 vs. 15.0 by GCS, p = 0.029; median 4.0 vs. 1.0 by mRS, p = 0.003), and had higher prevalence of impaired consciousness (75.0% vs. 19.3%, p = 0.035) and focal neurologic deficit (100.0% vs. 24.6%, p = 0.006). No significant differences were observed in age, sex, other clinical presentation, or cerebrospinal fluid findings (Table 1).

Table 1.

Baseline clinical characteristics in CVT patients stratified by 12-month functional outcome

Characteristic	All CVT (n = 61)	Poor outcome (mRS ≥ 2; n = 4)	Favorable outcome (mRS < 2; n = 57)	p-value
Age, years (mean ± SD)	43.2 ± 17.1	46.0 ± 5.7	43.0 ± 17.6	0.732
Male sex, n (%)	36 (59.0)	2 (50.0)	34(60.0)	1.000
Onset pattern, n (%)
Acute (≤ 48 h)	3 (4.9)	0(0.0)	3(5.3)	1.000
Subacute (2–30 days)	42(68.9)	3(75.0)	39(68.4)	1.000
Chronic (> 30 days)	16(26.2)	1(25.0)	15(26.3)	1.000
Smoking, n (% of yes)	11(18.0)	1(25.0)	10(17.5)	0.559
Alcohol use, n (% of yes)	10(16.4)	1(25.0)	9(15.8)	0.521
Prior medical history, n (%)
Coronary artery disease	1(1.6)	0(0.0)	1(1.75)	1.000
Diabetes mellitus	4(6.6)	1(25.0)	3(5.3)	0.243
Hypertension	15(24.6)	0(0.0)	15(26.3)	0.564
Hyperlipidemia	23(37.7)	2(50.0)	21(36.8)	0.628
Prior VTE (DVT or PE)	5(8.2)	0(0.0)	5(8.8)	1.000
Initial symptoms, n (%)
Headache	51(83.6)	3(75.0)	48(84.2)	0.521
Nausea and vomiting	34(55.7)	3(75.0)	31 (54.4)	0.623
Visual disturbances	16(26.2)	1 (25.0)	15 (26.3)	1.000
Impaired consciousness	14 (23.0)	3(75.0)	11 (19.3)	0.035*
Seizures	10(16.4)	1 (25.0)	9 (15.8)	0.521
Focal neurologic deficit	18(29.5)	4(100.0)	14(24.6)	0.006 **
Cognitive/psychiatric symptoms	5(8.2)	0(0.0)	5(8.8)	1.000
Cerebrospinal fluid findings (if lumbar puncture done), n (% tested)
Elevated opening pressure	21(34.4)	2(50.0)	19(33.3)	0.523
Elevated WBC count	6(9.8)	1(25.0)	5(8.8)	0.346
Elevated protein	15(24.6)	1(25.0)	14(24.6)	1.000
Functional parameters
NIHSS scores at admission, median (IQR)	0.0(0.0,4.0)	14.0(7.8,18.5)	0.0(0.0,4.0)	0.007 **
GCS scores at admission, median (IQR)	15.0(13.0,15.0)	10.0(8.5,12.0)	15.0(14.0,15.0)	0.029 *
mRS scores at discharge, median (IQR)	1.0(0.0,2.0)	4.0(3.5,4.3)	1.0(0.0,2.0)	0.003 **

Lumbar puncture data were missing for 28 cases (45.9%). Data for all other variables were complete

*: p-value < 0.05; **: p-value < 0.01

Imaging findings

The transverse (80.3%), sigmoid (77.1%), jugular (59.0%), and superior sagittal sinuses (52.5%) were the most frequently involved sites. Most patients (77.0%) had thrombosis involving multiple sinuses, and 65.6% had parenchymal brain injury on imaging. Notably, inferior sagittal sinus thrombosis (75.0% vs. 8.8%, p = 0.006), involvement of ≥ 3 venous sinuses (100.0% vs. 40.4%, p = 0.034), infarction (100.0% vs. 43.9%, p = 0.046), and midline shift (75.0% vs. 10.5%, p = 0.009) differ significantly by long-term outcome. At the 3-month follow-up, 45 (73.8%) patients showed partial recanalization, and 13 (21.3%) achieved complete recanalization. By 12 months, 21 (34.4%) achieved complete recanalization. Recanalization rates were not associated with long-term outcome (Table 2; Fig. 3).

Table 2.

Imaging findings in CVT patients stratified by 12-month functional outcome

Imaging Characteristic	All CVT (n = 61)	Poor outcome (mRS ≥ 2; n = 4)	Favorable outcome (mRS < 2; n = 57)	p-value
Thrombosed veous sinuses, n (%)
Superior sagittal sinus	32 (52.5)	3(75.0)	29(50.9)	0.614
Inferior sagittal sinus	8(13.1)	3(75.0)	5 (8.8)	0.006 **
Transverse sinus	49 (80.3)	4 (100.0)	45 (79.0)	0.576
Sigmoid sinus	47(77.1)	4 (100.0)	43 (75.4)	0.565
Straight sinus	13(21.3)	2 (50.0)	11 (19.3)	0.196
Confluence of sinuses	15(24.6)	2 (50.0)	13 (22.8)	0.251
Jugular vein	36(59.0)	3(75.0)	33 (57.9)	0.638
Superficial cerebral veins	15(24.6)	1 (25.0)	14 (24.6)	1.000
Deep cerebral veins	10(16.4)	2 (50.0)	8 (14.0)	0.122
Number of sinuses involved, n (%)
1	14 (23.0)	0 (0.0)	14 (24.6)	0.565
2	20(32.8)	0 (0.0)	20 (35.1)	0.293
≥ 3	27(44.2)	4(100.0)	23 (40.4)	0.034*
Brain parenchymal lesion, n (%)
Any parenchymal lesion (total)	40(65.6)	4 (100.0)	36 (63.2)	0.289
Hemorrhage	29(47.5)	3(75.0)	26 (45.6)	0.338
Infarction	29(47.5)	4 (100.0)	25 (43.9)	0.046*
Midline shift	9(14.8)	3(75.0)	6 (10.5)	0.009**
Brain herniation	4(6.6)	1 (25.0)	3 (5.3)	0.243
Recanalization at 3 months, n (%)
Complete	13(21.3)	0 (0.0)	13 (22.8)	0.569
Partial	45(73.8)	4 (100.0)	41 (71.9)	0.565
None	3 (4.9)	0 (0.0)	3 (5.3)	1.000
Recanalization at 12 months, n (%)
Complete	21(34.4)	2 (50.0)	19 (33.3)	0.602
Partial	36(59.0)	2 (50.0)	34 (59.7)	1.000
None	4(6.6)	0 (0.0)	4 (7.0)	1.000

Follow-up imaging data were missing for 5 cases (8.2%)

*: p-value < 0.05; **: p-value < 0.01

Fig. 3.

Venous sinus recanalization status at 3 months and 12 months. Recanalization rates showed no correlation with long-term outcomes

Binary logistic regression

This study featured a limited sample size and a low number of endpoint events, and the analysis presented here is exploratory in nature. All baseline variables were included as potential predictors in the regression models. To rigorously prevent model overfitting, the number of variables included in the binary logistic regression analysis was restricted to two. Variables with a p-value < 0.05 in univariate analysis were considered, among which admission NIHSS score and impaired consciousness — both crucial clinical indicators to evaluate neurological severity and consciousness level — were selected for further analysis. Poor long-term prognosis of CVT was set as the dependent variable. Using a forward stepwise regression approach, admission NIHSS scores (OR 1.212, 95% CI 1.052–1.396, p = 0.008) was identified as an independent predictor of poor long-term outcome in CVT. Each one-point increase in the NIHSS scores was associated with a 21.2% increase in the risk of long-term poor prognosis, and this association was statistically significant. Figure 4 shows the area under the ROC curve (AUC) for admission NIHSS scores was 0.873 (95% CI: 0.714-1.000, p = 0.013). The Hosmer-Lemeshow goodness-of-fit test (χ²=7.362, p = 0.118) indicated good model calibration. An optimal cutoff value of 9 points in NIHSS yielded a sensitivity of 75.0% and specificity of 93.0% for predicting poor long-term outcome.

Fig. 4.

ROC Curve of the predictive model for poor long-term outcome. An optimal cutoff value of 9 points in NIHSS yielded a sensitivity of 75.0% and specificity of 93.0% for predicting poor long-term outcome

Discussion

In this prospective cohort of patients with newly diagnosed CVT, we found a clearly local demographic shift characterized by a higher proportion of younger male patients compared to historic cohorts [4], aligning with recent global epidemiological trends [11]. This shift in sex distribution within our population may be potentially driven by multiple factors such as a decline in pregnancy- and postpartum-related CVT due to improvements in maternal care, and the broader use of low-estrogen formulation. Despite this gender shift in CVT, a majority of patients (93.4%) achieved a favorable outcome with no deaths or major bleeding complications occurring during standard anticoagulation therapy. This favorable outcome may be attributed to milder admission neurological deficits (lower NIHSS scores) and earlier anticoagulation (within 48 h of diagnosis). Notably, functional outcomes improved markedly over time, with 32.8% of patients showing disability at 3 months, decreasing to 6.6% at 12 months. These findings underscore the potential for neurological recovery in CVT but also highlight the persistence of residual symptoms—such as headache, visual disturbances, and cognitive deficits—in a subset of patients. Early identification of individuals at high risk for poor long-term outcome remains essential to optimize therapeutic strategies and rehabilitation planning.

Our study did no observed correlation between age, gender, onset pattern, prior medical history, smoking, alcohol use, and initial symptoms with poor long-term outcome, although headache was the most common initial symptom and typically presented with subacute onset. It is noteworthy that the status of vascular recanalization, irrespective at 3 or 12 months, was not observed to influence long-term outcome. Previous studies reported a vascular recanalization rate of 74% and a favorable functional outcome rate of 68% in CVT patients within 8 days, with no significant difference between early (28 days–3 months) and chronic-phase (1–3 years) vascular recanalization rates [18–20]. In this study, the rates of short (4.9%) and long-term (6.6%) non-recanalization were comparable. However, some patients with partial early recanalization achieved complete recanalization over time, leading to further neurological improvement. It should be noted that the relatively low rate of complete recanalization (21.3%) at 3 months may have limited our ability to detect a potential association between recanalization and neurological recovery. Altogether, these findings suggest that post-anticoagulation vascular recanalization primarily occurs in the early disease phase and may be more associated with early improvement of symptoms.

In the univariate analysis, higher admission NIHSS scores, lower admission GCS scores, higher discharge mRS scores, impaired consciousness, focal neurological deficit, inferior sagittal sinus thrombosis, involvement of ≥ 3 venous sinuses, brain parenchymal infarction and midline shift were risk factors for poor long-term outcome in CVT patients, suggesting that the more severe the disease condition at onset, the worse the long-term outcome might be [21]. Further binary logistic regression analysis revealed that the admission NIHSS score was an independent predictor of poor long-term outcome in CVT patients. The NIHSS score is normally used to assess the severity in patients with acute arterial stroke [22, 23]. The early ISCVT-RS, a score used to evaluate CVT outcome, did not include it [6], but the IN-REvASC score by the ACTION-CVT study found that the admission NIHSS score was highly correlated with poor 90-day neurological outcome [8]. In addition, the NIHSS also serves as the primary basis for the “neurological deficit state " component within the SI2NCAL2C score, a comprehensive stratification tool for CVT risk assessment [24, 25]. The present study revealed that when the admission NIHSS score was ≥ 9, its sensitivity for predicting poor long-term outcome in CVT patients was 75.0%, and the specificity was 93.0%, which could aid in the early and fast identification of CVT patients who are at high risk for poor long-term outcome. Patients identified as high-risk at admission could warrant an earlier and more intensive strategy, such as intensively pharmacological and interventional revascularization, enhanced monitoring, as well as proactive complication prevention and rehabilitation. In contrast, Hiltunen S et al. used an admission NIHSS score of > 2 as the cutoff value for poor outcome in CVT patients [26]. This discrepancy may be due to two reasons: Hiltunen S’s study was based on an early retrospective cohort in the Finnish population, where the NIHSS score was only treated as a binary variable (NIHSS score ≤ 2 or > 2). Additionally, with increased awareness of CVT, mild cases with lower NIHSS scores are now more likely to receive timely and adequate treatment during the acute phase compared to decades ago, potentially improving their long-term outcome.

This study also has several limitations. First, as a single-center exploratory subgroup study, due to the rarity of CVT, the sample size—determined by the primary endpoint of the original study protocol—was relatively small and may introduce selection bias. This limitation may lead to insufficient statistical power and a substantial risk of overfitting in multivariable models. Second, to minimize the risk of overfitting in the predictive model, only two variables were included in the binary logistic regression analysis. It may inadvertently exclude other clinically relevant factors that could have prognostic importance. Third, a baseline NIHSS ≥ 9 serves as an early warning signal for poor functional outcome at one year, applicable for rapid and simple clinical triage, but a multi-factor scoring systems such as SI2NCAL2C remains comprehensive and thorough risk assessment for prognosis. Future large-scale, prospective, multicenter studies are needed to definitively validate our findings.

Conclusion

In this prospective cohort of patients in the Guangdong region, China with newly diagnosed CVT, an NIHSS score of ≥ 9 at admission was an independent predictor of poor functional outcome at 12 months. This threshold demonstrated high specificity (93%) and moderate sensitivity (75%) for identifying patients at high risk of long-term disability. Our findings support the use of admission NIHSS as a practical tool for early risk stratification and targeting intensified monitoring or intervention in high-risk patients.

Abbreviations

CVT

Cerebral venous sinus thrombosis

CTV

CT venography

MRV

MR venography

DSA

Digital subtraction angiography

GCS

Glasgow coma score

NIHSS

National Institutes of Health Stroke Scale

mRS

Modified rankin scale

Authors’ contributions

Songjie LIAO and Rui WANG conducted patient recruitment, follow-up, and wrote the initial manuscript. Zhaorong LI performed sample collection and data collation. Lin XU made data analysis and critical revision. Jian YU conceived and conducted the study. All authors reviewed and approved the final manuscript.

Funding

This study was supported by grants from Guangzhou Municipal Science and Technology Project (No.2024B03J0121), Guangdong Provincial Clinical Research Center for Neurological Diseases (2020B1111170002), Guangdong Province International Cooperation Base for Early Intervention and Functional Rehabilitation of Neurological Diseases (2020A0505020004), Guangzhou Major Difficult and Rare Diseases Project (2024MDRD02), Guangdong Provincial Engineering Center for Major Neurological Disease Treatment, Guangdong Provincial Translational Medicine Innovation Platform for Diagnosis and Treatment of Major Neurological Disease, Guangzhou Clinical Research and Translational Center for Major Neurological Diseases.

Data availability

The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The study was approved by Ethics Review Committee of the First Affiliated Hospital of Sun Yat-sen University (No. [2024]032), and was registered with the Chinese Clinical Trial Registry (URL: https://www.chictr.org.cn; No. ChiCTR2400080136). The study followed the STROBE guidelines and the Declaration of Helsinki. All participants gave written informed consent for this study.

Consent for publication

Informed consent for publication of the anonymized medical images was obtained from the patient (or their legal guardian). The images have been processed to protect the patient’s identity to the fullest extent possible, and no other identifiable personal information is included in the manuscript.

Competing interests

The authors declare no competing interests.